List of Tables.. v. List of Figures.. viii. Acknowledgements.. xii. Section 1 Executive Summary.. 1

Transcription

1 i TABLE OF CONTENTS List of Tables.. v List of Figures.. viii Acknowledgements.. xii Section 1 Executive Summary.. 1 Section 2 Development of an Aerobic Fitness Standard.. 6 Introduction.. 6 Oxygen Cost of Firefighting Work.. 11 Section 2.2 The Effects of the Self-Contained Breathing Apparatus (SCBA) and Fire Protective Clothing on Maximal Oxygen Uptake (VO 2max ).. 20 Introduction.. 20 Methods.. 21 Results.. 23 Discussion.. 32 Conclusions.. 34 Section 2.3 Aerobic Demands of Fire-Rescue Training Scenarios.. 35 Introduction.. 35 Methods.. 36 Results.. 44 Discussion.. 54 Summary.. 61

2 ii Section 2.4 The Oxygen Cost of the CF/DND Fire Fit Test and Graded Treadmill Exercise in Males and Females.. 62 Introduction.. 62 Methods.. 62 Results.. 67 Discussion.. 82 Summary.. 91 Section 2.5 Description of the Applicant Treadmill Test and Interpretation of Test Results.. 93 Introduction.. 93 Description of the Treadmill Test Protocol.. 93 Interpretation of Treadmill Test Results.. 98 Section 3 Development of Job-Related Tests and Standards Introduction Methods Selection of Representative Tasks Pilot Studies Job-Related Performance Tests Equipment Carry/Vehicle Extrication Test Job-Related Test Protocol Main Data Collection Introduction.. 121

3 iii Validation of Applicant Test Protocols Reliability of the Applicant Tests Determining the Minimum Standard for Applicant Tests Determining the Optimal Performance Level for Applicant Tests Results Discussion Validity of the Job-Related Performance Tests Reliability of the Job-Related Performance Tests Determining Minimum Standards Determining Optimal Standards Interpretation of Test Scores Section 4 Evaluation of the Applicant Test Protocol Introduction The Effect of Fire-Training on Test Performance The Effect of Gender on Test Performance The Effect of Age on Test Performance Section 5 Recommendations Introduction Recommendations.. 169

4 iv Section 6 References Appendix A Sample Information Package for Applicants Appendix B Sample Medical Clearance for Testing Appendix C Sample Informed Consent for Testing Appendix D Sample Greeter Check-list Appendix E Sample Treadmill Test Data Sheet Appendix F Sample Job-Related Test Data Sheet Appendix G Sample Report Card of Test Results Appendix H Points Distribution for Firefighter Applicant Test Scores Appendix I Sample Script for Test Administration Appendix J Typical Timeline for the Complete Applicant Test Appendix K Schematic for Car Door Mock-up.. 213

5 v List of Tables Table Physiological responses at maximal exercise 26 Table Table Table Table Table Table Table Table Table Mean (±SD) physical characteristics and aerobic power of male (n=13) and female (n=12) participants. 47 Summary of fire-rescue competencies covered in the training plan. 48 Mean (±SD) split and total times for fire-rescue scenarios for male (26 observations) and female (24 observations) subjects. 49 Oxygen consumption and ventilation responses during fire-rescue scenarios for male (26 observations) and female (24 observations) subjects. 50 Heart rate, perceived exertion and blood lactate responses during fire-rescue scenarios for male (26 observations) and female (24 observations) subjects. 51 Mean (±SD) physical characteristics of male (n=30) and female (n=23) participants 70 Mean (±SD) values for selected physiological measurements from the experimental trials of the CF/DND FF Test circuit with the portable metabolic system (VmaxST) and the SCBA for male (n=30) and female (n=23) subjects. 71 Selected responses during the 8-min constant work phase of the GXT for males (n=30) and females (n=23). Values are mean ± SD 78 Selected characteristics of experienced firefighters (n=97) 100 Table VO 2max scores and associated points 100 Table Selected demographic characteristics of incumbent firefighters (n=37) 131

6 vi Table Table Table Table Selected demographic characteristics of incumbent firefighters (n=57) 131 Correlation coefficients between selected components of the FF-test and Applicant test with 37 and 57 firefighters 132 Rating of Perceived Exertion (RPE) values for selected components of the FF-test and the Applicant test with increased incumbent numbers 134 Frequency analysis of the questionnaire responses 135 Table Table Table Table Table Table Table Table Selected demographic characteristics of volunteers in the reliability sub-study (n=13) 136 Mean (±SD) performance times for (n=13) participants completing the job-related tests on two separate days 136 Mean (±SD) points for (n=13) participants completing the job-related tests on two separate days 137 Selected demographic characteristics of incumbent firefighters (n=93) 139 Breakdown of Pass and Fail responses of fire service Subject Matter Experts (n=23) for selected performance times on job-related tests 140 Selected demographic characteristics of firefighters (n=150) 140 Job-related test scores and associated point distributions where applicable 141 Mean physical characteristics and test results for trained and untrained fire-training students 153

7 vii Table Table Table Table Table Table Failing test scores for trained and untrained fire-training students 155 Selected characteristics of female firefighter applicants (n=28) 157 Selected characteristics of male firefighter applicants (n=399) 157 Distribution of passing and failing results on each individual test for female applicants (n=28) 158 Distribution of passing and failing results on each individual test for male applicants (n=399) 159 Physical and performance characteristics of older (n=34) and younger (n=34) male firefighter applicants. 167

8 viii List of Figures Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Mean (±SD) oxygen consumption during GXT PT and GXT PPE 27 Mean (±SD) pulmonary ventilation during GXT PT and GXT PPE. 28 Scattergram of the fractional change in VO 2max and peak ventilation between GXT PT and GXT PPE. 29 Mean (±SD) tidal volume during GXT PT and GXT PPE. 30 Mean (±SD) breathing frequency during GXT PT and GXT PPE. 31 Sample plots of average oxygen consumption for each of the five scenario segments for one male and one female subject in the Nozzle position during Scenario Two. 52 Relationship between average VO 2 and time to complete a simulated fire suppression protocol. 53 Relationship between completion times for the CF/DND FF Test during the SCBA and VmaxST conditions for males and females 72 Average oxygen consumption for males and females during the CF/DND FF Test. 73 Scatterplot of average oxygen consumption for males and females completing the CF/DND FF Test between 7.5 and 8.5 minutes. 74 Mean ± SD average oxygen consumption for males and females completing the CF/DND FF Test between 7.5 and 8.5 minutes. 75 Oxygen consumption record for one sample male and one sample female subject completing the CF/DND FF Test in approximately 7 minutes. 76

9 ix Figure Oxygen consumption (VO 2 ) and heart rate (HR) expressed as a fraction of maximal values for a sample male and a sample female subject completing the CF/DND FF Test in approximately 7 minutes. 77 Figure Regression analysis of maximal oxygen consumption, from the treadmill test for males and females and performance time from the Hard VmaxST trial of the CF/DND FF Test. 79 Figure Figure Figure Figure Figure Figure Regression analysis of maximal oxygen consumption, for males and females and peak 1-min oxygen consumption during the CF/DND FF Test. 80 Regression analysis of average VO 2 and work time from research where gas exchange has been measured during simulated fire-rescue work. 81 Treadmill test for evaluation of aerobic fitness of firefighter applicants with gas exchange analysis equipment for measurement of oxygen consumption. 97 Treadmill test for evaluation of aerobic fitness of firefighter applicants without gas exchange analysis equipment for measurement of oxygen consumption. 98 Distribution of VO 2max scores for 97 experienced firefighters 101 Front (A) and side (B) views of an applicant dressed in the correct attire for the job-related tests. 112 Figure View of the Charged Hose Advance Test. 113 Figure View of the Rope Pull Test. 114 Figure View of the Forcible Entry Test. 115 Figure View of the Victim Rescue Test. 116

10 x Figure View of the Ladder Climb Test. 117 Figure Figure View of the large and small spreader tools used in the Equipment Carry/Vehicle Extrication Test. 118 Panel A shows the correct technique for carrying the spreader tools. Panel B shows correct technique for lifting and lowering the spreader tools during the Equipment Carry/Vehicle Extrication Test. 119 Figure View of the Vehicle Extrication Test. 120 Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure The relationship between FF Test Rope Pull time and the Applicant Test Rope Pull time. 133 Scatterplot showing the scores for the Charged Hose Advance test on Day one and Day two of the reliability sub-study. 138 Scatterplot showing the scores for the total points earned on the five performance tests on Day one and Day two of the reliability substudy. 138 Distribution of performance times for the Hose Drag test (n=93) 139 Frequency distribution of treadmill test scores for female applicants (n=28) 161 Frequency distribution of Hose Drag scores for female applicants (n=25) 161 Frequency distribution of Rope Pull scores for female applicants (n=24) 162 Frequency distribution of Forcible Entry scores for female applicants (n=24) 162 Frequency distribution of Victim Rescue scores for female applicants (n=24) 163 Frequency distribution of Ladder Climb scores for female applicants (n=23) 163

11 xi Figure Frequency distribution of Equipment Carry/Vehicle Extrication test scores for female applicants (n=22) 164

12 xii ACKNOWLEDGEMENTS The investigators would like to thank the many people who contributed to this research. The research was funded by the Canadian Forces Fire Marshall and managed through the Canadian Forces Personnel Support Agency under the direction of Dr. Wayne Lee and with the assistance of Sue Jaenan, Daryl Allard, Ben Oullette, Patrick Gagnon and Kelly Lupton. Additional support for various aspects of the research was provided from Fire ETC (Emergency Training Center, Vermilion, AB), Emergency Services Academy (ESA, Edmonton, AB), and Edmonton Fire-Rescue. Without the support and cooperation from these agencies, this research would not have been possible. Many individuals worked with our research team at the University of Alberta and we wish to acknowledge the significant contributions of Allison Branston, Neil Eves, Mike Gilpin, Tim Hartley, Natasha Hutchinson, Merrin Lloyd, Reg Nugent, Mike Stickland, Damien Wild and Tina Wong. We would also like to thank our colleagues, Dr. David Docherty (University of Victoria), Dr. Richard Jones (University of Alberta), Dr. Olive Triska (University of Alberta), and Dr. Bruno Zumbo (University of British Columbia) for their interest, advice and support. Finally, we extend our sincerest thanks to the many, many individuals who participated as subjects in the various aspects of the project. Stewart Petersen and Randy Dreger

13 1 SECTION 1 EXECUTIVE SUMMARY The University of Alberta (U of A) entered into a contractual agreement with the Canadian Forces Personnel Support Agency (CFPSA) to conduct research into the development of bona fide physical fitness selection standards for Canadian Forces (CF) and Department of National Defence (DND) firefighters. The research has been conducted in the following steps to: 1. Identify the problem 2. Review the related literature 3. Identify appropriate methodology 4. Brief the Canadian Forces Fire Marshall (CFFM) on the findings 5. Submit a research proposal summarizing the work necessary to develop appropriate tests and standards 6. Conduct the research to develop appropriate tests and standards 7. Brief CFFM on the results of the research to develop the selection tests and standards 8. Submit a final report on the research conducted on the development of the selection tests and standards Steps 1 5 were completed and summarized in a previous report submitted to Dr. Wayne Lee dated January Step 6 has been completed according to the original contract and subsequent amendments (September 2002; May 2003; September 2004). From time to time as appropriate, the CFFM, Dr. Wayne Lee and/or Ms. Sue Jaenen and others have been briefed on the progress of the research (October 2001 in Montreal, November 2001 in Ottawa, February 2002 at CFB Comox, September 2002 in Ottawa, January 2003 in Edmonton, September 2003 in Toronto, March 2004 in Edmonton, August 2004 in Victoria, and November 2004 in Ottawa). The final briefing for CFFM took place in Ottawa in February 2006.

14 2 This document represents the final report on the research leading to the selection tests and standards that were discussed at the February 2006 briefing. For the purposes of this report, the test that has been developed for CF and DND firefighter applicants will be referred to as the Canadian Forces/Department of National Defence Fire Fighter Applicant Test (CF/DND FF Applicant Test). The development of bona fide selection tests and standards for firefighter applicants was based on several principles that were identified in the preliminary research leading to the main research proposal. These principles are summarized briefly below. First, the standard for applicants to the CF/DND fire service must be demonstrably similar to the standard for incumbents. That is, the physical fitness levels required to meet the applicant standard should be very similar to the physical fitness level associated with successful completion of the CF/DND Fire Fit (FF) Test (Deakin et al., 1996) at the 8-minute standard. Wherever possible, relationships between the CF/DND FF Test and the CF/DND FF Applicant Test have been demonstrated. However, in addition to demonstrating relationships with the CF/DND FF Test, it was equally important to demonstrate similar relationships to the physical demands of firefighting work. While the applicant test and standard is clearly linked to the incumbent test and standard, the former must also stand on it s own merit as a legitimate test for firefighters. Second, the investigators have a strong bias towards what is frequently referred to as a task simulation approach rather than the fitness component approach to occupational fitness testing. The merits of each approach were discussed in detail in the research proposal. Briefly, the former uses performance on simulations of critical or common tasks as the basis for testing. The latter approach relies on relationships between job performance and scores from common tests of fitness components. Since one of the main objectives was to

15 3 maintain close relationships between the applicant and incumbent standards, this approach was highly appropriate. Third, an important extension of the desire to match the applicant and incumbent standards (first principle) and also of accurate task simulations (second principle) required that the clothing ensemble worn during physical testing be as close as possible to the protective clothing ensemble worn by firefighters in the field. The importance of this principle is discussed elsewhere in this report. Briefly, the protective ensemble influences physical capacity due to the additional weight and the heat stress that results from exercise, even in a thermo-neutral environment. Fourth, the CF/DND FF Applicant Test was designed to allow the option of being administered by Canadian Forces Personnel Support Agency (CFPSA) staff at selected CF bases. While a series of implementation recommendations are provided later in this report, development of a specific implementation plan was not part of the research program. However, a guiding principle in the design of the CF/DND FF Applicant Test was to allow for in-house testing at CF bases as well as the possibility for contracting the testing to outside agencies such as University laboratories. Fifth, the CF/DND FF Applicant Test provides standards for both screening and selection. These terms have been used systematically throughout this report. For the purposes of this report, screening refers to the process of distinguishing between applicants who pass or fail the test. This is a very important step and the criteria for passing are consistent with what is known about bona fide occupational requirements (BFOR) for firefighting. If an applicant passes the test, the organization can be reasonably certain that the individual has the physical fitness required to move to the next step (e.g., fire training or probationary employment as a firefighter). The concept of selection takes this process one step further, allowing the ranking of successful applicants based on performance. Therefore, if the organization is interested in using physical fitness

16 4 as one of the criteria for selecting firefighters, then the ability to do so is provided. However, this matter and method of how this test might be used for selection should be addressed by CFFM and CFPSA while considering how other information (e.g., aptitude test results, previous experience or training) is used to select the most favorable applicants. The Applicant Test has been implemented on a trial basis at the University of Alberta. Our laboratory has long-standing relationships with numerous municipal fire departments in Alberta, as well as two fire-training schools. The pilot implementation allowed the collection of test data on various groups of individuals applying for full-time employment in the fire service, or alternately, either entering or graduating from full-time fire-training programs. Consequently, the effectiveness of the protocol for both screening and selection of applicants has been evaluated. The logistics of conducting the protocol have been established. Finally, as shown in Section 4 of this report, preliminary analysis of the data from real applicants shows no compelling evidence that the test protocol is biased on the grounds of skill, age or gender. Further evaluation of these questions is essential, however this work can only be done as the data on real applicants becomes available. The balance of this report is organized in sections that provide detailed explanations of various phases of the overall research project. Section 2 presents the work related to development of an aerobic fitness test and standard. Section 3 presents the work related to the development of job-related fitness tests and standards. Section 4 presents the results of pilot implementation of the tests and standards. Section 5 presents a series of recommendations for implementation of the CF/DND FF Applicant Test. Cited and related references are listed in Section 6. At the time of submission of the final report, a version of Section 2.2, the effects of the self-contained breathing apparatus (SCBA) and fire protective clothing on

17 5 maximal oxygen uptake (VO 2max ), has been accepted for publication in Ergonomics. Other parts of the research have been presented at scientific conferences and will be submitted for publication in scientific journals.

18 6 SECTION 2 DEVELOPMENT OF THE AEROBIC FITNESS TEST AND STANDARD Introduction This section summarizes the research related to the overall problem of developing a suitable test of aerobic fitness for firefighters. Other components of the main research project have considered the development of task-related simulations to evaluate muscle strength, power and endurance in firefighting. These are very important to safe and effective performance of firefighting work and cannot be overlooked if one wishes to accurately consider the stresses of firefighting work. Up to a point, in a task-simulation test (or in a real work situation), there is the possibility that an individual may compensate in order to get the work done. For example, a very strong or powerful individual may be able to rely on his/her muscular attributes to compensate for a lower aerobic fitness level, or vice-versa. This is one of the great advantages of task-simulation tests. On the other hand, we need to recognize that within this conceptual model, minimal levels of each relevant physical fitness component are required. Firefighting work often involves moving heavy objects (strength requirement) for long enough periods of time that the cardiovascular system is taxed at a high level (aerobic requirement), and frequently, because of the need for a timely response to an emergency, power and speed are essential (anaerobic requirement). Up to a point, individual firefighters can find a way to get the work done safely and effectively by relying on their physiological strengths to offset their weaknesses. Notwithstanding the above, it bears repeating that there is a fundamental or minimal level of each attribute that must exist before these weaknesses can effectively be compensated for.

19 7 The purpose of the first sub-study was to document the effects of fire protective clothing and the self-contained breathing apparatus on the maximal rate of oxygen consumption (VO 2max ). The second and third sub-studies were undertaken to document: the rate of oxygen consumption (this may also be expressed as the aerobic demand or the VO 2 associated with a particular exercise) of men and women performing selected firefighting tasks ( work samples ) in a fire-rescue training environment; the CF/DND Fire Fit Test (FF Test or the circuit ); and, treadmill exercise while dressed in firefighting ensemble (including personal protective clothing and self-contained breathing apparatus). There are several studies in the literature documenting the aerobic demands of firefighting on men, however there is almost no information available on women. It is important to compare male and female responses to simulated firefighting work in order to improve the understanding of male and female performance on both the new physical tests for applicants and the CF/DND task based circuit used by incumbents. Some preliminary comments are in order. First, there is no argument that the best method of studying the aerobic demands of firefighting would be to document the responses of incumbent firefighters in real emergency situations. In other words, the scientist would make observations during representative samples of work. For many reasons this is not safe, practical or possible. Consequently, scientists have been forced to rely on simulations of firefighting. The various types of simulations have inherent limitations, advantages and disadvantages, but at the end of the day, it must be recognized that it is impossible to study the aerobic demands of real firefighting, and that therefore, simulations are the only alternative. In this report, the term firefighting work is frequently used to infer that the work of the occupation is under investigation. An example of the work could be a specific task such as dragging a charged hose an appropriate distance over a surface with specific friction characteristics. It is possible to document the

20 8 physical work required to complete this task or alternately, to document the physiological responses in a human subject performing the task at a particular work-rate. Some authorities have used this approach to characterize the physical demands of firefighting (Gledhill and Jamnik, 1992a) and to develop physical testing programs for firefighters (Gledhill and Jamnik, 1992b). Alternately, a selection of individual tasks may be combined to create a specific sequence of events and therefore, a particular quantity of work. The Firefighter Combat Challenge (Davis et al., 1982) and the CF/DND FF Test (Deakin et al, 1996) are good examples of this approach. For sport competition or physical abilities testing, the time required to complete the specific sequence of jobrelated tasks has meaning. It becomes more difficult to accurately describe the nature of any individual task that is embedded in the sequence, however the physiological responses to the challenge of completing the series of events may provide a somewhat better insight into the physical demands of combinations of tasks. In the real world, fire suppression and rescue work rarely involves one isolated task. Instead, firefighters are expected to recognize and respond to a variety of problems in a rapidly changing environment. Therefore, the physical testing approach that involves moving rapidly from one task to the next may provide a better approximation of the physical demands of fire-rescue operations. The third possibility that will be discussed in this report is a more loosely configured sequence of events that may be described as a job-related scenario (Sothmann et al., 1990; 2004). This could also be described as a type of work sample. It is important to keep the terminology precise however, in that the intent is to create a sample of the work, not the job. If the work sample or scenario is designed carefully, elements of problem-solving can be introduced in a controlled manner. For example, the task of victim rescue is an important part of the series of individual tests described by Gledhill and Jamnik (1992b), the Combat Challenge (Davis et al., 1982) and the CF/DND FF Test (Deakin et al, 1996). However, in each case the individual being evaluated is aware of when

21 9 and where this particular task is placed in the sequence of events. The individual doing the work can see the victim and is informed in advance as to the distance it must be moved. In contrast, during the fire-rescue training scenarios described in Section 2.3 of this report, the research subjects were trained to enter a series of dark, smokefilled rooms in a building and search for victims. The number and location of the victims was unknown. The research subjects had to first find the victim(s) and then use appropriate rescue strategies to get the victim(s) to a point of safety. In this situation, the physiological responses to the work can be studied under somewhat more realistic conditions. This model is very common in fire-training and we have adopted it in our research. Students are first introduced to individual skills (e.g., the Gledhill model ), then challenges that combine the skills in highly controlled situations (e.g., the Deakin model ), and ultimately, more complex problems that require the integration of skills in more life-like situations with obstacles, smoke, heat, and/or darkness (e.g., the Sothmann model ). Finally, if the intent were to study the job, it would be essential to utilize only trained firefighters as research subjects. However, if the intent is to study the physiological responses to samples of the work, then other subjects with appropriate physical characteristics may be used. In some of our research, physically active volunteers that were physically capable of doing the work were studied. Typically, this type of volunteer is very interested in firefighting and may or may not have any fire-training or job experience. In each case, a specific set of skills and experiences must be introduced and practiced until the volunteer demonstrates a level of competency allowing him/her to perform the work in the prescribed manner. For example, when studying the physiological responses in females doing firefighting work, an immediate logistical problem was the relatively small number

22 10 of female firefighters. Even in a large metropolitan area like Edmonton there are only a few females employed as firefighters. It is also vital to note that in Canada, research with human subjects is strictly regulated and all participants must freely volunteer. It would have been impossible to conduct these studies if we had restricted our subject pool to incumbent female firefighters who were willing to volunteer to participate. This problem is evident in several examples from the recent literature. For example, in our first experiment on the effects of hyperoxia on performance of simulated firefighting work, two out of 17 subjects were female firefighters (Petersen et al., 2000). This type of representation precludes any possibility of gender comparisons. Bilzon et al. (2001) studied the oxygen cost of 15 female and 34 male Royal Navy personnel during simulations of selected shipboard firefighting operations. Unfortunately, in the most demanding task, only 4 of the 15 females were able to complete the work, which again compromises the usefulness of the findings. Thus, in order to make gender comparisons, it is essential to have reasonable representation of female subjects who are physically capable of doing the work at an acceptable workrate. In our research, a logical alternative strategy was to utilize female volunteers who were willing to invest the time to be trained to complete the work in an acceptable fashion. It bears repeating that the research objective in both of the projects presented in this report was to study selected physiological demands of the work, not the workers on the job. Therefore, while we recognize that most of the research subjects were not firefighters, we argue strongly that the physiological responses to the samples of firefighting work are valid since the volunteers were thoroughly trained to complete the specific samples of work in a safe and efficient manner.

23 Oxygen Cost of Firefighting Work The oxygen cost (VO 2 ) of firefighting work is an important aspect in the characterization of the job demands. Several brief points of clarification are in order. When reading this literature, it is important to distinguish between the average oxygen cost of work and the peak oxygen cost of the work. The average VO 2 is an appropriate expression of the average aerobic energy cost of work of at least several minutes duration where the work-rate is variable. The CF/DND FF Test (or, the circuit ) provides a good example. The type of work (e.g., upper body dominated, lower body dominated) changes throughout the circuit, as does the intensity (e.g., specific event or transition between events). Consequently, the record of oxygen consumption could literally appear as a series of peaks and valleys. One can usually pick out the highest VO 2 quite easily, however the average VO 2 during the period of work is also of significant interest. This is potentially confusing since physiological data are frequently reported as the mean value plus or minus the standard deviation from the mean of a group of subjects. In the case where the physiological variable of interest is the average VO 2 of a group of subjects doing a task, the reader must be careful to distinguish between the average VO 2 and the mean of the average VO 2 (or mean average ) with respect to the group of subjects. In this report, the terms average and mean have been used systematically. The energy demands of firefighting work are of interest for a number of reasons. First and foremost, it is necessary to understand the rate of energy production that must be sustained in order to safely and effectively complete the work. Second, many researchers have used this data in order to identify a standard for the maximal oxygen consumption (VO 2max ) that firefighters must attain in order to safely and effectively meet the demands of the work. The VO 2max is often considered the gold standard measurement of aerobic fitness (Sutton, 1992) since it represents the peak or maximal rate of oxygen consumption. VO 2max is typically measured during a graded exercise test on a

24 12 cycle ergometer or treadmill. A graded exercise test (GXT) typically consists of minutes of progressively more difficult exercise that ends when the subject is too exhausted to continue. Expired gases are collected and analyzed to calculate the rate of oxygen consumption. The VO 2max is typically recorded at or near the point of exhaustion. The rate of oxygen consumption is usually expressed in liters of oxygen per minute (L. min -1 ). Another approach is to scale the oxygen consumption relative to body size. The most common method used for weight-bearing activities (where the individual is moving his/her own body mass) is to express the VO 2 relative to body mass (ml. kg -1. min -1 ). Since firefighting is mainly a weight-bearing activity, this is the most common method used to describe the oxygen cost of work or physical fitness in this occupation. A healthy, physically active individual could be reasonably expected to exercise for a relatively long time (e.g., 60 min) at an intensity of 60% of his/her VO 2max. However, as the intensity of work increases, the time that can be sustained decreases (Kilbom, 1980). For example, that same individual might be able to work for 30 min at 75% of VO 2max, 10 min at 85% of VO 2max, and 3 min at 95% of VO 2max. The values for time and intensity for any given individual are usually dependent on the level of training and may vary significantly between individuals. The values given above are estimates provided for the purposes of illustrating the relationship between work intensity and the time that specific work-rates may be sustained. While these are realistic values for many individuals, it should be emphasized that they do not apply universally. This point also underscores the potential problem of relying on identification of a maximal value and then inferring the capability to work for a given period of time at a specific submaximal level. Much of firefighting appears to involve work at submaximal levels. However, since the VO 2max is universally recognized as an important measurement of physical fitness and is relatively easy to measure, researchers have used this physiological variable to set fitness standards for workers.

25 13 Several methods have been employed in an attempt to determine the oxygen cost of firefighting work. Direct measurement of VO 2 during actual emergencies would, in theory, be the most accurate method, however this is impossible to accomplish. The utility of this approach is limited by: the expense and technical requirements of the necessary equipment; the potentially disruptive effects on execution of fire suppression duties when public safety is at risk; and, the environmental limitations of the instrumentation (e.g., inability to work in high temperatures and smoke). Therefore, direct measurement of oxygen uptake has normally been performed under simulated firefighting conditions. Published literature on the oxygen cost of simulated firefighting work has included: stairclimbing in the laboratory (O Connell et al., 1986); simulated smoke-diving (Lusa et al., 1994); simulated fire scenarios (Sothmann et al., 1990; 1992b); individual task performance (Lemon and Hermiston, 1977b; Gledhill and Jamnik, 1992a) and simulated shipboard firefighting (Bilzon et al., 2001). One of the first studies to examine the oxygen cost of firefighting utilized the Douglas bag method to directly measure the VO 2 of incumbents performing selected firefighting tasks (ladder climb, victim rescue, hose drag and ladder raise) in a safe and effective manner (Lemon and Hermiston, 1977b). Each of the firefighters wore standard turnout gear while performing the tasks. The results of the study indicated that the selected firefighting tasks required a VO 2 of 2.19 to 2.55 L. min -1. Using demographic data reported in this study, the VO 2 relative to body mass can be calculated as ml. kg -1. min -1. These data should however, be treated with caution (the reader should note that the data have not been used in any of the calculations of average VO 2 that appear below). The work times were very brief and the method of gas collection was unlikely to provide accurate data under those conditions. O Connell et al., (1986) measured the oxygen cost of simulated stair climbing in 17 firefighters. The firefighters were dressed in turnout gear and carried a length

26 14 of hose over their shoulder. The mean VO 2 observed during 5 minutes of exercise at a stepping rate of 60 steps. min -1 was 38.6 ml. kg -1. min -1. Gledhill and Jamnik (1992a) also used the Douglas bag method to measure the VO 2 of incumbent firefighters while performing a variety of firefighting tasks. Subjects performed the tasks in standard turnout gear including SCBA (without breathing from it) under 'emergency-like' conditions. These investigators reported that the average oxygen cost of 90% of the tasks analyzed required a VO 2 of 23.4 ml. kg -1. min -1 (range: 16.8 to 44.0 ml. kg -1. min -1 ). In the final analysis, the authors recommended that the most significant oxygen consumption values were obtained from the tasks that were most physically demanding and most frequently encountered. The mean VO 2 for these tasks was 40.7 ml. kg -1. min -1, and the corresponding work time was 2.35 minutes (sufficient duration to allow accurate measurements with the Douglas Bag method). One of the shortcomings of this research was that the tasks were performed in isolation. While this approach provided an accurate determination of the VO 2 required for each individual task, it did not allow for an accurate estimation of the VO 2 required for fire suppression work that involves a series of tasks or alternately, prolonged work on one task. One of the best investigations of the oxygen cost of firefighting studied incumbent firefighters during a firefighting scenario in a hot, smoke-filled environment that replicated the demands of a live-fire emergency (Sothmann et al., 1990). Oxygen consumption was directly measured utilizing a metabolic measurement system integrated with the SCBA. Subjects were requested to perform the scenario in a manner similar to an actual emergency. It was reported that the scenario required absolute and relative oxygen consumption rates of 2.5 L. min -1 (range: L. min -1 ) and 30.5 ml. kg -1. min -1 (range: ml. kg -1. min -1 ), respectively. The corresponding average completion time was 8.93 minutes.

27 15 Lusa et al. (1993) estimated VO 2 during a live-fire scenario by measuring the volume of air consumed from the SCBA cylinder. This method relies on some assumptions about respiratory gas exchange and consequently it is important to treat the results as estimates rather than measurements. The oxygen cost for the scenarios was estimated to be 2.4 L. min -1 (range: L. min -1 ) or 31 ml. kg -1. min -1 (range: 22 to 55.0 ml. kg -1. min -1 ). In contrast to the relative abundance of research describing the aerobic requirements of firefighting work in males, there has been only one study published in the scientific literature that examined the VO 2 responses of females during simulated firefighting work (Bilzon et al., 2001). This investigation studied 34 male and 15 female Royal Navy personnel who had a minimal level of shipboard firefighter training (3 days). Each subject performed five firefighting activities while dressed in appropriate ensembles of protective equipment. Each of the five activities was meant to be completed at a prescribed work-rate for 4- minutes. The activities were: Boundary cooling (BC) this activity involved operating a charged hose for a total work time of 4-min; Hose running (HR) this activity involved reeling out 12.3 m lengths of hose (7.1 kg) for a total time of 4-min; Ladder climbing (LC) subjects entered a hatch and descended a vertical distance of 2.8 m with a charged hose (approximately 10 kg). The subjects then released the hose, walked 14 m and ascended a ladder (vertical height 2 m) to the start position. Each subject completed 6 circuits within a 4-min period; Extinguisher carry (EC) subjects carried a liquid foam extinguisher (11.2 kg) along a predetermined route that included ascending two external sloping ladders (total vertical height of 4 m), walking across a horizontal deck (17 m), descending the ladders, and walking back to the start point. Each subject undertook 4 circuits of this route at a controlled work-rate of one circuit per minute; and,

28 16 Drum carry (DC) subjects carried liquid foam drums (30 kg) from the start point to the firefighting team in a lower-engine space at a rate of 1 drum per minute. In this scenario, each delivery involved carrying the drum down two sloping ladders (total vertical height of 4 m) and walking 25 m. The subject then put the drum down and returned to the start point. Respiratory gas exchange data were recorded using a VmaxST (SensorMedics, Yorba Linda, CA) portable metabolic measurement system. The oxygen cost (average for all subjects) for the various firefighting activities ranged from 23 ± 6 ml. kg -1. min -1 for the boundary cooling to 43 ± 6 ml. kg -1. min -1 for the drum carry. The mean oxygen cost to perform all of the fire suppression activities described in this experiment was approximately 34 ml. kg -1. min -1. There were no significant gender differences in the oxygen cost of doing the work in the BC, DC, or HR activities. However, the aerobic cost of the work was significantly higher for the males during the EC and LC tasks. When the oxygen cost of the work was expressed as a percentage of the VO 2max, this pattern was reversed. The mean VO 2max of the males and females was 53 ± 5 and 43 ± 6 ml. kg -1. min -1, respectively. Consequently, when expressed as a percentage of the VO 2max, the oxygen cost of the EC task was 76 and 86% for males and females, respectively. In other words, the females were required to work at a greater fraction of their maximal capacity in order to do the work at the prescribed rate. It is of interest that 11 of the 15 females failed to maintain the endorsed workrate in the DC task. The females who did complete the task at the acceptable work-rate demonstrated a VO 2 consistent with the average VO 2max for the female subjects. That is, the aerobic demand of the DC task for the four successful females was 42 ± 6 ml. kg -1. min -1, while the average VO 2max for the female subjects was 43 ± 8 ml. kg -1. min -1.

29 17 The results of this study reported the mean oxygen cost for females and males performing simulated firefighting work to be ml. kg -1. min -1, with a range of ml. kg -1. min -1. This is similar to most of the values reported in the literature for males, but it is important to recognize that the number of females studied was very small on the most demanding activity (DC), so the results should be interpreted with some degree of caution. With this one exception, the study by Bilzon et al. (2001) was well-designed, and the data are both interesting and valuable. However, one might question whether the results would apply to structural firefighting where the activities are somewhat different. In section 2.3 of this report, we describe in detail the experiment to document the aerobic demand of fire-rescue training scenarios. This work has been presented at a national exercise physiology conference and published in abstract form (Petersen and Dreger, 2004). In brief, the average VO 2 observed during approximately 12 minutes of simulated fire-rescue work was 26.6 ml. kg -1. min -1. By averaging the VO 2 data reported in the literature (O Connell et al., 1986; Sothmann et al., 1990, 1992a; Gledhill and Jamnik 1992a; Lusa et al., 1993; Bilzon et al., 2001), the mean oxygen cost of fire suppression and rescue work appears to be approximately 32 ml. kg -1. min -1 (range: 16 to 55 ml. kg -1. min -1 or 1.7 to 4.3 L. min -1 ). Sothmann et al. (1992b) reviewed the available literature and concluded that firefighters typically consume between ml. kg -1. min -1 while suppressing a structural fire. A number of the reported values are predictions of oxygen consumption (Sothmann et al., 1992a; Lusa et al., 1993), and the methods of prediction may have inherent errors. A more conservative and prudent approach might be to consider only the values where the VO 2 was actually measured (O Connell et al., 1986; Sothmann et al., 1990; Gledhill and Jamnik 1992a; Bilzon et al., 2001; Petersen and Dreger 2004). The average VO 2 from these research studies is 34.5 ml. kg -1. min -1.

30 18 At first glance, it may be tempting to speculate from the broad range of oxygen consumption values that scientists cannot make up their minds about the VO 2 associated with firefighting work. However, it is very important to bear in mind several important points when reviewing these values. First, there is a fundamental inverse relationship between intensity and duration of exercise. Secondly, experienced workers will tend to pace themselves at a level consistent with completing the work at hand (Manning and Griggs, 1983). Thirdly, sensory restriction (e.g., smoke and darkness) necessitates that the average work-rate be reduced. If a worker cannot see the object of his search, then he or she must slow down in order to search by feel. This process includes aspects of actual searching as well as maintaining spatial orientation in the search area. Variability in research design has led to the wide variability in observed VO 2. For example, the work-rate could be dictated by the investigators or self-selected by the subjects. Alternately, the subjects may have been given a discrete task and directed to complete it as fast as possible or, a series of tasks complicated by sensory deprivation. Finally, some measurement techniques are more precise than others. For example, Bilzon et al., (2001) directly measured respiratory gas exchange while Lusa et al., (1993) estimated oxygen consumption from measurements of air consumption. Careful review of the unique aspects of each research design is required prior to interpreting the values. Despite the apparently large number of studies that have tried to document the aerobic demands of firefighting work, there is actually very little consistency between research design factors. Therefore, each data set must be judged on its own merit. However, when combined, these studies can provide general insight into the oxygen cost of firefighting work. As noted previously, the energy demands of firefighting work are of interest for a number of reasons. First and foremost, it is necessary to understand the rate of energy production that must be sustained in order to safely and effectively complete the work. Second, many researchers have used this data in order to identify a minimum VO 2max standard necessary for firefighters to safely and

31 19 effectively meet the demands of the work. Recommendations for minimum VO 2max described by various authors range from 33.5 to 45 ml. kg -1. min -1 (Bilzon et al., 2001; Davis and Dotson, 1987b; Gledhill and Jamnik, 1992a; Lemon and Hermiston, 1977, Lusa et al., 1993; O Connell et al., 1986; Sothmann et al., 1990; and Sothmann et al., 1991). With the exception of Bilzon et al. (2001), the previous research has utilized male subjects. Consequently, descriptions of the aerobic demands of firefighting and suggested fitness standards from previous research are based on male performance. The aerobic demand on females performing firefighting tasks has not been adequately documented in the scientific literature. Given the general physical and physiological differences between males and females, it seems prudent to determine whether these differences influence the energetics of firefighting work.

32 THE EFFECTS OF THE SELF-CONTAINED BREATHING APPARATUS (SCBA) AND FIRE PROTECTIVE CLOTHING ON MAXIMAL OXYGEN UPTAKE (VO 2max ) Introduction Firefighting is considered to be one of the most physically demanding and hazardous civilian occupations (Gledhill and Jamnik 1992a and b; Guidotti and Clough 1992). Lusa et al., (1994) determined that regardless of age or rank, one of the most physically demanding tasks faced by firefighters is that of smokediving (search and rescue). This task typically involves entry into a dark, smokefilled structure where the firefighter must search, by feel, for casualties and then evacuate them to safety. Research has shown that search and rescue work during actual fire emergencies elicits near-maximal heart rate responses (Sothmann et al.,1992a) and places a significant demand on aerobic metabolism (Gledhill and Jamnik 1992a; Bilzon et al., 2001) Environmental hazards require that firefighters wear personal protective equipment (PPE) and a self-contained breathing apparatus (SCBA). Although there have been improvements in ergonomic design and reduced weight, there is still evidence to suggest that the protective clothing and respiratory devices used by firefighters negatively affect maximal oxygen consumption (VO 2max ) (O Connell et al., 1986; White et al., 1991; Eves et al., 2005). Maximum oxygen consumption is often measured to ensure that firefighters are physically capable of performing their duties safely and effectively (Gledhill and Jamnik 1992b; Sothmann et al., 1992). However, VO 2max tests are typically performed on a treadmill or cycle ergometer, with the individual dressed in normal exercise clothing (e.g., shorts, t-shirt and running shoes), while breathing through a low-resistance valve, which is not consistent with the PPE required at work (Kilbom 1980; Louhevaara et al., 1985; Gledhill and Jamnik 1992b). Previous investigations have assessed either the effect of the SCBA (Eves et al.,

33 ) or PPE (Louhevaara et al., 1995) on VO 2max, however none have examined the combined effects. Since firefighters are required to work in PPE and wear the SCBA, logically, aerobic fitness should also be evaluated in this condition. The purpose of this study was to investigate the combined effects of fire protective clothing and the SCBA on maximal oxygen consumption Methods Subjects Twelve healthy males, who were very familiar with exercise in firefighting ensemble and breathing apparatus, served as subjects in this investigation. The physical characteristics of the subjects were (mean ± SD): age 31.0 ± 9.1 years; height ± 6.4 cm; mass 83.0 ± 7.1 kg; mass in PPE ± 7.28 kg. Volunteers provided written informed consent for participation in the project that had previously been approved by the appropriate institutional ethics review board. Experimental Design Each subject performed two graded exercise tests (GXT) to determine maximal oxygen consumption. In one condition, subjects were dressed in firefighting personal protective ensemble while breathing through the SCBA (GXT PPE ). In the second condition, subjects were dressed in shorts, t-shirt and running shoes, while breathing through a typical low-resistance breathing valve (GXT PT ). Each test was separated by at least 24 hours during which participants were asked to refrain from strenuous exercise. The order of the tests was randomized. Exercise Test Protocol All tests were carried out at normal room temperature (21-24 ºC). The exercise protocol was performed on a motor driven treadmill (Model ; Standard Industries, Fargo, ND). The same loading protocol was used for both tests. Each subject initially walked at a constant speed (93.8 m min -1 ) while grade increased by 2% every two minutes until 10% grade was reached.

34 22 Subsequently, grade was increased by 2% every minute until 20% grade was reached. If the subject was able to continue, combinations of grade (2% increments) and/or speed (13.4 m min -1 increments) increases were individually selected in order to elicit volitional exhaustion. The highest 20 second VO 2 reading was accepted as VO 2max if at least two of the following criteria were met: a plateau in oxygen consumption was observed despite an increase in work rate; a respiratory exchange ratio (RER) >1.10; and/or, a heart rate greater than or equal to age-predicted maximum was reached. Power output at VO 2max was calculated using the equation [mass x speed x grade]. The mass used in this calculation was the combined weight of the subject and any clothing and equipment carried during the test (either PT or PPE condition). Clothing During the GXT PT condition, subjects wore athletic shorts, t-shirt and running shoes. During GXT PPE subjects added duty coat and pants (System 300 CGSB; SafeCo MFG. Inc., Scarborough, ON), helmet (Model 911; SafeCo MFG. Inc., Scarborough, ON), anti-flash hood (Model: 30176; PGI, Inc., Green Lake, WI) and firefighter gloves (Model: Firefighter, The Glove Corp., Alexandria, IN). The subjects also wore a Scott 4.5 SCBA system (Scott Health and Safety, Monroe, NC) comprising a face piece assembly (AV 2000), regulator (Scott Presur-Pak, E-Z Flo ), and backpack (harness and a full 60-min Scott Air-Pac fibre composite air cylinder). The total weight of the equipment and SCBA was 21.4 ± 1.8 kg. Measurements Respiratory gas exchange (oxygen consumption [VO 2 ], carbon dioxide production [VCO 2 ], respiratory exchange ratio [RER], fraction of expired oxygen [F E O 2 ] and carbon dioxide [F E CO 2 ]) and ventilatory data (ventilation [V E ], tidal volume [V T ], breathing frequency, inspiratory and expiratory time) were acquired during each GXT with a TrueMax 2400 (ParvoMedics, Salt Lake City, UT) computerized metabolic measurement cart (MMC). The gas analysers were

35 23 calibrated immediately prior to each test using gases of known concentration. Calibration of the gas analyzers was checked immediately following each test to verify that calibration had been maintained during the data collection period. The pneumotachometer (Hans Rudolf 3813, Kansas City, MO) was calibrated according to manufacturer specifications using a 3-L syringe (Hans Rudolf 5530 series). In the GXT PT condition, expired gasses were collected via a Hans Rudolph 2700 series low-resistance breathing valve. In order to capture expired gases during the GXT PPE, the SCBA regulator was fitted with a Plexiglas cone that formed an airtight seal over the exhalation ports (for a detailed description see Eves et al. 2002). The distal end of the cone and breathing valve were connected to the MMC via a two meter long corrugated flexible plastic hose with an inside diameter of 2.8 cm. The same hose was used for the GXT PT condition. Prior to experimental testing, the SCBA regulator was tested by an authorized technician to ensure proper working order. Statistical Analysis Respiratory gas exchange data were averaged for each minute during the exercise protocol, except at maximal exercise where the highest 20 s reading was used to detect VO 2max. Differences between conditions were analyzed using a one-way repeated measures analysis of variance. Significant F-ratios were examined on a post hoc basis using the Scheffe procedure for multiple comparisons. Pearson Product-Moment correlation coefficients were used to examine the relationships between variables of interest. Descriptive statistics are presented as means and standard deviations. A probability value equal to or less than 0.05 was considered significant Results All subjects completed both test protocols up to and including 18% grade. Figures 2.2-1, 2.2-2, 2.2-4, and were drawn to show the physiological

36 24 responses for all subjects. The max data point on these figures represents the average maximum value for each variable regardless of the workload. Gas Exchange Responses The oxygen cost of walking on the treadmill in the GXT PPE condition was significantly greater than the GXT PT condition at all grades up to 18% (Figure 2.2-1). At peak exercise, oxygen consumption and carbon dioxide production during GXT PPE were significantly lower compared to the GXT PT condition. It should be noted that power output at VO 2max was significantly higher in the PT condition than in the PPE condition (354 ± 41 W vs. 321 ± 38 W). At the ventilation rate corresponding to VO 2max during the GXT PPE, the F E O 2 was significantly higher while F E CO 2 was significantly lower during the GXT PPE than the GXT PT condition (Table 2.2-1). The respiratory exchange ratio and the ventilatory equivalent (V E /VO 2 ) were not significantly different between conditions at maximal exercise (Table 2.2-1). Ventilatory Responses Ventilation during GXT PPE was significantly higher during each of the test stages 4% to 18% grade; however, V E was significantly lower at maximal exercise by 14.3 ± 9.9% (Figure and Table 2.2-1). Correlation analysis was used to examine the relationship between VO 2max and V E at maximal exercise. The reduction in VO 2max during the PPE condition was highly related (r=0.81) to the decrease in V E (Figure 2.2-3). Tidal volume and breathing frequency displayed different patterns for both GXT PPE and GXT PT (Figures and 5). Tidal volume during GXT PPE was significantly higher during the earlier stages of exercise but lower during the later stages and at maximal exercise V T was lower than the GXT PT condition (Figure and Table 2.2-1). Breathing frequency was significantly higher during most of GXT PPE but there was no significant difference at maximal exercise (Figure

37 ). Inspiratory time during the GXT PPE was significantly shorter than GXT PT at maximal exercise, whereas expiratory time was the same (Table 2.2-1).

38 26 Table Physiological responses at maximal exercise (mean ± SD) Variable GXT PT GXT PPE Oxygen consumption (ml. kg -1. min -1 ) 52.4 ± ± 5.7* Oxygen consumption (L. min -1 ) 4.32 ± ± 0.4* Carbon dioxide production (L. min -1 ) 5.31 ± ± 0.5* Fraction of expired oxygen (%) ± ± 0.28* Fraction of expired carbon dioxide (%) 4.32 ± ± 0.26* Respiratory exchange ratio 1.24 ± ± 0.1 V E /VO ± ± 3.5 Pulmonary ventilation (L. min -1 ) ± ± 18.0* Tidal volume (l) 3.17 ± ± 0.4* Breathing frequency (breaths. min -1 ) 53 ± 7 55 ± 7 Inspiratory time (s) 0.48 ± ± 0.06* Expiratory time (s) 0.72 ± ± 0.07 Heart rate (beats. min -1 ) 188 ± ± 7 * = significant difference between GXT PT and GXT PPE (p <0.05).

39 27 Oxygen Consumption (ml. kg -1. min -1 ) * * GXT PT GXT PPE * * * * * * * Grade (%) max Figure Mean (±SD) oxygen consumption during GXT PT and GXT PPE. * = significant difference between GXT PT and GXT PPE.

40 GXT PT GXT PPE * 160 * Ventilation (L. min -1 ) * * * * * * * Grade (%) max Figure Mean (±SD) pulmonary ventilation during GXT PT and GXT PPE. * = significant difference between GXT PT and GXT PPE.

41 % Change in VO 2max % Change in V E Figure Scattergram of the fractional change in VO 2max and peak ventilation between GXT PT and GXT PPE.

42 GXT PT GXT PPE * * Tidal Volume (L) * * * * * * Grade (%) max Figure Mean (±SD) tidal volume during GXT PT and GXT PPE. * = significant difference between GXT PT and GXT PPE.

43 31 70 GXT PT Breathing Frequency (breaths. min -1 ) * * GXT PPE * * * * * Grade (%) max Figure Mean (±SD) breathing frequency during GXT PT and GXT PPE. * = significant difference between GXT PT and GXT PPE.

44 Discussion The submaximal data presented in this study are comparable with previous research, showing that at equivalent external workloads, VO 2, V E, and heart rate are higher in PPE conditions versus PT (Davis and Santa Maria 1975; Louhevaara et al., 1985; O Connell et al., 1986; White et al., 1991; Lusa et al., 1993). It has been suggested that the increased physiological responses during submaximal exercise in PPE are attributed to a variety of factors. Aspects such as increased weight of the equipment (Louhevaara et al., 1985; Duggan 1988), a hobbling effect (Teitlebaum and Goldman 1972), or pronounced forward lean (Soule and Goldman 1977; Gordon et al., 1989) could individually or in combination affect the efficiency of exercise and work tolerance. The main finding of this investigation was the substantial reduction in VO 2max in the PPE condition compared to PT. The 18% reduction is similar to the results of a study by Louhevaara et al., (1986) where the subjects wore only the SCBA facemask, harness and tank (~17%). More recently, Eves et al., (2005) reported similar reductions in VO 2max that were explained mainly by the resistance of the SCBA regulator. Louhevaara et al., (1995) reported that VO 2max of subjects wearing firefighting clothing alone was reduced by 4% (ns), which supports that most of the attenuation of VO 2max is caused by the breathing apparatus. The results from the current study indicate that the decrease in maximal V E during the GXT PPE was the main reason for the decreased VO 2max. The cause for the decreased V E while breathing from the SCBA is not clear, but may be related to an increased work of breathing due to expiratory resistance caused by the regulator (Eves et al., 2002a and b) and/or impairment in thoracic excursions due to the weight of the SCBA harness (Louhevaara et al., 1985). Tidal volume plateaued at approximately 80% of VO 2max during the PPE condition which corresponded to 83% of maximal V T during the PT condition (Figure 2.2-4). Inspiratory time was significantly shorter in the PPE condition, however despite

45 33 the reduction in tidal volume, expiratory time was not changed. This suggests that more effort was required to exhale against the resistance of the regulator. While maximal breathing frequency was not different in the two conditions, there is clear evidence that when tidal volume reached a plateau in the PPE condition (Figure 2.2-4), ventilation was maintained by a rapid increase in breathing frequency (Figure 2.2-5). The SCBA may cause a change in alveolar ventilation (V A ) during heavy exercise (Donovan and McConnell 1999a), which in turn could contribute to lower arterial saturation (S a O 2 ). When the expired fraction of O 2 and CO 2 at similar V E were compared between test conditions, significantly higher F E O 2 and lower F E CO 2 were found in the PPE condition. Eves et al. (2002), using pulse oximetry to predict S a O 2, found that during a PPE GXT, a moderate level of exercise-induced hypoxemia occurred, which may be related to an inadequate V A. These observations suggest the possibility of altered gas exchange during heavy exercise with the SCBA, however for confirmation, further research is required. The effects of the SCBA and PPE on VO 2max have some practical implications regarding the health and safety of firefighters. Several researchers have measured or estimated oxygen consumption during firefighting work (Sothmann et al., 1990; Gledhill and Jamnik 1992a; Lusa et al., 1993; Bilzon et al., 2001). This research indicates that fire suppression and rescue work typically requires an average VO 2 of approximately 34 ml. kg -1. min -1. Based on the understanding of the aerobic demands of fire rescue work, recommendations for minimum VO 2max levels for firefighters range from ml. kg -1. min -1 (O Connell et al., 1986, Gledhill and Jamnik 1992b; Sothmann et al., 1992; Bilzon et al., 2001; NFPA 2003), with the average being approximately 42 ml. kg -1. min -1. Typically, the oxygen cost of firefighting has been determined during work simulations while wearing PPE and SCBA. However, none of the recommendations for VO 2max have included guidelines for equipment or clothing configurations during testing.

46 34 By way of example, if the average VO 2 during fire rescue work is assumed to be 34 ml. kg -1. min -1, and the average recommended VO 2max is 42 ml. kg -1. min -1, the aerobic demands of the work would represent approximately 80% of VO 2max. However, as shown in the present study, the VO 2max in PPE and on SCBA is significantly reduced from the level measured in the traditional test. Therefore, the firefighter would have to work at a higher fraction of maximal aerobic power. The requirement to work at greater than 80% of VO 2max may compromise the reserve for safe and effective performance and total potential work time (Louhevaara et al., 1986; Ilmarinen 1992a and b). Consequently, the results of this study need to be considered when evaluating aerobic fitness standards for firefighters Conclusions It is evident that firefighter protective equipment and especially the SCBA, has a significant negative effect on maximal oxygen consumption. Nearly two decades of equipment design have not alleviated the negative effects of the SCBA on VO 2max as reported by previous researchers. Therefore, when testing the aerobic fitness of firefighters, the effects of the SCBA and protective clothing on maximal work capacity must be taken into consideration. A logical alternative to the practice of testing aerobic fitness in traditional exercise clothing would be to adopt a method of testing similar to that described above. This approach to measuring VO 2max should provide a more functional assessment of aerobic work capacity in firefighters.

47 AEROBIC DEMANDS OF FIRE-RESCUE TRAINING SCENARIOS Introduction In contrast to the relative abundance of research describing the aerobic requirements of firefighting work in males, there has been only one study published in the scientific literature that examined the VO 2 responses of females during simulated firefighting work (Bilzon et al., 2001). This investigation studied 34 male and 15 female Royal Navy personnel who had a minimal level of shipboard firefighter training (3 days). Each subject performed five firefighting activities while dressed in appropriate ensembles of protective equipment. Each of the five activities was meant to be completed at a prescribed work-rate for 4- minutes. The results of this study reported the mean oxygen cost for females and males performing simulated firefighting work to be ml. kg -1. min -1, with a range of ml. kg -1. min -1. This is similar to most of the values reported in the literature for males, but it is important to recognize that the number of females studied was very small on the most demanding activity (DC), so the results should be interpreted with some degree of caution. With this one exception, the study by Bilzon et al., (2001) was well-designed, and the data are both interesting and valuable. However, one might question whether the results would apply to structural firefighting where the activities are somewhat different. As previously noted, there are several methods of simulating firefighting work: completion of individual tasks; completion of a predetermined series of tasks in a circuit format; and, completion of a series of tasks attempting to replicate the problem-solving requirements of actual fire-rescue work. The latter has been described as a scenario and has previously been utilized in the scientific literature (Sothmann et al., 1990; 2004).

48 36 The purpose of the present study was to document the aerobic demands of firerescue training scenarios in males and females Methods Participants Thirteen males and 12 females volunteered to participate in this investigation. All participants were healthy and reported regular physical activity patterns at the time of enrollment. Participants responded to word-of-mouth invitations and poster advertisements posted in local fitness centers. The intent was to simulate the work done in a fire-rescue training environment, and consequently, an interest in firefighting was essential, however previous experience was not. The physical characteristics of the subjects at entry are reported in Table Each subject provided written informed consent to participate in the study that had been previously approved by the appropriate institutional Research Ethics Board. Experimental Design The participants completed the following stages of the research project: orientation to firefighting personal protective equipment and the CF/DND FF Test circuit; 3-6 practice sessions of the circuit in firefighting ensemble while breathing from the SCBA; graded exercise test (GXT) to determine maximal oxygen consumption (VO 2max ) in firefighting ensemble (FPE); selected theoretical and practical elements of fire-rescue training to acquire specific skills related to indoor hose attacks, search, rescue and removal of victims to triage; and, two fire-rescue training scenarios where respiratory gas exchange data were acquired.

49 37 All stages of the research were carried out indoors in a thermo-neutral environment (21-25 ºC). Firefighting Protective Equipment (FPE) During each stage of the research project, subjects dressed in National Fire Protection (NFPA) Standard 1500 (NFPA, 1997) duty coat (System 300 CGSB; SafeCo Mfg. Inc., Scarborough, ON), firefighting pants (System 300 CGSB; SafeCo Mfg. Inc., Scarborough, ON), helmet (Model 911; SafeCo Mfg. Inc., Scarborough, ON), anti-flash hood (PGI, Inc, Green Lake, WI), leather work gloves (CF/DND FF Test) or firefighting gloves (training and scenarios), and rubber firefighting boots (Black Diamond, Kaufman Footware, Kitchener, ON), except during the GXT where running shoes were worn in the interest of comfort and safety. Subjects wore a Scott 4.5 harness with 60-min Scott Air-Pac fibre composite air cylinder (Scott Aviation, Monroe, NC). During the CF/DND FF Test and training, the subjects breathed from the SCBA. During the GXT and the scenarios, they carried the SCBA with a fully charged cylinder. Simulated Firefighter Work Circuit The CF/DND FF Test (Deakin et al., 1996) was used in this study for two reasons. First, it was essential that the subjects were able to demonstrate the capacity to do physically demanding work with firefighting PPE and the SCBA. The ability to complete the CF/DND FF Test provided evidence of an appropriate physical capacity to enter the fire-rescue training phase. Second, as completing the CF/DND FF Test correctly involves a significant amount of equipment handling (e.g., advancing charged hose, climbing ladders, dragging victims, etc), the test was a very appropriate vehicle for practicing these fundamental workrelated activities. This test was set up indoors on a concrete floor where all the course dimensions and equipment (type and weight) were completely accurate.

50 38 Graded Exercise Test (GXT) Maximal oxygen consumption was determined while walking on a motor driven treadmill (Model # , Standard Industries, Fargo, ND). The GXT was performed in the following four stages: warm-up 5-minutes at a constant speed (93.8 m. min -1 ) with a progression from 0-6% grade; 8-minutes at 10% grade and 93.8 m. min -1 ; grade was increased 1% every minute until 15%, then speed was increased by 13.4 m. min -1 each minute until volitional exhaustion; 5-minutes of cool-down at 0% grade and 53.6 m. min -1. The highest 1-minute VO 2 reading was accepted as VO 2max if a plateau in oxygen consumption was observed despite an increase in work rate, or alternately, if the subject was too fatigued to continue exercise. Physiological Measurements Continuous gas exchange measurements (e.g., VO 2, VCO 2, V E, and RER) during the GXT and the scenarios were made with a VmaxST (SensorMedics, Yorba Linda, CA) portable metabolic measurement system (MMC). Data were acquired in breath-by-breath mode and subsequently averaged over appropriate time periods for analysis. The software version was Meta Soft (Cortex Biophysik, Leipzig, Germany). Each subject wore a specially designed Hans Rudolph 8930 series full-face mask with an Ultimate Seal TM gel (669204/201236) and head cap assembly (P/N ). The gas analyzers were calibrated immediately prior to each test using gases of known concentration and calibration was verified immediately following the test. The Triple-V flow volume transducer was calibrated using a 10-stroke calibration of a 3.00-liter Hans Rudolph 5530 series syringe. In all conditions, subjects wore a Polar telemetry system (Vantage NV, Polar USA, Inc., CT), which continuously measured and recorded heart rate at 5- second intervals during exercise. A Polar Advantage computer interface operating with Polar HR Analysis Software Version (Polar Electro, OY,

51 39 Finland) was used to transfer the heart rate data to a computer for analysis. Rating of perceived exertion (RPE: Borg, 1982) was recorded at the end of each minute of exercise. Development of Scenarios or Work Samples Development of the work sample phase of the research project was a critical step requiring a significant amount of organizational support from DND. Our approach was to devise a series of reproducible tasks that reflected the physical demands of critical elements of firefighting work in terms of load patterns, intensity, and duration. Each participant ran through a series of repetitions of each task in a manner consistent with what would typically occur during fire training or fire suppression work. Initially, most of the participants were relatively unskilled in fire-rescue operations. We recognized the need to provide training so that the participants were able to execute the requisite skills related to certain aspects of firefighting work. The training was extensive enough that the participants could perform the required tasks in a realistic manner to the satisfaction of the subject matter experts (SME) tasked to administer the training. We recognized the possibility that different teams of subjects would move through the scenarios at different rates and potentially with a different order of events. The scenarios were designed to maximize consistency in these factors as much as possible so that we could compare results for parts of the scenario (e.g., split times for scenario segments) rather than simply the overall completion time. Finally, to make comparisons between the male and female participants, it was essential that the tasks were reproducible to the extent that any physiological differences could be explained by gender rather than some other factor. The training program and scenario development occurred as follows: The principal investigators from the U of A (S. Petersen and R. Dreger) met with selected personnel at the Canadian Forces Fire Academy

52 40 (CFFA) at CFB Borden. The main objective was to define the nature of work samples that would reflect the physical demands of the most important aspects of structural fire training. Staff at CFFA provided input on how the work samples could be reproduced in a remote facility; SME from CFB Edmonton were tasked to develop the training facility. A vacant hangar adjacent to the fire hall provided a large indoor space with a variety of rooms on two levels. The windows were blacked out to eliminate natural light; A series of fire-rescue scenarios were developed for use in training the participants. Two scenarios were selected for use during the data collection phase; The CFFA Standards Officer and local SME (from CFB Edmonton) identified the methodology and protocols for training the participants. An essential part of this step was to establish criteria to evaluate the skill level of the participants before proceeding to the data collection phase; The investigators provided a small group of participants to pilot the training plan and work sample data collection; The Standards Officer from CFFA visited the training site in Edmonton. The main outcome was that the delivery of the training program and the running of the scenarios met his satisfaction for standards of safety and effectiveness; and, After the pilot research was completed, the actual data collection began. Fire-Rescue Training Plan Practicing the CF/DND FF Test ensured that the participants had some of the basic skills related to firefighting work, (e.g., using the SCBA, advancing charged hose, raising and climbing ladders) prior to starting the work sample phase. The Standards Officer from CFFA developed a training plan that was delivered by senior staff from the fire department at CFB Edmonton. Successful completion of the training plan ensured that the subjects were adequately skilled to perform the scenario tasks in a safe and effective manner. The training involved both

53 41 theoretical and practical elements of fire-rescue work. The training plan was administered to small groups of 4-6 research participants by 3-4 instructors. Two of the instructors had previously served at CFFA and were fully qualified to teach basic firefighting courses. The senior instructors were assisted by qualified firefighters from CFB Edmonton. In the practical component, the research subjects were introduced to individual skills (e.g., hose handling, stair procedures, casualty evacuation, etc). They then progressed to more complex iterations involving partner work. The subjects were taught to work in teams of two with carefully defined roles under the job-descriptions of Rescue and Nozzle. Each subject was required to demonstrate competency in the skills associated with both positions. The teams of two were then introduced to simple search and rescue situations in blacked-out rooms filled with non-toxic smoke. The research subjects were always under the direct supervision of one of the training instructors who typically adopted the role of a supervising firefighter. That is, the instructor would coordinate the efforts of the two firefighters, but would not assist them in their own work. Over the training period the scenarios became more complicated and ended with searches of up to eight blacked-out, smoke filled rooms on two levels involving two fire props, two hose-lines, and two victims that had to be evacuated up or down stairs. A wide variety of furniture props (e.g., tables, chairs, sofas, file cabinets, desks) were used to change the training environment. The final scenario completed in the training program was equivalent to the complexity of the data collection scenarios. Upon completion of the training plan, each participant met the criteria for safe and effective completion of the essential tasks under the two job titles. The supervising instructor completed a checklist consistent with the expectations for fire students involved in the National Fire Protection Agency 1001 training program. The skills are listed in Table under the appropriate job title.

54 42 Fire-Rescue Scenario Data Collection As noted above, two scenarios were developed that each included: 8 blacked out rooms filled with non-toxic smoke on two levels staircase between levels two fire props two hose-lines (charged 38 mm hose) two victims (70 kg Rescue Randy mannequins) to be found and removed to triage area 15 m from point of entry When completing Scenario One, the fire-rescue team entered on the ground floor and finished their search on the upper level. This scenario ended when the second victim was evacuated down the stairs to the triage area. When completing Scenario Two, the fire-rescue team entered on the upper floor and finished their search on the lower level. This scenario ended when the second victim was evacuated up the stairs to the triage area. In each case, the triage area was located 15 m from the point of entry into the search area. The order of the scenarios (One or Two), the assignment of team members to job positions (Rescue or Nozzle), and the order of the run (first or second) were randomly assigned. The furniture props were consistently arranged for each scenario. Each participant completed two scenarios, one as Rescue and one as Nozzle. Respiratory gas exchange data and heart rate were recorded throughout the scenario using the portable metabolic measurement system (VmaxST) and Polar telemetry system, respectively. Prior to the experiment, pilot studies revealed that the VmaxST metabolic system operated normally in the non-toxic smoke. The measurement of minute ventilation, VO 2, and VCO 2 were not altered by the smoke. Measurement variables for each scenario included: heart rate; blood lactate; respiratory gas exchange;

55 43 perceived exertion; split times for 5 designated segments; and, overall completion time. Two metabolic systems were used so that each participant could be monitored. Research assistants recorded the elapsed times for each aspect of the scenarios. Ultimately, the time measurements were collapsed into a series of consistent split times as follows: Split 1 This included the elapsed time from start of scenario on floor one to finding the first victim. Split 2 This included the elapsed time between finding the first victim and removing the victim to the triage area. Split 3 This included the elapsed time required to leave the triage area, re-enter the search area, find the first fire, get the second hose-line, complete the search on floor one and move to floor two. Split 4 This included the elapsed time from point of entry on floor two to finding the second fire and second victim. Split 5 This included the elapsed time required to remove the second victim to the triage area. As soon as the second victim had been safely delivered to the triage area, the scenario was finished. Most of the FPE was removed as quickly as possible and at exactly five minutes post-exercise a small (~1 ml) blood sample was drawn by venipuncture from a convenient forearm vein. A 200 µl aliquot of blood was immediately added to a Yellow Springs Instrument 2315 lactate preservative tube (YSI; Yellow Springs, OH). The tube was capped and the contents were mixed by vortexing. The tubes were stored at -80 C until analyzed in duplicate using a YSI 2300 lactate analyzer. The analyzer was calibrated after every 5 th sample. The mean of each set of duplicates was used in the data analysis. No differences greater than 0.3 mmol. L -1 were found between duplicate samples.

56 44 Statistical Analysis Descriptive statistics, paired and unpaired t-tests and repeated-measures analysis of variance (ANOVA) were used for the statistical analysis of the data. All statistical analyses were performed using StatView (SAS Institute Inc., Carry, NC) with the exception of the ANOVA, which was conducted with Statistica Version 7 (StatSoft, Tulsa, OK). A probability value 0.05 was considered significant Results The physical characteristics of the participants are displayed in Table The males and females were similar in age, however the females were shorter and lighter than the males. The differences in size have important implications for some of the physiological differences in oxygen consumption and ventilation that are reported later. The average weight of the FPE was 23.9 and 23.5 kg for males and females, respectively. The FPE weight, when expressed as a fraction of body mass, represented 28 and 35% of body mass for males and females, respectively. Absolute VO 2max was higher (p<0.05) in the male group, but was the same when aerobic power was expressed relative to body mass. Based on published fitness classifications (McArdle et al., 2001, p. 163), the mean relative VO 2max of both groups would be categorized as good, with individual scores ranging from average to excellent. Table displays the individual fire-rescue competencies that were covered in the training plan developed by the Standards Officer at CFFA and delivered by the senior instructors from the fire department at CFB Edmonton. Each research subject received a passing grade on each skill prior to the experimental data collection. Each subject completed two scenarios, one in the rescue position and one in the nozzle position. As noted previously, the number (Scenarios One and Two) and order of the scenarios (first or second run of the day) and the job positions

57 45 (Rescue and Nozzle) were randomly assigned. There were no significant differences with respect to completion time or physiological responses (e.g., VO 2 ) between Scenario One and Two, between Rescue and Nozzle or between the first and second run of the day. Consequently, all scenario data were collapsed and are presented as 26 male observations and 24 female observations (Tables 2.3-3, 2.3-4, and 2.3-5). Table displays the split and overall scenario completion times for males and females. Typically, the females required more time (p<0.05) to complete the scenarios, however it must be stressed that the subject matter experts were completely satisfied that the work was done at an acceptable rate taking both safety and performance into consideration. On average, the females took approximately 17 s longer to find the first victim (Split 1) and approximately 24 s longer to extricate the first victim to triage (Split 2). There were no gender differences for the segments of the scenario (Splits 3 and 4) where the subjects returned from triage, completed the balance of the search on the first level, found the first fire, brought in the second hoseline, moved to the second level (either up or down), completed their search of the second level and found the second victim. The final segment (Split 5) required the subjects to extricate the second victim to triage and this involved carrying the mannequin either up or down the stairway. Notwithstanding that the task was accomplished safely and at an acceptable work-rate, the females required, on average, 1.95 minutes longer to complete this segment. It should be noted that, without exception, the subjects reported that the segment was the most physically demanding aspect of the entire scenario. The physiological and psycho-physical responses for males and females separately and all subjects combined are reported in Tables and Absolute oxygen consumption and minute ventilation were higher in the males. There were no differences in heart rate, perceived exertion, or blood lactate between males and females.

58 46 The pattern of oxygen consumption responses for one male and one female in the Nozzle position during Scenario Two is displayed in Figure These subjects were selected for comparison since the overall times for completion of the scenario by these subjects were approximately 11 minutes. The VO 2 responses of these two subjects (same position, same scenario, same completion time, but different partners) were remarkably similar.

59 47 Table Mean (±SD) physical characteristics and aerobic power of male (n=13) and female (n=12) participants. Variable Male Female Age 25.6 ± ± 3.4 Height (cm) ± ± 6.4* Body mass (kg) 86.2 ± ± 6.1* Body mass with FPE (kg) ± ± 6.7* VO 2max (L. min -1 ) 3.72 ± ± 0.40* VO 2max (ml. kg -1. min -1 ) 44.2 ± ± 4.0 *p<0.05 (significant difference between males and females)

60 48 Table Summary of fire-rescue competencies covered in the training plan. Fire-Rescue Skill Rescue Position Nozzle Position Personal Protective Equipment X X Donning and use of SCBA X X Hose handling techniques Support in advancing hoseline Door procedures Stairway procedures Search techniques X X Casualty evacuation X X Communication X X Team effort X X Response to commands X X Safety X X X X X X

61 49 Table Mean (±SD) split and total times for fire-rescue scenarios for male (26 observations) and female (24 observations) subjects. Scenario Time Segment Males (min) Females (min) Mean Difference (min) Split (0.41) 1.16 (0.45) 0.28* Split (0.27) 1.16 (0.45) 0.40* Split (1.08) 4.14 (1.52) 0.37 Split (0.98) 3.29 (1.78) 0.48 Split (0.87) 4.42 (1.46) 1.95* Total (1.77) (2.49) 2.83* *p<0.05 (significant difference between males and females) Time segment notes: Split 1 start on level one to first victim Split 2 first victim to triage Split 3 from triage to start of search on level two Split 4 to second victim Split 5 second victim to triage Total total time to complete the scenario

62 50 Table Oxygen consumption and ventilation responses during fire-rescue scenarios for male (26 observations) and female (24 observations) subjects. Variable Males Females Combined Avg VO 2 (ml. kg -1. min -1 ) 27.1 (3.5) 26.1 (3.1) 26.6 (3.3) Avg VO 2 (L. min -1 ) (0.297) 1.770* (0.264) (0.390) Avg VO 2 (% VO 2max ) 61.3 (8.7) 59.2 (9.3) 60.3 (9.2) Peak VO 2 (ml. kg -1. min -1 ) 35.3 (4.8) 34.2 (4.8) 34.7 (4.8) Peak VO 2 (% VO 2max ) 79.9 (12.2) 77.1 (10.3) 79.6 (11.4) Avg V E (L. min -1 ) 69.5 (12.1) 54.3* (7.7) 61.9 (12.6) Peak V E (L. min -1 ) 96.3 (16.1) 76.8* (14.8) 86.5 (18.2) Avg V E /VO (3.3) 30.8 (2.4) 30.4 (2.9) Values are mean (±SD) *p<0.05 (significant difference between males and females)

63 51 Table Heart rate, perceived exertion and blood lactate responses during fire-rescue scenarios for male (26 observations) and female (24 observations) subjects. Variable Males Females Combined Avg HR (bpm) 154 (14) 155 (16) 154 (15) Avg HR (% maximum) 79.3 (7.2) 79.2 (8.1) 79.2 (7.6) Peak HR (bpm) 179 (14) 177 (14) 178 (14) Peak HR (% maximum) 92.2 (6.6) 90.4 (7.3) 91.3 (7.0) Avg RPE 14 (1) Peak RPE 16 (1) 14 (2) 16 (3) 14 (2) 16 (2) Blood Lactate (mm) 5.8 (2.2) 5.9 (2.0) 5.8 (2.0) Values are mean (±SD)

64 VO 2 (ml. kg -1. min -1 ) Female Nozzle Male Nozzle 0 Split 1 Split 2 Split 3 Split 4 Split 5 Activity Figure Sample plots of average oxygen consumption for each of the five scenario segments for one male and one female subject in the Nozzle position during Scenario Two. Completion time for both subjects was approximately 11 min. Time segment notes: Split 1 start on level one to first victim Split 2 first victim to triage Split 3 from triage to start of search on level two Split 4 to second victim Split 5 second victim to triage

65 VO 2 (ml. kg -1. min -1 ) Time (min) Figure Relationship between average VO 2 and time to complete a simulated fire suppression protocol. Re-drawn from the data of Sothmann et al., 1990.

66 Discussion The purpose of this study was to document the physiological responses in males and females completing fire-rescue training scenarios. The main finding was that the average oxygen consumption during the scenarios was identical for men and women. All work met SME requirements for safe and effective performance. However, on average, the men completed the work at a faster rate than the women. The female subjects were shorter and lighter than their male counterparts (Table 2.3-1). The added mass of the FPE represented approximately 28 and 35% of the body mass for males and females, respectively. Consequently, the relative burden of the firefighting clothing and SCBA (approximately 23 kg for all subjects) was substantially greater for the female subjects. This is an important point because much of the work encountered in simulations of firefighting tends to be based on absolute loads. That is, the weight of charged hose and rescue mannequins are constant for all subjects. Logically, the absolute loads of the FPE and the object(s) to be moved place a greater relative load on a lighter individual. All other things being equal, a smaller person with less muscle mass will have to work relatively harder than a larger person to accomplish the same absolute task. In firefighting work, the relative workload is primarily related to size, not gender per se. However, since as a group, the females in this study were smaller than the males, the problem of higher relative loading will tend to be more generally applicable to the females. Applying the age and gender stratified norms for Canadians (CSEP, 1986) to the group means for height and weight revealed that our male subjects were ranked at the 85 th and 65 th %ile for body weight and height, respectively, and the females at the 85 th and 70 th %ile, respectively. This suggests that, on average, our male and female subjects tended to be larger than average, but equally so compared to their reference groups. While the absolute VO 2max was lower for the females, when adjusted for body mass, the difference in VO 2max disappeared.

67 55 The mean VO 2max for both gender groups was classified as good (McArdle et al., 2001, p. 163). The physical characteristics of the subjects in this project were very similar to the mean values for several thousand firefighter applicants tested in our laboratory during the past decade (unpublished observations). In summary, based on their interest in firefighting and their physical characteristics, the participants were very representative of the type of individual entering the fire service. The scenarios were developed to present a series of physical challenges related to searching with charged hose-lines, and finding and removing victims to safe areas. The two scenarios were configured to present similar challenges in roughly the same order so that physiological responses could be compared. As noted previously, sources of natural light were eliminated and the rooms were filled with non-toxic smoke. Therefore, the research subjects were forced to work without the benefit of vision and spatial orientation. These environmental factors influenced the work-rate in several ways. First, most of the searching was accomplished by feel and the subjects were frequently bent over or crawling rather than walking upright. Secondly, because of the sensory deprivation and the elements of uncertainty (number and location of fire-props, victims, etc), occasionally the subjects would become momentarily disoriented and have to re-group before proceeding. Finally, tasks such as bringing in the second hoseline and extricating the victims were more difficult. Collectively, these factors had the effect of slowing down the rate of movement through the training area. It bears repeating that the firefighter teams were under the constant supervision of a training officer. This individual would give the subjects directions in the same manner that a supervising firefighter would direct junior firefighters during an emergency. The supervisor also was charged with ensuring that the research subjects completed the scenario following correct procedures and executing the skills that had been taught in the training plan. The research subjects met all of the previously determined criteria for safe and effective completion of the work.

68 56 Average oxygen consumption was calculated for each subject during each segment of each scenario. The mean responses are shown in Figure 4-1. The first time segment was relatively short in duration (approximately one minute) and involved mostly searching. The second time segment was of similar duration but involved the more strenuous work of removing the first victim to triage. The third and fourth segments were longer in duration (approximately 3-4 minutes each) and involved searching, advancing charged hoses, and moving down stairs. The final segment in the scenario required moving the second victim up the stairs to triage, and lasted approximately 3-4 minutes. There were no differences in the VO 2 responses of men or women during any of the split times. Without exception, all subjects rated the final activity of moving the victim up or down the stairs as the most physically demanding aspect of the scenarios. The peak RPE shown in Table was associated with this activity. While there was significant variability in the actual VO 2 and the duration of each time segment, the pattern was quite consistent. That is, after the first brief time segment, the rate of oxygen consumption was elevated to a relatively high level and remained quite consistent for the balance of the scenario. Sample plots of average oxygen consumption during each of the five scenario segments are shown (Figure 2.3-1) for representative male and female subjects in the Nozzle position. Both individuals completed the scenario in approximately 11 minutes. The main physiological results from the scenarios are displayed in Tables and When reviewing these data, it should be remembered that each scenario involved an absolute amount of work that had to be accomplished. That is, each scenario involved 8 rooms on two levels, a staircase, two charged hoses, two fire props and two victims. At the design stage, great care was taken to try to equate the work between the two scenarios. However, the subjects were not informed of the specific work involved. They were instructed to search the

69 57 entire area for fire and victims following the fire-rescue procedures learned during training. The reader is also reminded that while the females were generally smaller in size, the relative VO 2max was the same and the mean VO 2max for both groups was classified as good. Table shows the main respiratory gas exchange results for the scenarios. The values for the two groups are remarkably consistent, suggesting that when aerobic fitness is equated, the oxygen cost of accomplishing a specified amount of firefighting work is the same for men and women. The significant differences in absolute VO 2 (L. min -1 ) and minute ventilation (V E, L. min -1 ) are explained by the difference in body mass. When the VO 2 is expressed relative to body mass (ml. kg -1. min -1 ), the difference disappears. The ventilatory equivalent for oxygen (V E /VO 2 ) was not different, which is logical because the minute ventilation and absolute VO 2 are proportional regardless of gender. The V E /VO 2 ratio is frequently used during graded exercise testing as an indicator of metabolic stress. The values shown in Table are clinically normal and are consistent with strenuous work. The average work rate for both males and females represented approximately 60% of the VO 2max. The peak VO 2 during the scenarios was similar for males and females at approximately 80% of VO 2max. The peak minute ventilation showed an increase of approximately 40% above the average value for both males and females. Rough calculations of total ventilation made from average minute ventilation and the time required to complete the scenarios suggest that total air consumption was the same for males and females at about 750 L. Average and peak heart rate, perceived exertion and post-exercise blood lactate results are shown in Table There were no significant differences between males and females in any of these parameters. The heart rate responses are consistent with expected values during strenuous work and correspond with the fraction of VO 2max. A slightly disproportionate HR response relative to VO 2 might

70 58 be expected since there was a significant amount of upper limb work involved in the scenarios (e.g., dragging charged hose, carrying the victims). As well, while no attempt was made to quantify the psychological responses, it would stand to reason that the requirement to solve the scenarios in the dark, smoky environment would result in some level of anxiety which might be reflected in heart rate. Finally, despite the thermo-neutral environment, the subjects experienced mild thermal stress from working in the protective equipment. The average and peak RPE values indicate that the subjects classified the overall physical exertion during the scenarios as hard. This rating of exertion is not surprising since many authorities have classified firefighting work as strenuous. Furthermore, the scenarios in this research were designed by subject matter experts from the Canadian Forces fire service to be faithful simulations of firerescue work. Lactate is a metabolite that is produced from the anaerobic degradation of glucose. At rest and during moderate exercise, the rate of lactate production is low and normal blood levels are approximately 1 mm. Blood lactate is frequently used as a marker of exercise intensity and/or as an indicator of the degree of reliance on anaerobic metabolism. The blood samples were drawn five minutes after exercise, allowing for equilibration of the lactate metabolite between the site of production (active muscle tissue) and the venous blood. The values of approximately 6 mm are indicative of strenuous work with significant involvement of anaerobic metabolism (McArdle et al., 2001, p 294). Gledhill and Jamnik (1992a) reported somewhat higher blood lactate responses ( mm) to three separate activities that they classified as the most physically demanding firefighting operations. The operations described by Gledhill and Jamnik (1992a) were isolated simulations of hoisting equipment, victim rescue, and stair climbing under load. The simulations were completed as quickly as possible. A fair description of simulations of firefighting work would be brief (1-2 minutes) periods of exercise of near maximal effort. The lactate

71 59 responses reported by Gledhill and Jamnik (1992a) appear very reasonable however, the work done in their research was quite different from the longer duration, lower intensity exercise performed in the current study. In previous sections of this report, some of the factors influencing the aerobic demands of firefighting work have been discussed. In order to make valid comparisons between studies, factors such as the intensity/duration relationship, sensory restriction, and controlled or self-selected work-rate must be considered. For example, comparisons between the present results and the work of Bilzon et al., (2001) are confounded by design factors. In that study, subjects could rely on sight to help navigate the task simulations, the work times were fixed at four minutes and the work-rate was controlled. Considering the differences in the way the work was done, it is not surprising that the VO 2 reported in that study was higher than the present results. On the other hand, a noteworthy and legitimate comparison shows that the VO 2 responses for males and females were the same in both studies. The aerobic demands of the scenarios in this project were very similar to the findings reported by Sothmann et al., (1990). In that study, male firefighters completed a circuit consisting of seven tasks in a hot (130ºC), smoky (non-toxic) environment. The firefighters were asked to complete the tasks at a self-selected work-rate that was consistent with their experience in real emergencies. A device to measure oxygen consumption was interfaced with the SCBA and VO 2 data were recorded three times during the circuit. These values were averaged to determine a VO 2 value representative of the work protocol. The average duration of the protocol was approximately nine minutes (range: 5:30 13:54) and the mean VO 2 value was 30.5 ml. kg -1. min -1. Figure was drawn from the data of Sothmann et al., (1990), and shows the relationship between average VO 2 and time to complete the firefighting work simulation.

72 60 Successful completion of the protocol was defined as mean time plus two standard deviations (13:06) and the VO 2 associated with that criterion was 25.5 ml. kg -1. min -1. In the present study, the average time for our male and female subjects to complete the scenarios was and minutes, respectively. The average VO 2 calculated from continuously recorded data was 27.1 and 26.1 ml. kg -1. min -1 for males and females, respectively. While the specific details of the work simulations were different, the work time and VO 2 responses were similar. The results of these two studies suggest that experienced firefighters or firefighter trainees working for approximately minutes during a simulated fire-rescue protocol are required to sustain a VO 2 of approximately ml. kg -1. min -1. On the basis of their observations, Sothmann et al., (1990) speculated that a VO 2max of at least 41 ml. kg -1. min -1 was required in order to ensure successful completion of their work protocol. A number of authorities have noted the importance of muscular strength and endurance in firefighting. Gledhill and Jamnik (1992a) summarized the results of a comprehensive task analysis by concluding that: the most commonly encountered muscular strength and endurance applications required of firefighters under emergency conditions were (a) lifting and carrying objects (e.g., sand pail of 79 lb, nozzle and halligan tool of 72 lb); (b) pushing, pulling or dragging objects (e.g., raising ladders of 135 lb, hoisting equipment via ropes of 111 lb, dragging victims of 200 lb or more); (c) working with objects in front of the body (e.g., extrictator of 72.7 lb, removing 67-lb ladder from truck). (pp ) Bearing in mind that most of this work would normally be done while the firefighter is wearing protective clothing and SCBA (typical weight of kg), it should be obvious that there is a significant strength component in firefighting work.

73 61 As previously noted, all the subjects in this experiment reported that carrying the 70 kg mannequin up or down the stairs was the most difficult aspect of the scenarios. The most substantial difference between gender groups in split times for the various scenario segments was found in the last segment that involved this difficult activity. There is little doubt that the females took longer to complete this segment because of differences in body mass and strength. The work-rate was acceptable to the supervising training officer but was significantly slower than that of the larger, stronger males. There was no gender difference in the VO 2 during this segment, which further supports the suggestion that the performance limitation was due to strength, not aerobic fitness Summary In summary, after appropriate training, 25 young, healthy, physically active subjects completed two fire-rescue scenarios of approximately the same configuration and difficulty. Subject matter experts from the Canadian Forces fire service were responsible for the design and delivery of the training plan and the design of the scenarios. The SME also supervised each of the research subjects during each scenario to ensure that the fire-rescue skills were applied correctly with respect to safety and task-related job performance. Male and female pairs completed the scenarios while respiratory gas exchange data were collected with Sensormedics VmaxST portable metabolic measurement systems. On average, the males completed the scenarios faster than the females however the performance of the females was rated as satisfactory by the SME. When adjustments were made for body size, there were no gender differences in respiratory gas exchange variables, HR, perceived exertion or blood lactate. The results of this study clearly demonstrate that the aerobic demands of firefighting work are the same for males and females.

74 THE OXYGEN COST OF THE CF/DND FF TEST AND GRADED TREADMILL EXERCISE IN MALES AND FEMALES Introduction The CF/DND FF Test (Deakin et al. 1996) was developed as a simulated firefighting work circuit to evaluate the physical readiness of firefighters. The 8- minute cut-score for acceptable performance on the CF/DND Fire Fit Test was partly based upon the circuit completion time corresponding to VO 2max values cited in the scientific literature from research that only studied men. The actual aerobic demands of the FF Test were not studied in the original research, however it is important to recognize that the instrumentation necessary to make these measurements was in the developmental stages when the original research was done. That is, the instrumentation was not readily available or perhaps more importantly, not particularly precise. Improvements in this technology during the last decade have made it possible to now make ambulatory measurements of respiratory gas exchange parameters. Furthermore, it is important to examine whether or not the aerobic demands of simulated firefighting work are the same for males and females. The purpose of this phase of the research project was to directly measure the oxygen cost of a standardized simulation of firefighting work (CF/DND FF Test or the circuit ) and a standardized treadmill protocol. Efforts were made to adequately account for the effects of body mass, fire protective equipment (FPE), and the self-contained breathing apparatus (SCBA) on work capacity Methods Subjects Thirty males and 23 females volunteered as subjects in this investigation. All participants were healthy and reported regular physical activity patterns at the time of enrollment. Participants responded to word-of-mouth and poster

75 63 advertisements posted in local fitness centers. The physical characteristics of the subjects at entry are reported in Table Prior to participation, each subject provided written informed consent. The study had been previously approved by the appropriate institutional Research Ethics Board. Experimental Design All participants completed the following sequence of events: Orientation to firefighting equipment, clothing ensemble, the SCBA and the circuit; Several (usually 3-5 depending on previous experience) practice sessions of the circuit in firefighting ensemble while breathing with the SCBA; paced circuit with oxygen consumption involved completing the circuit approximately 2 min slower than the subject s best time in practice; hard circuit with oxygen consumption performed as quickly as possible, while abiding by all regulations of the test protocol; hard circuit with SCBA performed as quickly as possible, while abiding by all regulations of the test protocol; and, graded exercise test (GXT) to determine maximal oxygen consumption (VO 2max ) in firefighting ensemble (FPE). The three experimental tests on the circuit (listed above as paced and hard ) and the GXT were performed in random order. All tests were carried out indoors in a thermo-neutral environment (21-24 ºC). Firefighting Protective Equipment (FPE) During each of the exercise tests, subjects dressed in National Fire Protection Association (NFPA) Standard 1500 (NFPA, 1997) duty coat (System 300 CGSB; SafeCo Mfg. Inc., Scarborough, ON), firefighting pants (System 300 CGSB; SafeCo Mfg. Inc., Scarborough, ON), helmet (Model 911; SafeCo Mfg. Inc., Scarborough, ON), anti-flash hood (PGI Inc., Green Lake, WI), leather work gloves, and rubber firefighting boots (Black Diamond; Kaufman Footware,

76 64 Kitchener, ON), except during the GXT where running shoes were worn in the interest of comfort and safety. Subjects used a Scott 4.5 harness and a full 60- min Scott Air-Pac fibre composite air cylinder (Scott Aviation, Monroe, NC). Simulated Firefighter Work Circuit The protocol used to simulate firefighting work was the CF/DND FF Test as described by Deakin et al., (1996). This test was set up indoors on a concrete floor where all the course dimensions and equipment (type and weight) were completely accurate. The CF/DND FF Test consists of the following steps: Hose carry: The subject carried one section (15.24 m) of rolled 65 mm rubber jacketed hose (Red Chief; Angus Fire, Thame, Oxfordshire, UK) weighing 16.5 kg in one hand a distance of m, then returned the same distance carrying the hose in the other hand. The subject set down the rolled hose and walked m to the next event. Ladder carry and raise: The subject lifted and carried a 3.6 m aluminum roof ladder (13.6 kg) a distance of m and raised it against a brick wall. The subject then walked m to the next event. Hose drag: The subject gripped a hose nozzle (Pistol Grip; Elkhart Brass Mfg. Co. Inc., Elkhart, IN) over the shoulder and dragged two charged lengths (30.48 m in total) of 38 mm hose (Red Chief; Angus Fire, Thame, Oxfordshire, UK) a distance of m. The force required to move the hose bundle was approximately 260 N. The subject then walked m to the next event. Ladder climb 1: using a 7.2 m ladder (Duo-safety Ladder Corp., Oshkosh, WIS), the subject climbed 10 rungs (3.45 m) up and down, 3 times. The subject then walked m to the next event. Rope pull: while standing in a stationary position, the subject pulled a 16 mm nylon rope (static) attached to a bundle of hose (one m length of 100 mm hose and one m length of 65 mm; Red Chief; Angus Fire, Thame, Oxfordshire, UK) m using a hand-over-hand

77 65 movement. The subject then walked m and repeated the pull. The force required to move the hose bundle was approximately 200 N. The subject then walked m to the next event. Forcible entry: using a 4.5 kg steel-head sledge hammer (DF0832C; Garant, Saint-Francois, QUE) the subject hammered a 71 cm diameter rubber tire filled with sandbags (total weight kg) a distance of 30.5 cm across a 76.2 cm high wooden picnic table. The tabletop was reinforced with 19 mm (¾ ) good-one-side plywood. The subject then walked m to the next event. Victim rescue: The subject dragged a 68.2 kg mannequin (Rescue Randy 1434; Simulaids Inc., Woodstock, NY) a total distance of m (15.24 one way, around a pylon and then back m). The subject then walked m to the next event. Ladder climb 2: using a 7.2 m ladder (Duo-safety Ladder Corp., Oshkosh, WIS), the subject climbed 10 rungs (3.45 m) up and down, 2 times. The subject then walked m to the next event. Ladder lower and carry: The subject lowered and carried (15.24 m) a 3.6 m aluminum roof ladder (13.6 kg). The subject then walked m to the next event. Spreader tool carry: The subject picked up and carried a 36.4 kg spreader tool (Hurst 32B; Hale Products Inc., Conshohocken, PA) m and then returned m. During the circuit, the following time measurements were recorded: the elapsed time for each individual event; the time to move from one event to the next; and, the total time to complete the entire circuit. In addition, each subject provided a rating of perceived exertion (RPE: Borg, 1982) corresponding to each individual component of the FF Test. The 15-point scale was shown to the subject immediately after each event was completed and a recorder noted the response. The subject was familiarized with the scale during the orientation and practice

78 66 sessions so that this information could be obtained quickly without causing any delay during the test. Graded Exercise Test (GXT) Maximal oxygen consumption was determined while walking on a motor driven treadmill (Model # ; Standard Industries, Fargo, ND). The GXT was performed in the following four stages: warm-up 5-minutes at a constant speed (93.8 m. min -1 ) with a progression from 0-6% grade; 8-minutes at 10% grade and 93.8 m. min -1 ; grade was increased 1% every minute until 15%, then speed was increased by 13.4 m. min -1 each minute until volitional exhaustion; 5-minutes of cool-down at 0% grade and 53.6 m. min -1. The highest 1-minute VO 2 reading was accepted as VO 2max if a plateau in oxygen consumption was observed despite an increase in work rate or alternately, if the subject was too fatigued to continue exercise. Physiological Measurements Continuous (breath-by-breath) gas exchange measurements (e.g., VO 2, VCO 2, V E, and RER) during the GXT, paced and hard circuits were made with a VmaxST (SensorMedics, Yorba Linda, CA) portable metabolic measurement system (MMC). The software version was Meta Soft (Cortex Biophysik, Leipzig, Germany). Each subject wore a specially designed Hans Rudolph 8930 series full-face mask with an Ultimate Seal TM gel (P/N /201236) and head cap assembly (P/N ). The gas analyzers were calibrated immediately prior to each test using gases of known concentration, and calibration was verified immediately following the test. The Triple-V flow volume transducer was calibrated using a 10-stroke calibration of a 3.00-liter Hans Rudolph 5530 series syringe. In all conditions, subjects wore a Polar telemetry system (Vantage NV; Polar USA Inc., CT), which continuously measured and recorded heart rate at 5-

79 67 second intervals during exercise. A Polar Advantage computer interface operating with Polar HR Analysis Software Version (Polar Electro, OY, Finland) was used to transfer the heart rate data to a computer for analysis. The rating of perceived exertion was recorded at the end of each minute of exercise. Statistical Analysis Descriptive statistics, paired and unpaired t-tests, Pearson Product-Moment correlation, ANOVA, and linear regression were used for the statistical evaluation of the results. All statistics were performed using StatView for Windows version (SAS Institute Inc., Carry, NC). A probability value 0.05 was considered significant Results The physical characteristics of the male and female subjects from this project are displayed in Table The female subjects were shorter and lighter (p<0.05) than the male subjects. The VO 2max for the female subjects was lower (p<0.05) than the male subjects. Selected physiological responses during the CF/DND FF Test, from two of the experimental conditions, are displayed in Table There were no significant differences in these variables when the subjects completed the circuit breathing from the SCBA or when the VmaxST portable metabolic measurement system was used (Figure 2.4-1). It was important to show that the VmaxST neither advantaged nor disadvantaged the subjects with respect to performance time, perceived exertion, or heart rate compared to doing the same work with the SCBA. Average oxygen consumption for males and females during the CF/DND FF Test is reported in Figure This figure contains 60 data points for males and 45 data points for females (one hard and paced result for each subject). The

80 68 regression equation (all male and female data combined) to predict the oxygen cost (Y) of completing the circuit at a specific time (X) from this data is as follows: Y = ; r 2 = 0.529; r = ; SEE 3.68 When X = 8 minutes, the predicted oxygen cost of completing the simulation of firefighting work is 34.7 ml. kg -1. min -1. Average oxygen consumption for males and females completing the circuit between 7.5 and 8.5 minutes is displayed in Figure This figure contains 16 data points for males and 15 data points for females. The resulting regression equations for males and females and for both genders combined predict the oxygen cost (Y) of completing the circuit at a specific time (X). When X = 8 minutes, the predicted oxygen cost of doing this work was calculated to be: Combined (males and females): 34.5 ml. kg -1. min -1 Males only: 34.7 ml. kg -1. min -1 Females only: 34.3 ml. kg -1. min -1 Subsequent coincidence testing revealed no significant differences between the regression lines. Figure displays the mean average VO 2 for all males and all females completing the FF Test between 7.5 and 8.5 minutes. The mean completion time for both males and females was 7.9 ± 0.3 min. The average VO 2 values were the same when gender group means were compared. Furthermore, the mean values shown in Figure are in close agreement with the predicted values for 8- minute completion times obtained from the regression equations developed from the data shown in Figures and

81 69 Representative plots of heart rate and VO 2 for one male subject and one female subject are shown in Figures and These plots are included to show the normal pattern of HR and VO 2 responses to this simulation of firefighting work. These particular subjects were selected because: their response patterns were highly typical; both subjects completed their hard trials in approximately seven minutes; and, their stature, mass, and VO 2max were similar. Figure displays the actual VO 2 response averaged for each minute of the FF Test. Figure displays the HR and VO 2 expressed as a fraction of the individual maximal values and averaged for each minute of the FF Test. Table represents selected responses during the eight-minute constant work phase of the GXT test. The oxygen cost of treadmill exercise, expressed relative to body weight, was very similar to the values associated with completion of the FF Test at the 8-minute standard (Figures and 2.4-4). Furthermore, there was no significant difference between males and females. Figure illustrates the predictive ability of performance time on the CF/DND FF Test from maximum oxygen consumption (VO 2max ) during treadmill exercise. There is a significant relationship between these two variables. The regression equation suggests that in order to complete the FF Test in 8 minutes, most individuals would require a VO 2max of 41.2 ml. kg -1. min -1. When the VO 2max data from the treadmill test are compared with the peak 1-min VO 2 during the hard VmaxST trial, (Figure 2.4-8) there is a strong relationship (r = 0.891).

82 70 Table Mean (±SD) physical characteristics of male (n=30) and female (n=23) participants Variable Male Female Age 29.0 ± ± 5.6 Height (cm) ± ± 6.4* Body mass (kg) 84.6 ± ± 7.6* Body mass with FPE (kg) ± ± 7.5* VO 2max (L. min -1 ) 4.17 ± ± 0.43* VO 2max (ml. kg -1. min -1 ) 49.2 ± ± 4.8* p< 0.05

83 71 Table Mean (±SD) values for selected physiological measurements from the experimental trials of the CF/DND FF Test circuit with the portable metabolic system (VmaxST) and the SCBA for male (n=30) and female (n=23) subjects. Male Male Female Female Variable VmaxST SCBA VmaxST SCBA Total time (s) Work time (s) Relief time (s) 365 ± ± ± ± ± ± ± ± ± ± ± ± 24 Average RPE 14 ± 2 14 ± 2 15 ± 2 14 ± 2 Peak RPE 18 ± 2 18 ± 2 19 ± 2 18 ± 2 Average HR (beats. min -1 ) Peak HR (beats. min -1 ) 169 ± ± ± ± ± ± ± ± 11* * significantly different between VmaxST and SCBA trials within gender groups

84 VmaxST Trial (min) SCBA Trial (min) Figure Relationship (r= 0.939) between completion times for the CF/DND FF Test during the SCBA and VmaxST conditions for males (circle: 29 data points) and females (square: 21 data points). Line of identity is shown..

85 VO 2 (ml. kg -1. min -1 ) Time (min) Figure Average oxygen consumption for males (circle: 60 data points) and females (square: 45 data points) during the CF/DND FF Test. Combined regression equation Y = X; r 2 = 0.529; r = ; SEE 3.68 (34.7 ml. kg -1. min -1 when X = 8 min) Male regression equation Y = X; r 2 = 0.489; r = ; SEE 3.71 (35.6 ml. kg -1. min -1 when X = 8 min) Female regression equation Y = X; r 2 = 0.310; r = ; SEE 3.23 (33.5 ml. kg -1. min -1 when X = 8 min)

86 VO 2 (ml. kg -1. min -1 ) Time (min) Figure Scatterplot of average oxygen consumption for males (circle: 16 data points: dashed line) and females (square: 15 data points: dotted line) completing the CF/DND FF Test between 7.5 and 8.5 minutes. Combined equation: Y = X; r 2 = 0.009; r = 0.096; SEE 3.45 (34.5 ml. kg -1. min -1 when X = 8 min) Males: Y = X; r 2 = 0.021; r = 0.144; SEE 3.55 (34.7 ml. kg -1. min -1 when X = 8 min) Females: Y = X; r 2 = 0.002; r = 0.045; SEE 3.60 (34.3 ml. kg -1. min -1 when X = 8 min)

87 75 60 min -1 ) Average VO 2 (ml. kg Male Female Figure Mean ± SD average oxygen consumption for males (16 data points) and females (15 data points) completing the CF/DND FF Test between 7.5 and 8.5 minutes. Male - 16 data points, 34.6 ± 3.3 ml. kg -1. min -1 Female - 15 data points, 34.3 ± 3.6 ml. kg -1. min -1

88 VO 2 (ml. kg -1. min -1 ) Time (min) Figure Oxygen consumption record for one sample male (circle) and one sample female (square) subject completing the CF/DND FF Test in approximately 7 minutes.

89 Percentage Maximum Male VO 2 Male HR Female VO 2 Female HR Time (min) Figure Oxygen consumption (VO 2 ) and heart rate (HR) expressed as a fraction of maximal values for a sample male and a sample female subject completing the CF/DND FF Test in approximately 7 minutes.

90 78 Table Selected responses during the 8-min constant work phase of the GXT for males (n=30) and females (n=23). Values are mean ± SD Variable Males Treadmill Exercise Females Treadmill Exercise Heart Rate (beats. min -1 ) 158 ± ± 13 + % HR max 83.3 ± ± VO 2 (L. min -1 ) VO 2 (ml. kg -1. min -1 ) 2.88 ± ± ± ± 2.8 % VO 2max 70.2 ± ± RPE 13 ± 2 13 ± 2 + = significant (p < 0.05) difference between males and females.

91 VO 2max (ml. kg -1. min -1 ) Time (min) Figure Regression analysis of maximal oxygen consumption, from the treadmill test for males (circle: 29 data points) and females (square: 22 data points) and performance time from the Hard VmaxST trial of the CF/DND FF Test. Combined regression equation Y = X; r 2 = 0.328; r = ; SEE 5.9 Predicted VO 2max is 41.2 ml. kg -1. min -1 when X = 8 min

92 ) VO 2 max (ml. kg -1. min Peak 1-min VO 2 (ml. kg -1. min -1 ) Figure Regression analysis (solid line; Line of identity - dashed line) of maximal oxygen consumption, for males (circle: 29 data points) and females (square: 22 data points), and peak 1-min oxygen consumption during the CF/DND FF Test.

93 81 50 Avg VO 2 (ml. kg -1. min -1 ) Time (min) Figure Regression analysis of average VO 2 and work time from research where gas exchange has been measured during simulated fire-rescue work. From left to right: Gledhill and Jamnik 1992a; Bilzon et al, 2001; O Connell et al, 1986; Sothmann et al, 1990; Petersen and Dreger 2004 (also see section 2.3 of this report). Regression equation to predict average VO 2 for a specific work time: Y = X; r 2 = 0.936; r = 0.968; SEE 1.70 (32.4 ml. kg -1. min -1 when X = 8 min)

94 Discussion The main purpose of this project was to document the oxygen cost of the CF/DND FF Test. Of particular interest was the aerobic demand associated with completing this simulation of firefighting work at the 8-minute standard. A secondary purpose was to investigate whether the oxygen cost associated with completing the work at the minimum standard of 8-minutes was the same for males and females. The third purpose was to design a treadmill exercise protocol to replicate the aerobic demands of the CF/DND FF Test at the 8-minute standard. Two important methodological points should be made before the results are discussed. First, in order to get the best possible data during exercise in simulated firefighting work, it would be ideal to interface the portable metabolic measurement system with the SCBA. While it is possible to combine the two systems, extensive pilot investigations in our laboratory have failed to produce gas exchange data that meet acceptable standards for reliability and precision. While others have attempted to do so, our experience suggests that the resulting data may be suspect. Some of the technical difficulties can be explained, however a detailed explanation is not germane to this report. The reader should note that great care was taken to ensure that the gas exchange data presented herein could be interpreted with confidence. From a methodological perspective, it is important to bear in mind that there were no substantial or systematic differences between performance, heart rate, or perceived exertion when our subjects completed the CF/DND FF Test breathing through the portable metabolic measurement system (VmaxST) or the SCBA. The reader should remember that the data reported in Table and Figure are from the two experimental trials in which the subjects were meant to complete the FF Test circuit as quickly as possible (or, in the terminology of this experiment, the hard circuit). Our research shows that when the FF Test is completed as fast as possible, the peak VO 2, VE, HR, and RPE reach, on

95 83 average, approximately 95% of the peak values observed during graded exercise testing. These observations indicate that the typical subject must provide a nearmaximal effort to complete the circuit under the hard condition. The data reported in Table and Figure show consistency whether the subject is breathing through the VmaxST system or the SCBA. The importance of this step in the research methodology cannot be overstated. In the absence of any substantive differences between the two performance trials on the circuit under the most demanding conditions, it can safely be concluded that the VmaxST system neither advantaged nor disadvantaged the subjects compared to the SCBA. Since the SCBA represents standard practice for the correct administration of this test, the comparisons shown in Table were absolutely essential in order to accept that the respiratory gas exchange measurements were made under similar physiological conditions. The consistency of the physiological and psycho-physical responses in Table permits the confident assumption that respiratory gas exchange variables obtained during the experimental trials with the VmaxST should be completely consistent with the physiological responses to the SCBA condition. A second methodological consideration was the unique approach to measurement of VO 2max during graded treadmill exercise. In this study, the maximal oxygen consumption was measured while the subjects were dressed in firefighting clothing and carrying the SCBA. To date, other researchers have compared the physiological responses to firefighting work (subjects dressed in firefighting gear) to exercise test results obtained when the subject is dressed in normal exercise clothing (e.g., shorts and T-shirt). The full extent of the impairment in maximal work capacity and VO 2max caused by the firefighting protective ensemble is not completely understood, however, there is no doubt that physiological responses are significantly altered.

96 84 Previous research has documented impairments associated with protective clothing (Teitlebaum and Goldman, 1972; Duggan et al., 1988), the components of the SCBA (Eves et al., 2000; Eves et al., 2005), and the SCBA combined with firefighting clothing (Louhevaara et al., 1985, 1986a, 1995; Eves et al., 2002a, 2002b; 2003a, 2003b; Dreger et al., in press). Consequently, any comparisons of physiological responses during simulated firefighting work and treadmill exercise will be more valid if the clothing ensemble is matched between test conditions. The data shown in Figure reveal a moderately strong relationship (r = ) between the average rate of oxygen uptake and performance time on the CF/DND FF Test. The FF Test involves a standardized, absolute amount of work. The relationship, shown in Figure 2.4-2, demonstrates that the greater the average rate of oxygen consumption, the faster the standardized amount of work can be completed, and underscores the importance of aerobic fitness to performance in this work simulation. The regression equation from this relationship predicts that an average rate of 34.7 ml. kg -1. min -1 would be associated with a completion time of 8-minutes. As previously noted in the introduction to this report, the average VO 2 reported in the literature for fire suppression work appears to be approximately 32 ml. kg -1. min -1. The values contributing to this average were derived from a wide variety of research designs where the range was ml. kg -1. min -1. The requirement to sustain approximately 34 ml. kg -1. min -1 during 8 minutes of simulated fire-rescue work in the CF/DND FF Test seems reasonable in light of what others have reported to be representative of the demands of the work. This is especially true if one bears in mind the previously mentioned inverse relationship between exercise intensity and duration. Many of the work samples that contributed to the average VO 2 requirement of 34 ml. kg -1. min -1 found in the literature were of longer duration than 8 minutes.

97 85 One of the potential limitations of the relationship shown in Figure is that there are non-physiological limits imposed on the time range shown on the X- axis. According to the protocol specified by the Department of National Defence, the FF Test must be completed in accordance with a set of rules that places considerable emphasis on safety. For example, running between events is strictly prohibited. As well, firefighters must step on every rung during the ladder climb. For good reason, these safety rules artificially constrain the maximal work-rate. At the lower end of the time range, a completion time of greater than approximately minutes is typically associated with insufficient muscular strength and endurance to accomplish some of the individual tasks (e.g., the victim drag). In our current subject pool, only one female subject was capable of finishing the circuit in less than six minutes. Therefore, the ability to make gender comparisons throughout the entire performance range is constrained. However, in spite of these limitations, Figure shows a strong and consistent relationship between average oxygen consumption and performance. The regression equation shown in Figure suggests that in order to meet the 8- min standard, the average person must sustain a VO 2 of approximately 35 ml. kg - 1. min -1. This rate of oxygen consumption was equivalent to approximately 71% of VO 2max for the males and 83% of VO 2max for the females in this study. Therefore, while the females in this study were capable of meeting the standard, the work was done at a significantly higher relative intensity. Figure displays the average oxygen consumption for males and females completing the CF/DND FF Test between 7.5 and 8.5 minutes, or plus and minus 0.5 minute from the 8-minute BFOR standard, respectively. Within this narrow time range bracketing the actual standard, there are almost equal numbers of data points (16 male; 15 female), which allows for a fair comparison of the average oxygen cost with respect to gender. Based on the gender-specific regression equations generated from the relationships between average VO 2 and

98 86 test time, the average oxygen cost of completing the test at the 8-minute standard was 34.7 and 34.3 ml. kg -1. min -1 for males and females, respectively. Figure shows the result of averaging the data points for males and females completing the test between 7.5 and 8.5 minutes (shown in Figure 2.4-3). This process revealed that the mean average VO 2 was 34.6 and 34.3 ml. kg -1. min -1 for males and females, respectively. Within this one-minute range in performance time, the average VO 2 was the same for males and females. As noted above, while the actual VO 2 was the same, the oxygen cost expressed as a fraction of VO 2max was different. That is, 34.3 ml. kg -1. min -1 represented 82% of VO 2max for the females, while 34.7 ml. kg -1. min -1 represented 70% of VO 2max for the males. The mean test completion time for both the male and female subjects in this subset was 7.9 ± 0.3 minutes. Support for the validity of the aerobic demands of the FF Test may be found in the research reported by Sothmann et al. (1990). In that study, male firefighters completed a circuit consisting of seven tasks in a hot (130ºC), smoke-filled (nontoxic) environment. The firefighters were asked to complete the tasks at a selfselected work-rate that was consistent with their experience in real emergencies. A device to measure oxygen consumption was interfaced with the SCBA and VO 2 data were recorded three times during the circuit. These values were averaged to determine a VO 2 value representative of the work protocol. The average duration of the protocol was approximately nine minutes (with a range between 5:30 and 13:54) and the mean VO 2 value was 30.5 ml. kg -1. min -1 (range of 23.5 and 49.3 ml. kg -1. min -1 ). Using the data reported in this article, it is possible to generate a regression equation to predict the average VO 2 for any completion time. This equation predicts that an average metabolic rate equivalent to approximately 32 ml. kg -1. min -1 would be required to complete the work simulation described by Sothmann et al., (1990) in 8 minutes. While direct comparisons between the results of Sothmann et al., (1990) and the present study are not possible, it is of interest to note the similarity of the VO 2 responses.

99 87 To emphasize this point, the results of the best studies that have actually measured average oxygen consumption during simulated firefighting can be pooled to predict the average oxygen consumption that would be expected for the CF/DND FF Test. The data shown in Figure use the results of O Connell et al., (1986); Sothmann et al., (1990), Gledhill and Jamnik (1992a), Bilzon et al., (2001), and our fire-rescue study (Section 2.3 of this report). In each case, the average VO 2 is plotted against the average work time reported in these studies. The line shown in Figure describes the relationship between the two variables as calculated by linear regression. The relationship is highly significant as the r 2 value for the regression analysis is This suggests that based on the input data from the five studies, there is very strong relationship between average oxygen uptake and work time during simulated firefighting. Using the equation generated from the regression analysis, the predicted VO 2 for eight minutes of simulated firefighting work, is approximately 32 ml. kg -1. min -1. This prediction is very similar to the measured value for completion of the CF/DND FF Test at the 8-minute standard as described earlier in this report. The scatter above and below the regression lines in Figures and reveals that while average VO 2 may be significantly related to test performance time, it is also clear that other factors contribute to performance. The nature of the work being done in many of the circuit tasks (e.g., moving heavy equipment) is suggestive of the importance of strength and anaerobic power in addition to aerobic fitness. Review of the physical characteristics of the subject pool suggests that body mass and stature may influence how individuals accomplish this work. The relative importance of strength to the performance of simulated fire suppression tasks has been reported previously (Davis et al., 1982; Williford et al., 1999). This should not be overlooked when evaluating the factors contributing to performance on any firefighting work simulation. While the focus of the current project was on the aerobic demands of this work simulation, it

100 88 would be a gross oversimplification to suggest that aerobic fitness alone determines test performance time. The sample heart rate and oxygen consumption records illustrated in Figures and indicate the typical pattern of physiological response during the CF/DND FF Test. It can be observed that shortly after the onset of the test, both HR and VO 2 rise rapidly to a high level that is sustained for the balance of the exercise time. This profile is shown to underscore the importance of the average VO 2 and HR responses in this type of test. The HR profiles shown for the two subjects in Figure are consistent with the HR profiles previously reported for the CF/DND FF Test (Deakin et al., 1996) and other simulated firefighting work (Petersen et al., 2000). One of the goals of the main research project was to develop a treadmill protocol to replicate the aerobic demands of the CF/DND FF Test. Pilot research revealed that treadmill walking at 93.9 m. min -1 (3.5 mph) and 10% grade while dressed in FPE was a good simulation of the oxygen cost of completing the CF/DND FF Test at the 8-minute standard. The data in Table show that the average VO 2 required to complete the circuit at the 8-minute standard can be matched with the average VO 2 during steady-rate treadmill exercise. The treadmill protocol replicates the aerobic demands of the FF Test in a more controlled mode of exercise and provides an excellent opportunity to compare male and female responses to the same external workload. Table reveals that there were no gender differences in the oxygen cost of the treadmill exercise. The males and females followed an identical treadmill protocol for warm-up and the steady-rate exercise phase. Each subject was dressed in properly fitting FPE. The VO 2 response to this highly controlled bout of work was identical. This observation is consistent with the VO 2 responses during the circuit and is supported by the work of Bilzon et al. (2001) who found no gender differences in the oxygen cost of simulated

101 89 firefighting work. The ability to replicate the aerobic demands of the FF Test with a standardized exercise protocol on a treadmill will be very useful in screening applicants for the CF/DND fire service. Furthermore, it allows both applicants and incumbents the opportunity to train for the aerobic demands of the FF Test in a safe and simple manner. The traditional approach to evaluating aerobic fitness in firefighters has required meeting a standard for VO 2max. The results from the current study indicate that an alternative approach would be to require the applicant to maintain a rate of oxygen consumption (in this case, approximately 34 ml. kg -1. min -1 ) consistent with the requirements of firefighting work. The data presented here has documented the aerobic demands associated with meeting the minimum standard (8-min) for CF and DND firefighters. This aerobic demand can be replicated on a treadmill. It is suggested that a standardized bout of treadmill exercise (dressed in FPE) could satisfy the aerobic aspect of a physical fitness standard. Numerous authorities have described the physically demanding nature of firefighting work. For example: firefighting is one of the most hazardous and physically demanding jobs in the public sector. (Brownlie et al., 1985) Firefighting can be a physically demanding occupation. (Davis et al., 2002) Ample evidence exists supporting the notion that structural firefighting is a physically demanding occupation. (Davis et al., 1982) Firefighting is widely acknowledged to be one of the most physically demanding and hazardous of all civilian occupations (Gledhill and Jamnik, 1992a) Firefighting involves performing strenuous physical work (Smith and Petruzzello, 1998) One of the most complete descriptions of the physical demands of firefighting may be found in the most recent (2003) edition of the National Fire Protection Association Standard 1582 entitled, Comprehensive Occupational Medical

102 90 Program for Fire Departments. This standard articulates 13 Essential Job Tasks with the intent of informing examining physicians of the nature of the physical strain associated with firefighting. On several occasions in the document, it is noted that firefighters require a maximal exercise tolerance of 42 ml. kg -1. min -1 (or 12 metabolic equivalents). This level is consistent with other recommendations for VO 2max found in the literature (Gledhill and Jamnik, 1992b; Bilzon et al., 2001). While it was not the purpose of this investigation to address the VO 2max requirements for firefighters, some interesting observations are possible. The data in Figure show the relationship between VO 2max recorded during the treadmill test and performance time on the FF Test. The relationship (r = ) is statistically significant, but apparently weaker than the relationship between average VO 2 and performance time shown in Figure (r = ). More research would be required to specifically address this question however, these data suggest that the ability to sustain a level of energy expenditure consistent with approximately 34 ml. kg -1. min -1 may be a stronger predictor of completion time on the FF Test than is VO 2max. The data shown in Figure illustrate a very strong relationship (r = 0.865) between the peak VO 2 during the FF Test and the VO 2max obtained during graded exercise testing. Figures and present data to show that VO 2max may be a significant factor in successful completion of the FF Test. The regression equation shown in Figure predicts that, on average, a completion time of 8 minutes is consistent with a VO 2max of 41 ml. kg -1. min -1. This value is in accordance with the recommendations of NFPA 1582 (42 ml. kg -1. min -1 ), as well as the research of Bilzon et al., (2001) on naval firefighters (41 ml. kg -1. min -1 ) and Gledhill and Jamnik (1992b) on civilian structural firefighters (45 ml. kg -1. min -1 ). Logically, firefighters must have a level of physical fitness consistent with the capacity to perform strenuous physical labor in a safe and effective manner. The NFPA Standard for the Organization and Deployment of Fire Suppression

103 91 Operations, Emergency Medical Operations and Special Operations to the Public by Career Fire Departments states an early aggressive and offensive primary interior attack on a working fire is usually the most effective strategy to reduce loss of lives and property damage. (NFPA 1710: 2001, page 15). This is a powerful endorsement of the time-sensitive nature of fire suppression operations. Structural firefighting has been classified as very heavy manual labor according to the DOT Occupational Code (US Dept of Labor, 1993). Other authorities have offered suggestions on the energy requirements of very heavy manual labor. For example, Astrand and Rodahl (1986, p. 491) have categorized the oxygen cost of this classification of work as ranging from kcal. min -1 or L. min -1. Considering the mean subject weight in the current study (76.4 kg), these values convert to estimated VO 2 values of approximately ml. kg -1. min -1. McArdle et al. (2001, p. 195) independently suggested that the energy cost of very heavy physical activity ranges from approximately ml. kg -1. min -1 for men and from approximately ml. kg -1. min -1 for women. These estimates are in close agreement with the suggestions of Astrand and Rodahl (1986) for very heavy work and also the VO 2 values found in the literature for firefighting (approximately 34 ml. kg -1. min -1 ). Moreover, there is close agreement with the values reported in the present study for men and women completing the CF/DND FF Test at the 8-minute standard (approximately 34 ml. kg -1. min -1 ) Summary In summary, the results of the present study have revealed a significant relationship between the average oxygen consumption and test completion time for the CF/DND FF Test, underscoring the importance of aerobic fitness to safe and effective completion of firefighting work. The average VO 2 consistent with meeting the 8-minute standard is the same for males and females, however the VO 2 required for achieving the standard represented a higher fraction of the VO 2max for females than males.

104 92 This research has documented the oxygen cost of the CF/DND FF Test. The results show that the average VO 2 required to meet the 8-minute standard falls within the range of other reports of the aerobic demands of firefighting work. Notwithstanding the importance of other physical attributes such as strength and power, individuals with average to good aerobic fitness should be capable of meeting this standard. The results of this study show that the aerobic demands of the standard are the same for males and females. It is suggested that applicants to the CF/DND fire service be evaluated for aerobic fitness using the treadmill test described above. Applicants must be dressed in properly fitting FPE and the test protocol must be standardized. Completion of the standardized warm-up (5 min) and the constant work phase (8 min) would demonstrate the ability to maintain the VO 2 that is consistent with the aerobic demands of the FF Test when completed at the 8-minute standard. Thus, completion of 18 minutes of standardized walking exercise (5 minutes of standardized warm-up, 8 minutes of the constant test workload, and 5 minutes of standardized cool-down) would constitute a pass on the aerobic component of the fitness evaluation. The second phase of this test was designed to measure VO 2max and if so desired, applicants who meet the passing criteria could be ranked on the maximal oxygen consumption. There is reasonable evidence that persons with greater aerobic fitness have a greater probability of completing the FF Test in less time.

105 DESCRIPTION OF THE APPLICANT TREADMILL TEST AND INTERPREPTATION OF TEST RESULTS Introduction Aerobic fitness is indisputably recognized as an important component of physical fitness in firefighting. The design of the CF/DND FF Test requires firefighters to sustain a relatively high rate of oxygen consumption (VO 2 ) for approximately 8 minutes. While the actual VO 2 is not evaluated as an outcome measure of the FF Test, the assumptions made by the original research team (Deakin et al., 1996) have been borne out and validated by some of the work described in this report. The overall design of the Applicant Test required a separate evaluation of aerobic fitness, and the results of the treadmill protocol allow both screening and selection of applicants based on aerobic fitness. In this report, the term screening has been used consistently in reference to the process of determining whether an individual meets the minimum fitness requirement. The term selection has been used consistently with reference to a process for ranking individuals based on their fitness scores Description of the Treadmill Test Protocol The development of the aerobic fitness test was based on common principles of exercise testing, along with consideration of the organizational needs of the CF/DND fire service. The test protocol may be adapted to suit the implementation plan that is ultimately developed for applicant testing. Two options exist including: completing the screening and selection portions of the test while measuring oxygen consumption (in this case, VO 2max is measured); or, only completing the screening portion of the protocol (in this case, gas exchange measurement is not required). The first option provides the most precise data,

106 94 and consequently, is strongly recommended. The test description below is for this option. Mode The treadmill was chosen for several reasons. First, many aspects of firefighting involve walking or climbing exercise under load (Gledhill and Jamnik, 1992a). Workload on a treadmill can be varied through adjustment of speed and grade, allowing for a reasonable simulation of these aspects of firefighting. Second, treadmills are readily calibrated so that the test protocol can be reproduced at various test sites. Third, treadmills are also widely available to the public, so prospective applicants to the CF/DND fire service can train in the same mode as they would be tested. Clothing Firefighters typically work while wearing a personal protective ensemble (PPE) that weighs approximately 22 kg. As previously discussed, this ensemble also restricts normal movement; increases thermal stress; and negatively affect both maximal and sub-maximal exercise performance. Consequently, it is important that the aerobic test should be completed while wearing a very close simulation of complete PPE. During the treadmill exercise test, participants dress in National Fire Protection Association (NFPA) Standard 1500 (NFPA, 1997) duty coat, firefighting pants, helmet, anti-flash hood, leather gloves, running shoes, and a self-contained breathing apparatus. Warm-up A common practice in exercise testing involves a warm-up phase (ACSM, 2006). Each participant performs an identical warm-up on the treadmill to allow for preparation of the body for physical exertion. The warm-up consisted of 5- minutes walking at a constant speed (93.8 m. min -1 ) with a progression from 0-6% grade (3 minutes at 0%, 1 minute at 2%, and 1 minute at 6% grade).

107 95 Screening Phase Section 2.4 of this report describes in detail the evaluation of the aerobic demands of the FF Test. It is essential that the aerobic requirement for the applicant test mirror the requirement associated with the incumbent test. Therefore, the treadmill test includes an 8-minute, constant work phase that simulates the average VO 2 of the FF Test at the 8-minute standard. As described in Section 2.4, the oxygen cost of completing the test in 8 minutes has been documented and found to be consistent with the results of other reports of the aerobic demands of fire-rescue work. It was also noted in Section 2.4 that the oxygen cost of the work was the same for the male and female subjects tested. In this project we have taken a novel approach to screening aerobic fitness in firefighters. The screening criterion is based on the ability to complete a reasonable period of exercise (8 minutes, after a standardized warm-up of 5- minutes) at an intensity that is consistent with our knowledge of energy expenditure during initial attack firefighting. This work-rate on the treadmill typically requires a rate of oxygen consumption of ml. kg -1. min -1 (see Table 2.4-3). The workload for this stage is 10% grade and 93.8 m. min -1 (3.5 mph). The speed represents a brisk walk for most individuals. The selection of speed and grade was based upon extensive pilot work. The physical characteristics of the male and female participants in this project reflect a wide range in age, stature, mass and aerobic fitness. The walking pace appears to be universally suitable. Selection Phase Upon completion of the screening stage, participants, if able, proceed immediately to the second phase of the test. The outcome of this phase is a measure of maximal oxygen consumption (VO 2max ). Attainment of VO 2max, requires that the applicant be taken to the point of maximal exertion, which is

108 96 achieved through a combination of small increases, first in grade (1% every minute until 15%) and then speed (increased by 13.4 m. min -1 or 0.5 mph each minute) until volitional exhaustion. Applicants can then be ranked on the VO 2max score. As noted previously in this report, there is general agreement that aerobic fitness is important for firefighters and that VO 2max is a readily measured gold standard. While it is difficult to find consensus among researchers regarding the minimum VO 2max that is required for firefighters, there is little doubt that higher is better. The data in Figure show that VO 2max is significantly related to performance time on the FF Test. Consequently, we suggest that applicants be ranked on VO 2max for purposes of selection. Cool-down (recovery) Upon completion of the selection stage, participants are required to perform a standardized cool-down consisting (5-minutes at 0% grade and 53.6 m. min -1 or 2.0 mph). The breathing valve for measurement of oxygen consumption is removed immediately. After the exercise test, most applicants will be very warm, and the cool-down process is facilitated by removing the helmet, flash-hood and gloves, and consuming cool water. In our extensive experience, most applicants are recovered enough to stop exercise after 5 minutes. It should be emphasized that the recovery stage is considered essential for a safe transition from maximal exertion to stopping exercise. Therefore, except in exceptional circumstances where the safety of the applicant is in question, the 5-minute recovery period is part of the test protocol that must be completed. Measurements During the test, it is essential to keep an accurate record of exercise time and the corresponding workload parameters (speed and grade) for each minute of the test. A properly calibrated metabolic measurement system must be used for analysis of oxygen consumption. A Polar heart rate monitor (or similar system)

109 97 should be used to record HR during each minute of exercise. The rating of perceived exertion (RPE: Borg, 1982) should be recorded at the end of each minute of exercise. While the RPE is not part of the test result per se, the data provide very useful insight into the fatigue state of the subject during and after the test. The highest 1-minute VO 2 reading should be accepted as VO 2max if a plateau in oxygen consumption was observed despite an increase in work rate or alternately, if the subject was too fatigued to continue exercise. The degree of fatigue can be evaluated through direct observation and the RPE responses at peak exercise. Figure Treadmill test for evaluation of aerobic fitness of firefighter applicants with gas exchange analysis equipment for measurement of oxygen consumption.

110 98 Figure Treadmill test for evaluation of aerobic fitness of firefighter applicants without gas exchange analysis equipment for measurement of oxygen consumption. The treadmill test can be completed in the configuration shown in Figure for the purpose of screening, since the measurement of oxygen consumption is not required. At this time, it is recommended that the maximal oxygen consumption (VO 2max ) be used as the aerobic fitness criterion for selection Interpretation of Treadmill Test Results Screening An applicant passes the screening portion of the test by completing the required amount of exercise consisting of: The standardized warm-up (5-minutes);

111 99 Constant work phase (8-minutes); and, The standardized cool-down (5-minutes). If an applicant is unable to complete the screening protocol, that would be grounds for failure on the aerobic test. It is essential to recognize that the test protocol must be completed safely, and hence any unsafe behavior or signs of excessive fatigue could also be grounds for terminating the test before the required time has been completed. The implementation plan should clearly specify the criteria for allowing the test to continue or alternately, be terminated. Selection Upon completion of the second phase of the treadmill test, the VO 2 values should be averaged over 60-s intervals. The highest score can then be used for selection purposes if so desired. Many authorities have acknowledged the significance of VO 2max in relation to evaluating physical fitness for firefighters. We report test results from 97 (87 male and 10 female) experienced firefighters using the same treadmill protocol as described above. The characteristics of these firefighters are reported in Table These data are similar to the characteristics reported by Gledhill and Jamnik (1992b) for 51 civilian male firefighters in Toronto. The VO 2max categories and associated points in Table were derived from the analysis of the VO 2max scores of 97 experienced firefighters who were tested according to the applicant protocol. The VO 2max data for the group of 97 firefighters were quite normally distributed as shown in Figure For the purposes of ranking applicants based on VO 2max, a 6-point scale (0-5) was established. The highest point value (5) was assigned to those scores who met or exceeded the 90 th percentile and the lowest point value (0) was assigned to scores at or below the 10 th percentile. Scores

112 100 associated with 1 to 4 points were determined based on the creation of equal segments between the 10 th and 90 th percentiles. The mean score for the firefighters fell within the range associated with 3 points. Table Selected characteristics of experienced firefighters* (n=97) Variable Mean SD Minimum Maximum Age (yr) Height (cm) Weight (kg) Fire Service (yr) VO 2max (ml. kg -1. min -1 ) *Male = 87 and Female = 10 VO 2max (ml. kg -1. min -1 ) Table VO 2max scores and associated points Points < >

113 101 Count VO 2max (ml. kg -1. min -1 ) Figure Distribution of VO 2max scores for 97 experienced firefighters

114 102 SECTION 3 DEVELOPMENT OF JOB-RELATED TESTS AND STANDARDS 3.1 Introduction Logically, any new testing program for applicants should be linked closely to the current physical fitness requirement for incumbent firefighters (Gerkin, 1995), which is the 8-minute standard for the CF/DND FF Test. This requirement was developed following a substantial research program at Queen s University and the outcomes of that process have been widely recognized for excellence in the field of occupational testing. The Department of National Defence and specifically the CFPSA, have been recognized for leadership for the manner in which the FF Test was developed and implemented. Several approaches to testing applicants were considered that would build on the strengths of the existing test procedures and standard. A fundamental principle underlying the entire research project was the recognition that the requirements developed for applicants be closely linked with the minimum level of acceptable performance on the CF/DND FF Test. One approach would be to simply utilize the incumbent firefighter test for the assessment of applicants. The advantages of this approach are that the scientific validity of the circuit has already been established and that the test protocols, equipment and procedures are available at all CF and DND fire halls. The main disadvantage is that the test procedure requires the use of full protective gear and the self-contained breathing apparatus (SCBA). It would be inappropriate to test potentially naïve applicants in this manner since the ability to perform heavy exercise under these conditions represents an acquired skill. A second disadvantage is that completion of the circuit in the best possible time requires a certain amount of pacing ability that must be acquired through practice. The results of the Women s Sub-study reported by Deakin et al (1996) clearly showed the effect of practice on circuit performance in previously naïve

115 103 subjects (Deakin et al., 1996, Figure 8, p.66). It would be inefficient to train applicants prior to testing. A third disadvantage is that the performance on the test circuit is expressed as the total time for completion. This does not allow precise determination of the ability to meet a minimum performance level in each of the representative subset of job-related tasks. It is quite conceivable that an applicant could do very poorly on some elements but compensate by doing very well on other elements. The overall completion time might not reflect a potentially unacceptable performance on some elements. Effective screening of applicants requires that they demonstrate at least minimal competencies on all of the representative tasks. A final problem is that the cut score of eight minutes was set based on testing of experienced incumbents. It would be necessary to undertake research to ensure that either the standard applies to inexperienced applicants as well, or alternately a new applicant standard would need to be developed. A second approach could involve a modification of the incumbent test circuit procedures. For example, applicants could carry the SCBA but not be required to breathe from it and could wear running shoes instead of fire fighting boots. As well, it may be argued that circuit elements such as the ladder raise involve skills that are acquired in training, consequently this task might need to be replaced by another physically demanding (but unskilled) task. These types of changes would improve the likelihood that naïve applicants could complete the circuit safely and effectively. However, the disadvantages listed above would still apply. The standard for successful completion would have to be modified extensively since in this case, several elements of the test would be different. A third alternative would involve identification of a series of discrete fitness component tests that show strong correlation with circuit performance. The data of Deakin et al., (1996) suggest the conclusion that firefighters with higher levels of aerobic fitness tend to have faster performance times on the circuit. This conclusion is supported by the results of two analyses. First, from the pilot study

116 104 data (Deakin et al., 1996, Final report, Table 7, page 40) it can be seen that circuit time was significantly correlated with VO 2max (r= -0.83). Second, the data from the main study show that the mean circuit time increases as predicted VO 2max decreases (Deakin et al., 1996, Final report, Table 11, page 50). However, these analyses also show statistically insignificant correlations between circuit performance and measures of upper and lower body power output (Deakin et al., 1996, Final report, Table 7, page 40). Finally, the relationship between VO 2max and circuit performance at best only explains about 69% of the variance in circuit performance. Our research (see Figure 2.4-7) suggests that the relationship between VO 2max and circuit performance is not nearly as strong as suggested by Deakin et al., (1996). Therefore, a high VO 2max alone is not a strong predictor of success on the circuit. If this approach were selected for assessment of applicants, extensive research would need to be undertaken to identify discrete fitness tests that account for the unexplained variance. Of equal importance would be the identification of minimal levels of performance on those discrete fitness tests that relate to the BFOR of 8 minutes on the circuit. This would be a formidable task. The approach that was chosen maintains the close links with the incumbent circuit and can be administered at selected CF bases. The plan involved developing a series of individual tests based on a selection of physically demanding aspects of the circuit. As previously described in Section 2.4, a separate laboratory test for aerobic fitness was developed. The main points of the job-related tests are described briefly below: A number of representative tasks were selected from the incumbent circuit and a series of individual job-related performance tests were created. The representative tasks are physically demanding, but do not require firefighting skills for successful completion. Applicants are encumbered with the equivalent weight of full protective gear, but are not required to breathe from the SCBA.

117 105 The tests are completed in a specified order with standardized rest periods between attempts. A minimum standard was developed for each test that corresponds to an acceptable work rate as identified by subject matter experts. A range of performance standards were developed for each test in order that applicants may be ranked on each aspect of physical fitness. 3.2 Methods Selection of Representative Tasks The physically demanding tasks that comprise the circuit were selected based on a complete job analysis (Deakin et al., 1996). These tasks represent the activities that may be reasonably expected of any single firefighter during fire suppression. Some of the tasks involve certain skills that would normally be acquired during training and job experience. On the other hand, most of the tasks require little skill for completion and as such, are well suited to testing applicants as well as incumbents. As previously noted, a number of representative tasks from the CF/DND FF Test were selected in order to develop a series of individual tests. It is important to note that these tests were based on the physical demands of the corresponding element of the circuit, both with respect to the procedures and the performance standards. The circuit elements were chosen carefully after a thorough task analysis and with the expert input from CF/DND personnel. Consequently, we believed that these elements should not be significantly changed for the applicant assessment. The general principles on which the job-related tests were developed are described below: Each task must be classified as either a common task (e.g., hose advance) or a critical task (e.g., victim rescue). The former classification represents tasks that are routinely completed during fire suppression. The latter classification represents tasks that may be

118 106 encountered on a rare or infrequent basis, however each firefighter must always be in a state of readiness for such tasks. Each task must be normally completed by a single firefighter. That is, tasks that are normally completed by a team effort were excluded. The tasks must be consistent with emergency conditions. That is, since performance is a key element in the proposed test protocol, tasks that are not completed with a sense of urgency were avoided. The distinct tests may generally be longer and harder than the corresponding element in the circuit. Any adjustment in, for example weight or distance, accounted for the fact that applicants will have the benefit of rest periods between each test and that they will not be required to breathe from the SCBA. However, the absolute weight of the loads (e.g., hose) must remain consistent with normal job demands for a single firefighter. For example, in the development of the applicant charged hose advance test, pilot work revealed the necessity to increase the load beyond the charged hose advance component of the circuit. We considered dragging a heavier hose (e.g., 65 mm) over a shorter distance (e.g., 15 m), however this task was no longer consistent with the normal expectations of a single firefighter. Therefore, further pilot studies revealed that rather than increasing the resistance imposed by the load, the desired effect could be attained by increasing the duration of exposure to the load. The representative subset of tasks from the work of Deakin et al., (1996) found in the CF/DND FF Test were: One arm hose carry Charged hose drag Forcible entry Victim drag

119 107 High volume hose pull Spreader tool carry Ladder carry Ladder raise and lower Ladder climb A number of these tasks appeared to have redundant characteristics. For example, the hose carry and the spreader tool carry both involve assessment of the strength necessary for lifting, carrying and moving equipment. It seemed reasonable to select the most demanding task that required the least skill on the assumption that if an applicant can complete that task, he/she can most likely complete the lesser demanding task as well. The ladder carry, raise and lower tasks all involve skills related to handling ladders, which are presumably learned in training or through on the job experience. These elements were eliminated from applicant testing. For the purposes of applicant testing, the following tasks were investigated: One arm hose carry Charged hose drag High volume hose pull Forcible entry Victim drag Ladder climb Spreader tool carry It was necessary to consider modification of the test procedures for these activities so that the procedures were more effective at screening applicants, however the essential physical demands remained consistent with the original tasks. This was readily assessed by having experienced firefighters provide feedback on the similarity of the two versions of each task.

120 Pilot Studies Job-related Performance Tests The first step in this research was a series of pilot studies to develop and refine the job-related test protocols. The general pattern for developing discrete jobrelated tests was as follows: Identify the task (e.g., charged hose advance) Identify the appropriate forces, distances and repetitions Identify test conditions (e.g., surface and clothing) Identify any obvious problems related to skill or safety Pilot testing with a small number of subjects to identify wrinkles in test procedures. Examine relationships between performance of the element within the circuit and performance of the task as a discrete test for this small group Use feedback from incumbents to improve protocol Standardize protocol In this pilot work, which lasted several weeks, a small group researchers and graduate students (n=6) from our laboratory worked to develop the job-related test protocols. Each of these individuals has had extensive experience with exercise testing and previous firefighting-related research projects (e.g., Petersen et al., 2000; Eves et al., 2002a, 2002b; Eves et al., 2003a, 2003b; Eves et al., 2005). These individuals also have had extensive experience with the CF/DND FF Test. Their understanding of exercise physiology, exercise in PPE and exercise testing allowed valuable insight into the development of test protocols that mirrored the physical demands of selected tasks from the CF/DND FF Test. Once the preliminary protocols were established, a small group (n=6) of CF firefighters from Edmonton Garrison volunteered to complete the tests and provide feedback. These individuals were recognized as subject matter experts with respect to the CF/DND FF Test and the fire-rescue work. The synergy

121 109 between these two groups of experts provided an excellent process for development of the job-related test battery. The main outcomes of the pilot work were: The order of the discrete tests remained consistent with the order of the related element within the circuit; Applicants should complete the test in the identical ensemble of PPE that incumbents would use during the circuit, except for the actual use of the SCBA; A number of the selected tasks were modified in the manner noted above. That is, as a general rule, rather than increasing the load (e.g., weight to be moved), the difficulty of the task was increased by altering the duration of exposure to the load (e.g., moving the same weight a greater distance) as long as the demands were consistent with normal firefighting duties of a single firefighter; Brief rest periods were incorporated between each of the discrete tests. In part, the duration of the rest periods was based on the theoretical principles (Fox and Mathews, 1981, p. 280) in exercise physiology for appropriate work:rest ratios. In part, the duration of the rest periods was also based on the practical requirement for the test subject to have reasonable recovery while avoiding excessive cool-down between tests. The consensus of the pilot subjects was that 3 minutes of rest between each test was adequate; The testing environment should be consistent with the environment normally encountered during the CF/DND FF Test. That is, indoors on a smooth concrete floor Equipment Carry/Vehicle Extrication Test The Equipment Carry/Vehicle Extrication simulation test was originally developed for the City of Edmonton fire department in At that time, the department identified the need for a test to evaluate upper-body strength and endurance specific to vehicle extrication incidents. In response to that need, the following

122 110 process led to the development of the test. First, the researchers (R. Dreger and S. Petersen) debriefed experienced crews on Rescue units in the Edmonton fire department. These crews reported that the most common physically demanding activity in vehicle extrication (VE) was removing a passenger car door. These firefighters informed the researchers that the task was frequently encountered, was physically demanding, and, was essential to a successful rescue. We then observed the crews performing what they described as a typical set-up and response at the training facility in Edmonton. Briefly, the Rescue vehicle arrived on-scene and parked approximately 23 m (~75 ) from the damaged vehicle. Firefighters moved rescue equipment and tools to a staging area approximately 8 m (~25 ) from the vehicle. Equipment such as a generator was operated at the staging area and provided power for the tools that were actually used at the vehicle (e.g., cutter and spreader tools, the jaws of life ). Once the set-up was completed, the firefighters removed the front drivers side door of the vehicle by breaking the pins at the two hinge points on the front of the door and the locking pin at the rear of the door. It was noted that the mean time that experienced firefighters required to break each pin with a spreader tool was 30 s. During this time, the firefighter was required to hold the spreader tool in various positions, mainly using the upper body muscles. The firefighters informed the researchers that the simulations that were observed were representative of the typical work done at a vehicle extrication incident. A test protocol was developed that required firefighters to move two pieces of equipment (spreader and/or cutter tools) from a start location (simulating the rescue truck) to a staging area. The firefighter then carried a spreader tool that weighed approximately 18 kg to a car-door mock-up located 8 m from the staging area. The spreader tool was held in a horizontal position in contact with a flat metal disc that represented the location of each pin on the driver s side door of a full-sized sedan. The firefighter held the tool in a standardized manner for 30 s at each pin, and was required to place it on the ground after each 30 s work period.

123 111 Finally, the firefighter carried the spreader tool all the way back to the starting point, retrieved the second tool from the staging area and returned it to the start point to end the test. Thirty Edmonton firefighters volunteered to go through the test protocol. Each was asked to complete the protocol at a self-selected pace consistent with their experience at vehicle extrication incidents. The slowest time for completion was 4 minutes (240 s). The firefighters reported that the physical demands of the simulation were very similar to the physical demands of similar tasks in vehicle extrication work. The test was adopted by the City of Edmonton as part of the screening protocol for firefighter applicants. The original test was modified slightly during our pilot studies to make it more consistent with the equipment used in the CF/DND FF Test. Briefly, the large spreader tool that is carried 30 m (100 ) in the FF Test was added in place of the smaller cutter tool Job-related Test Protocol The test protocols arising from the pilot studies are described below. Figure shows the applicant in the PPE ensemble utilized during the test. In each test, there are requirements that must be met for safety and for consistency. Running is not permitted during any test, although, the applicant is encouraged to complete the tasks as quickly as possible.

124 112 A B Figure Front (A) and side (B) views of an applicant dressed in the correct attire for the job-related tests. The applicant is wearing correctly fitting fire protective clothing including jacket, pants, rubber boots, flash-hood, helmet and leather work gloves. The applicant is carrying the self-contained breathing apparatus (in this case, a SCOTT 4.5 harness with empty one-hour cylinder). Note that the jacket collar is fully done up and the helmet visor is in the down position.

125 113 Charged hose advance starting from an erect position facing forward, the applicant bends and picks up a nozzle connected to 3 lengths of charged 38 mm hose (Red Chief, Angus Fire, Thame, Oxfordshire, UK) that are flaked behind the start line. With the nozzle and hose held securely over the preferred shoulder, the applicant advances the hose from the start line to the finish line. The applicant is encouraged to move as quickly as possible without running. The total distance the hose is advanced is 38.1 m. The applicant is given 3 minutes of rest before the next test. Figure View of the Charged Hose Advance Test. The applicant has just picked up the hose and has taken the first step towards the finish line.

126 114 Rope Pull starting from an erect position facing forward, the applicant bends and picks up a length of static 16 mm nylon rope attached to a bundle of hose (one m length of 100 mm hose and one m length of 65 mm; Red Chief, Angus Fire, Thame, Oxfordshire, UK). Keeping the feet securely in place, the applicant uses the rope to pull the bundle over the floor a distance of m. The subject then walks m and repeats the pull, walks back m and repeats the pull for the third and final time. The force required to move the hose bundle was approximately 200 N. The applicant is given 3 minutes of rest before the next test. Figure View of the Rope Pull Test. The applicant is pulling the hose bundle using the 16-mm static rope.

127 115 Forcible entry starting from an erect position, the applicant picks up a 3.64 kg plastic dead-blow sledge hammer (Stanley Tools) and uses it to move a 71 cm diameter rubber tire filled with sandbags (total weight kg) a distance of 30.5 cm across a 76.2 cm high table. The table is made of steel and is placed against a wall for stability. The table-top is good one-side,19-mm plywood. The applicant is given 3 minutes of rest before the next test. Figure View of the Forcible Entry Test.

128 116 Victim rescue starting from an erect position, the applicant lifts and drags a 68.2 kg mannequin (Rescue Randy 1434, Simulaids, Inc., Woodstock, NY) while walking backwards a total distance of m (15.24 one way, around a pylon and then back m). The applicant is given 3 minutes of rest before the next test. Figure View of the Victim Rescue Test. The applicant has the choice of lifting the rescue mannequin with the arms around the mannequin s torso or gripping a harness (as shown) while walking backwards. In this photograph, the applicant is approaching the pylon where the change of direction occurs. The applicant must drag the victim around the pylon without touching it.

129 117 Ladder climb starting from an erect position facing the ladder, the applicant climbed 10 rungs (3.45 m) up and down a 7.2 m ladder (Duo-safety Ladder Corp., Oshkosh, WI). This is repeated 5 times as quickly as possible. A repetition begins with both feet on the floor at the base of the ladder. The applicant climbs and places two feet on the 10 th rung, reverses direction and climbs down until both feet are again on the floor. The applicant is given 3 minutes of rest before the next test. Figure View of the Ladder Climb Test. The applicant must maintain three-point contact (two feet and one hand, or one foot and two hands) at all times on the ladder. Fall protection is provided by using a safety rope attached to the SCBA, run through belay points overhead and near the floor. The applicant is on belay throughout the test.

130 118 Equipment Carry/Vehicle Extrication - starting from an erect position facing the tools, the applicant lifts a 17 kg spreader tool, carries it m, sets it down, and then returns to the start to repeat the process with a 36 kg spreader tool (Hale Products Inc., Conshohocken, PA). The applicant then picks up the smaller tool, carries it 7.62 m and sets it down in front of a vehicle door mock-up. The tool must be held in a level position at right angles to the door mock-up with the jaws in firm contact with each of three flat metal discs that are similarly oriented to the three pins that must be broken to remove a car door. The tool is held in the correct position for 30 s on each disc. The tool is set down between each hold and the applicant must stand erect before lifting the tool and moving on to the next point of contact. After this sequence is completed, the applicant returns both tools to the starting point. This test is scored on a pass-fail basis. In order to pass the test, the applicant must complete all aspects of the simulation safely and with correct form within the time allowed. Figure View of the large (left) and small (right) spreader tools used in the Equipment Carry/Vehicle Extrication Test.

131 119 A B Figure Panel A shows the correct technique for carrying the spreader tools. Panel B shows correct technique for lifting and lowering the spreader tools during the Equipment Carry/Vehicle Extrication Test.

132 120 Figure View of the Vehicle Extrication Test. Note the orientation of the jaws of the spreader tool to the flat metal disk on the car-door mock-up. The spreader tool must be held in the correct orientation for 30 s at each of the three flat metal discs. The applicant must lift the tool from the floor and place the tool down on the floor before and after each hold.

133 121 It should be noted that the job-related test is preceded by the treadmill test for aerobic fitness, which is followed by a 60-minute rest period. The pilot research to develop the treadmill test was completed simultaneously with the pilot work to develop the job-related tests since both were required to be in close to final form before the main data collection phase could begin Main Data Collection Introduction The first purpose of this phase of the research was to validate the job-related test protocols. In order to satisfy this purpose, it was necessary to demonstrate that the physical demands of the applicant tests were consistent with the physical demands of the CF/DND FF Test and firefighting work. The second purpose was to use the performance data from a sample of the workforce to create standards for screening applicants. That is, to identify an appropriate standard of performance for each test that was consistent with a pass. The third purpose of this phase was to identify appropriate standards for selection of applicants based on physical fitness. That is, to create an appropriate system of ranking those applicants who met or exceeded the standards for screening Validation of Applicant Test Protocols During this step performance data and subjective feedback were acquired from a representative sample of the workforce who completed both the applicant test and the incumbent test. This process, in effect, was intended to validate the applicant test protocols. In other words, if the applicant test protocol was a valid representation of the physical demands of the CF/DND FF Test, there should be good agreement between the performances of experienced incumbents on the two tests. If on the other hand, the applicant tests were not valid representations of the physical demands of the CF/DND FF Test, then the pilot work stage would have to be repeated. With this possibility in mind, this step was completed according to the following general sequence:

134 122 A sample of volunteer incumbents completed the CF/DND FF Test (the circuit) and the Applicant Test on separate days. Relationships between the performance of selected tasks within the circuit and performance of the same task in the applicant protocol were evaluated. The researchers operated on the assumption that if the relationship was strong, then the applicant test protocol was an accurate reflection of the existing BFOR for incumbents. Conversely, if the relationship was weak, further research was required. In addition to measurements of performance time on the tests, the incumbents rated their perception of effort (Rating of Perceived Exertion or RPE) immediately upon completion of the selected tasks within the circuit or within the applicant protocol. After completing both test protocols, each incumbent responded to a questionnaire designed to evaluate the similarity of the physical demands of each individual job-related test to the physical demands of the related element within the CF/DND FF Test and the physical demands of similar activities encountered on the job. Since training is a significant part of the job in firefighting, the incumbents were directed to include training experiences as part of job experiences. A convenience sample of 37 incumbent CF firefighters completed this process and the data were analyzed. The results of this analysis are presented in the Results section of this chapter. In brief, the results supported the notion that the physical demands of the discrete tests within the applicant protocol were very similar to the physical demands of the circuit and firefighting work. In order to acquire data from the 37 volunteers, the research team visited three large CF fire departments at bases in Edmonton, Comox and Cold Lake. The participation rate (approximately 31% of eligible firefighters) was not satisfactory to the investigators and this concern was discussed with CFFM and CFPSA Research Manager. Upon the recommendation of CFFM, the researchers began

135 123 to recruit civilian firefighters from the Edmonton area. The researchers maintain long-standing professional relationships with the City of Edmonton Emergency Response Department, Strathcona County Emergency Services, City of St. Alberta Fire Department and Fire ETC (Emergency Training Centre). Since military and civilian firefighters train to the same standard and are charged with the same duties related to fire suppression, rescue and public safety, the decision to include civilians was made to facilitate the progress of the project without compromising the quality of the research. The procedures were modified to allow orientation to and additional practice on, the CF/DND FF Test. The civilian firefighters were routinely given an orientation the FF Test, followed by at least two practice runs before data collection. This level of preparation provided a level of exposure to the FF Test that was consistent with the experience of many of the CF incumbents previously tested. Data were acquired on 20 additional volunteers from civilian jurisdictions in the Edmonton area. The analysis comparing performance times, perceptions of effort and similarity of physical demands was repeated on the larger data set. The conclusions drawn from the first analysis (n=37) were supported by the second analysis (n=57). The firefighters participating in the validation phase of the project completed a questionnaire designed to evaluate the relatedness of the physical demands of each of the elements of the Applicant Test to the physical demands of the similar element within the FF Test and similar activities encountered during their experience in fire-rescue work or training. As shown below, the firefighters were asked to choose from one of seven responses to rate the similarity of the physical demands. The seven responses were subsequently assigned numerical values from 7 (Very similar) to 1 (Very different) by the researchers to permit analysis of the questionnaire.

136 124 The potential responses and the associated numbers (later assigned by the researchers) are listed below: Very similar 7 Similar 6 Somewhat similar 5 Uncertain 4 Somewhat different 3 Different 2 Very Different 1 The questions posed to the incumbent firefighters are listed below, and in each case, the firefighters chose their answers from the seven potential responses shown above: Question 1 In comparison to the charged hose drag part of the FF Test, would you say that the physical demands of the Applicant hose drag were: Question 2 In comparison to your experiences on-the-job or in fire-rescue training, would you say that the physical demands of the Applicant hose drag were: Question 3 In comparison to the rope pull part of the FF Test, would you say that the physical demands of the Applicant rope pull were: Question 4 In comparison to your experiences on-the-job or in fire-rescue training, would you say that the physical demands of the Applicant rope pull were: Question 5 In comparison to the forcible entry part of the FF Test, would you say that the physical demands of the Applicant forcible entry were: Question 6 In comparison to your experiences on-the-job or in fire-rescue training, would you say that the physical demands of the Applicant forcible entry were: Question 7 In comparison to the victim rescue part of the FF Test, would you say that the physical demands of the Applicant victim rescue were:

137 125 Question 8 In comparison to your experiences on-the-job or in fire-rescue training, would you say that the physical demands of the Applicant victim rescue were: Question 9 In comparison to the ladder climb part of the FF Test, would you say that the physical demands of the Applicant ladder climb were: Question 10 In comparison to your experiences on-the-job or in fire-rescue training, would you say that the physical demands of the Applicant ladder climb were: Question 11 In comparison to your experiences on-the-job or in fire-rescue training, would you say that the physical demands of the Applicant vehicle extrication test were: Question 12 In comparison to the FF Test, would you say that the physical demands of the Applicant treadmill test were: Question 13 In comparison to the FF Test, would you say that the physical demands of the overall Applicant test were: Question 14 In comparison to your experiences on-the-job or in fire-rescue training, would you say that the physical demands of the overall Applicant test were: Reliability of the Applicant Tests It was noted at the beginning of Section 3 that practice generally improves performance on the FF Test. The improved performance may be due to better pacing strategies or skill development (e.g., ladder raise), and/or improved fitness if the exposure to the test is sustained enough to elicit a training effect. One of the main reasons for adopting the discrete tests for the Applicant testing program was to avoid the problem of pacing. The elements of the FF Test that required relatively more skill (e.g., ladder raise) were eliminated. It must be acknowledged that extensive exposure to any physical task may improve fitness and hence, performance.

138 126 A sub-study was conducted to evaluate the reliability (or reproducibility) of test scores on the discrete job-related tests. Thirteen male University staff and students volunteered to complete the job-related tests on two separate days (Day One and Day Two). None of the subjects were firefighters. Participants were scheduled at approximately the same time of day, and normally 24 hours (in one case, 48 hours) elapsed between tests. Each test session was organized in exactly the same manner as for testing applicants. Participants in the sub-study were briefed on procedures and were dressed in the correct protective clothing ensemble. Participants were informed of the need to complete the first five tests as quickly as possible. They were also informed of the criteria for passing or failing the final test. Participants were asked to give a maximal effort on each test, both days. Following the standardized warm-up, each participant completed all of the job-related tests in the correct sequence and according to the standardized protocol. Participants were not informed of any of their test scores until the final test had been completed on the second day Determining the Minimum Standard for Applicant Tests The purpose of this step was to establish an appropriate minimum standard for each of the job-related tests within the applicant protocol. Previously, researchers have relied on statistical analysis of incumbent test data (Gledhill and Jamnik, 1992b) to set minimum standards. Given a normal distribution of test scores, the mean performance time plus two standard deviations should capture approximately 96% of the range of scores in the population. However, this method does rely on the assumption that the test scores from the sample of volunteers are, in fact, normally distributed. A further assumption is that the sample of volunteers is representative of the population. Subject matter expert evaluation of performance (IAFF, 1999; Sothmann et al., 2004) is a powerful method of setting the minimum acceptable performance

139 127 level. In this method, subject matter experts (SME) are asked to evaluate performance, and based on their experience, the SME can judge what is acceptable or not. Our approach was to incorporate both methods to determine the minimum standard for the job-related performance tests. This process involved several steps. First, in order to improve the chances of working with a more representative sample of experienced firefighters, we continued to recruit volunteers from local fire jurisdictions in the Edmonton area to complete the Applicant protocol. This increased the number of subjects to 93. Demographic characteristics of this group are shown in Table Second, we conducted an analysis of the performance results from this group. A sample frequency distribution for the hose drag test is shown in Figure We were satisfied that the distribution of scores was relatively close to normal, given the potential constraints that might limit achieving a true normal distribution. For example, the rules governing the completion of each test prevent subjects from achieving very low times. Using the hose drag test as an example, it is important to remember that applicants are not allowed to run. Therefore, the fastest acceptable time will be constrained by this rule and will likely prevent a true normal distribution. Also, it is important to remember that our subjects were all volunteers, and therefore, not a random sample of the population of interest (all incumbent firefighters). Third, we created a video showing a mock applicant (either S. Petersen or R. Dreger) performing each of the job-related performance tests at selected times corresponding to: The Mean time of the sample of firefighters (n=93) The Mean time plus one standard deviation The Mean time plus two standard deviations The Mean time plus three standard deviations

140 128 The Mean time plus four standard deviations The mean time was always shown first and the other performances then followed in random order. It should be noted that for the Rope Pull test, the video did not include the mean plus four standard deviations segment. When making the video, we found it impossible to have continuous movement at the slowest speed, and the intermittent (stop and start) nature of the performance appeared very different than the other performances of this task. Care was taken to ensure that certain aspects of the performance were kept as consistent as possible. For example, in the filming of the Victim Rescue test, the amount of time devoted to lifting the rescue dummy was kept consistent while the amount of time spent dragging the dummy was varied. Similarly, the walk time between repetitions of the Rope Pull test was kept consistent, while the pull time was varied. Fourth, the video was shown to 23 subject matter experts. The SME were all experienced firefighters who were: Familiar with the FF Test and the Applicant Test, and, In supervisory positions in either front line fire service (e.g., platoon chief) or at a nationally recognized training facility (e.g., training officer at Fire ETC or CFFA). Each SME viewed the video tape in the company of one of the researchers and indicated whether each performance was either acceptable or not acceptable according to his/her standards for safe and effective performance of similar fire rescue duties on the job or in training. Twenty-one of the SME were male (91%) and two were female (9%). Mean fire service experience of the SME was 20.8 (7.5) years. Jurisdictions included the Canadian Forces Base Comox, Canadian Forces Fire Academy, Fire ETC, and the City of Edmonton Fire Rescue Department. The researchers were satisfied

141 129 that these experts represented both fire suppression and training perspectives within civilian and military jurisdictions. For each test, regression analysis was performed (personal communications, Dr. Bruno Zumbo, University of British Columbia) to find the point of transition between acceptable and not acceptable responses from the SME. The transition point was then used as the minimum acceptable score for each test (Table ). A minimum acceptable completion time for the Vehicle Extrication/Equipment Carry Test was established by examining the scores from the group of 93 firefighters. The mean completion time for this group was s, with a range of 127 to 270 s. The mean time plus 2 SD was 276 s, which was very similar to the slowest firefighter. The selection of 270 s as the minimum time did not have a negative impact on any of the firefighters we studied, and consequently, this was chosen as the requirement to pass this test. Other criteria for successful completion of the Vehicle Extrication/Equipment Carry Test included: the requirement to lift, carry and set down the tools in a safe manner; and, the requirement to hold the small spreader tool in the correct position for 30 s at each of the three sites. Therefore, while the test is scored on a pass-fail basis, the three criteria must be met in order to pass Determining the Optimal Performance Level for Applicant Tests For the purposes of using physical fitness scores as a means of ranking firefighter applicants, a percentile analysis on test scores for the five performance tests from a group of 150 trained firefighters was conducted. The Equipment Carry/Vehicle Extrication Test was not included in this process, since the test is scored on a pass or fail basis. The 90 th percentile was selected as representative of an excellent level of performance, similar in principle that the top segment (usually 10%) of an academic class represents an excellent level of performance.

142 130 For the purposes of ranking applicants based on performance on five of the jobrelated tests, a 6-point scale (0-5) was established for each test. As noted above, the minimally acceptable time (passing level of performance) was determined from the analysis of the evaluations from the Subject Matter Experts. Scores failing to meet this criterion were assigned 0 points. The highest point value (5) was assigned to those scores who met or exceeded the 90 th percentile. Scores associated with 1 to 4 points were based on the creation of equal segments between the passing score and the score associated with the 90 th percentile. The mean score for the 150 firefighters fell within the range associated with 3 points.

143 Results Selected demographic characteristics of the incumbent firefighters who volunteered for the validation phase of the research are shown in Tables (n=37) and (n=57). Table Selected demographic characteristics of incumbent firefighters (n=37) Variable Mean SD Minimum Maximum Age (yr) Height (cm) Weight (kg) Fire Service (yr) FF Test time (s) Table Selected demographic characteristics of incumbent firefighters (n=57) Variable Mean SD Minimum Maximum Age (yr) Height (cm) Weight (kg) Fire Service (yr) FF Test time (s)

144 132 The results of the correlation analysis between selected elements of the FF Test and the corresponding element in the Applicant test protocol are shown in Table In each case, the correlation between the performance times with 37 firefighters was essentially unchanged or improved with the addition of more data. However, in all cases, the correlation coefficient was significant (p<0.01) Table Correlation coefficients between selected components of the FF-test and Applicant test with 37 and 57 firefighters Variable n = 37 n = 57 Hose Drag Rope Pull Forced Entry Victim Rescue Ladder Climb* * The time for the FF-Test Ladder Climb is a combination of the first climb time (3 times up and down) and the 2 nd climb (another 2 times up and down). The Applicant Ladder Climb consists of 5 consecutive times up and down the ladder. The data for the Rope Pull test are displayed in Figure as an example of the correlation analysis between performance times on the elements of the FF Test and the corresponding element of the Applicant Test. The addition of more data did not alter the relationship.

145 Applicant Rope Pull Time (s) n = 56 n = FF Test Rope Pull Time (s) Figure The relationship between FF Test Rope Pull time and the Applicant Test Rope Pull time. T he rating of perceived exertion (RPE) values obta ined from the smaller ( n=34) and larger (n=56) groups of incumbent firefighter s in th e validation phas e of the research are displa yed in Table The RPE value was a cquired fro m the subjects i mmediately upon completion of each element of the FF Test or the related element of the Applicant test p rotocol. There were several cases of missing data, where the firefighter did not respond when ask ed for the RPE value.

146 134 Table Ratin g of Perceived Exertion (RPE) values for selected components of the FF-test and the Applicant test with inc reased incumbent numbers Variable n = 34 n = 56 mean median mode mean median mode Hose FF-Test Drag (3) (20%) (3) (28%) Applicant (3) (30%) (3) (23%) Rope FF-Test Pull (2) (33%) (2) (38%) Applicant (2) (36%) (2) (34%) Forced FF-Test Entry (2) (44%) (2) (33%) Applicant 14 (3) (27%) 14 (3) (27%) Victim FF-Test Rescue (2) (35%) (2) (33%) Applicant * (2) (18%) (2) (23%) Ladder Climb FF-Test 1 13 (2) (35%) 12 (2) (40%) FF-Test (2) (30%) (2) (27%) Applicant 15 (2) (24%) 15 (2) (29%) Mean is represented (± SD). Median and Mode are the middle response and the most common response (and percentage), respectively. *RPE for the Applicant Victim Rescue was bimodal, with RPE of 13 and 14 being scored equally (18%). Thus the mode for the Applicant Victim Rescue is reported as The firefighters participating in the validation pha se of the project completed a questionnaire designed to evaluate the relatedness of the physical demands of each of the elements of the Applicant Test to the physical demands of the similar element within the FF Test and similar activities encountered during their experience in fire-rescue work or training. The responses to these questions are displayed in Table

147 135 Table Frequency Ana lysis of the Questionnaire Responses Question M ean Median Mode (1.8) (1.4) (1.3) (1.2) 6 6 (28%) 6 5 (34%) 6 6 (39%) 6 6 (46%) (1.8) (28%) (1.8) 6 6 (33%) (1.0) (61%) (1.7) 6 7 (30%) (1.5) (1.5) 6 7 (31%) 6 6 (40%) (1.7) 6 6 (37%) (1.9) (22%) (1.7) (1.0) Data are shown as Mean (± SD). 5 6 (34%) 6 6 (44%) Median and Mode are the middle response and the most common response (and percentage), respectively. The response to Question 5 was bimodal, with somewhat similar (5) and similar (6) returning equal responses (28%). Thus the mode for Question 6 is reported as 5.5.

148 136 Table Selected demographic characteristics of volunteers in the reliability sub-study (n=13) Variable Mean SD Minimum Maximum Age (yr) Height (cm) Body Mass (kg) Total Mass (kg) (Body + PPE) Table Mean (±SD) performance times for (n=13) participants completing the job-related tests on two separate days Test Time (s) Day One Time (s) Day Two Correlation Day 1 and Day 2 Charged Hose Advance 16.1± ± High Volume Hose Pull 59.3 ± ± Forcible Entry 8.6 ± ± Victim Rescue 26.2 ± ± Ladder Climb 64.7 ± ± Equipment Carry/Vehicle Extrication ± ±

149 137 The performance times for the first five tests where applicants are encouraged to perform the tas ks as fast as possible wer e added together to obtain a total work time. The total work time for Day One and Day Two was compared. Total time on Day One was ± 19.0 seconds, and for Day Two, ± 21.9 seconds. The mean difference of 1 second was not significantly different. Total work time was also calculated including the Equipment Carry/Vehicle Extrication test was added in, and total work time was then compared for Day One and Day Two. Total time on Day One was ± 21.6 seconds, and for Day Two, ± 32.0 seconds. The difference between the means was not significantly different. Table Mean (±SD) points for (n=13) participants completing the job-related tests on two separate days Test Points Day One Points Day Two Charged Hose Advance 4.1 ± ± 0.6 High Volume Hose Pull 3.7 ± ± 0.8 Forcible Entry 4.5 ± ± 0.5 Victim Rescue 4.0 ± ± 0.8 Ladder Climb 3.7 ± ± 0.7 Total Points 19.9 ± ± 2.6 By way of example, the distribution of scores between Day one and Day two on the Charged Hose Test is shown as a scatterplot in Figure below.

150 Day 1 Hose Advance Time (s) Day 2 Hose Advance Time (s) Figure Scatterplot showing the scores for the Charged Hose Advance test on Day one and Day two of the reliability sub-study ints Day 1 - Total Po Day 2 - Total Points Figure Scatterplot showing the scores for the total points earned on the five performance tests on Day one and Day two of the reliability sub-study.

151 139 Table Selected demographic characteristics of incumbent firefighters* (n=93) Variable Mean SD Minimum Maximum Age (yr) Height (cm) Weight (kg) Fire Service (yr) *Male = 90 and Female = 3 Count Hose Drag Figure Distribution of performance times for the Hose Drag test (n=93). Subject matter experts were shown video of the task performed at the mean time, mean + 1SD, +2S D, +3SD and + 4SD.

152 140 Table Breakdown of Pass and Fail responses of fire serv ice Subject Matter Experts (n=23) for selected performance times o n job-re lated t ests Hose Drag Rope Pull Forcible Entry Victim Drag Ladder Climb Time (s) Pass Fail Time (s) Pass Fail Time (s) Pass Fail Time (s) Pass Fail Time (s) Pass Fail Cut time Table Selected demographic characteristics of firefighters* (n=150). The 90 th percentile of this group was used to determine the optimal performance level for selection purposes Variable Mean SD Minimum Maximum Age (yr) Height (cm) Weight (kg) Fire Service (yr) *Male = 137 and Female = 12

153 141 Task Table Job-related test scores and associated point distributions where applicable. Hose Drag (s) Rope Pull (s) Forcible Entry (s) Victim Rescue (s) Ladder Climb (s) Pass/fail Points < > < < < < > 49 > 8.0 > 23.0 > 54.0 Vehicle Extrication 270 n/a n/a n/a n/a n/a (s) Notes: Pass/fail cut-points (0 points) for hose drag, rope pull, forcible entry and ladder climb are from analysis of SME evaluations. Optimal score (5 points) for hose drag, rope pull, forcible entry and ladder climb derived from 90 th percentile of experienced firefighters (n=150).

154 Discussion Validity of Job-Related Performance Tests The objective of this phase of the project was to establish that the physical demands of the Applicant tests were similar to the physical demands of the related element of the FireFit Test. The investigators accepted the evidence provided by Deakin et al., (1996) that the FireFit Test was a valid simulation of firefighting work. However, the part of the questionnaire that checked for similarity between the physical demands of the various parts of Applicant Test and the physical demands of related on-the-job experiences provided independent confirmation of the validity of these test items to the occupation of firefighting. Collectively, the results of the validation process showed that there was good agreement between the physical demands of the individual tests within the Applicant Protocol and the physical demands of the related elements from the FireFit Test. Correlation analysis revealed that performance time for each element of the Applicant Test was significantly related to performance time on the related element from the FireFit Test. Furthermore, the RPE data obtained immediately on completion of each test showed that the perception of physical exertion was very similar. The results of the questionnaire administered after the tests were completed revealed that, on reflection, the firefighters found the physical demands of the individual tests from the Applicant protocol and the associated element of the FireFit Test to be very similar. Finally, the results of the questionnaire also showed that the physical demands of the individual tests from the Applicant protocol were believed to be similar to the physical demands of related activities from on-the-job (or training) experiences. The congruence of the results from the various techniques used to check validity show that the Applicant Test can be considered a good replication of the physical demands of firefighting.

155 Reliability of Job-Related Performance Tests The purpose of this part of the research project was to evaluate the reliability of test scores from the Applicant Test. Repeated measures analysis of variance revealed no significant differences between either the raw scores (time) or points for any of the job-related tests. Pearson product-moment correlation analysis revealed significant (p<0.05) moderate to high relationships for Day 1 and Day 2 performances on the job-related performance tests. However, it should be noted that while the mean scores for the group did not show significant variability between test days, it is possible for individual scores to vary somewhat. A good example is shown in the scatterplot from the Charged Hose Advance test. Four of the data points lie on the line of identity, showing identical times for both tests. Five of the data points lie slightly above the line of identity, showing that these individuals performed better on the second test. Four of the data points lie slightly below the line of identity, showing that these individuals performed better on the first test. The highest variability between test days was approximately 2.5 s, or roughly 12% of the test time for that individual. On the other hand, the rest of the subjects showed far greater consistency, which is encouraging. Since raw scores are converted to points (see Table ), small variations in the raw score should not normally affect the associated point value. The obvious exception would be when an applicant s raw score is very close to the breakpoint between point categories. Using the Charged Hose Advance data as an example, the range of raw scores in performance time within a point category is approximately four seconds. The highest variation between test days was 2.5 seconds. Therefore, depending on where the initial test score fell with respect to the time range associated with a particular category, it is possible that on the second day, the score would warrant a higher or lower point value. On the other hand, it seems equally probable that even at the highest level of variability, the point value may not change. This is illustrated in Table where the mean

156 144 points for each test and total points (all tests added together) were not different between test days. Similarly, when all test times were added together, there was approximately one-second difference (much less than 1%) between total work time on Day One and total work time on Day Two. Therefore, while small differences in day-to-day performance on any test should be expected, two main conclusions may be drawn. First, the variability in performance from day-to-day is not systematic. Secondly, it seems unlikely that the small day-to-day variation should have any significant effect on evaluation of the applicant. While the overall picture shows good reproducibility between test days, it must be acknowledged that there will always be some variability in human performance. Part of this may be explained by biological variability, technical variability and mental state. Biological variability is generally minimized by advising the subject to maintain good sleep and nutrition habits and through careful scheduling. That is, the second test should occur at the same time of day as the first test. There must be adequate recovery time between the tests (e.g., hours), yet not enough time (e.g., days to weeks) to allow either training or detraining to occur. Technical variability is minimized by careful attention to the organization of the testing environment. For example, in the case of the job-related tests under consideration, factors such as wearing the same clothing each time, verifying that the distances, the friction of the testing surface, the weight of the objects, and the test instructions are identical will keep technical variability to a minimum. Mental state may include factors such as motivation and understanding of the task. Motivation is largely beyond the control of the tester, however maintaining a positive test environment with encouragement for the subject may help. Understanding the task is clearly an essential element of reliability and the tester must always ensure that the subject knows exactly what is expected each time. Notwithstanding every effort to minimize variability in biological, technical and mental state, there may always be some degree of variability in test scores from day to day. For example, our subjects were asked to refrain from other heavy

157 145 exercise during the sub-study so that fatigue would not influence performance. However, the willingness of the subject to comply with that request is beyond the control of the investigator. Similarly, the investigator cannot control the motivation of the subject to provide a maximal effort each time. While the potential variability in human performance must be recognized, overall there is little evidence suggesting that these tests do not show reasonable reliability from day to day. There is no compelling evidence to show that systematic improvement on the second day, which would be of serious concern. The Equipment Carry/Vehicle Extrication test is scored on a Pass-Fail basis and therefore, should not be considered in the same manner as the other five tests where the applicant is encouraged to complete the task as quickly as possible. In the test protocol for the Equipment Carry/Vehicle Extrication test, the applicant is advised first of the total time allowed to complete all of the activities within the test, and secondly, is advised throughout the test of the overall elapsed time. Thus, the applicant may choose to move through certain parts of the test protocol (e.g., the tool carries) somewhat faster or slower. However, since there is no direct benefit from completing the protocol faster than the minimum acceptable time of 270 s, our experience shows that most individuals are simply concerned with completing the test within the time limit. The correlation analysis revealed a much weaker relationship between completion times on the two days. This indicates a higher degree of individual variability from day to day. However, there were no differences in the mean times from the two tests, showing variability within the group was consistent. Overall, we conclude that the reliability of these tests is satisfactory when biological, technical and mental factors are controlled to the extent possible. When implementing the test protocol, it will be absolutely essential to create an environment that allows applicants to do their best. For example, the implementation process should address provision of knowledge of the tests and

158 146 optimal training methods well in advance (e.g., 3 to 6 months) of the test date. Applicants must be provided with guidelines for optimal preparation (e.g., nutrition and rest) in the days leading up to the test. On the test day, there must be full and consistent instruction to complement the warm-up and practice components. The testing process must include the opportunity for the applicant to question procedures to ensure understanding of each element of the protocol. All of these steps will improve the general concept of test reliability. In the final analysis, good reliability implies that the test result is determined by fitness level, not other factors such as experience (or ignorance) Determining Minimum Standards A multi-step process was used to set the minimum acceptable level of performance on each test. First, simple descriptive statistical analysis was used to evaluate the demographic characteristics of the group of 93 firefighters. As can be seen in Table 3.3-9, there was a wide range of age, experience and body size. Therefore, it appears that the characteristics of these volunteers are consistent with the diversity that might be expected within the fire service. Second, a frequency distribution analysis was performed to show the distribution of performance scores within the group of volunteers. Figure shows a typical frequency distribution, which approaches a normal distribution over a wide range of performance times. It seems unlikely that a normal distribution will ever result when using volunteers and when the rules of the test prevent very fast completion times. For example, in the case of the Charged Hose Advance, the firefighters were not allowed to run with the hose. However, when setting the minimally acceptable speed of work, the slower performances are of greater interest. The slower performances probably result from poor physical capacity and/or lower motivation levels. While it is not possible to explain the reasons for the slow performances, the important aspect is the wide range in performance times and a reasonably normal distribution. These two factors give some confidence in the average performance, and also in deviation from the mean.

159 147 Third, the analysis of the evaluations from the Subject Matter Experts (see Table ) identified the point of transition between what the SME judged to be acceptable and unacceptable work rates for each test. The use of subject matter expert opinion may be referred to as intuitive evaluation and recognizes the ability of experienced individuals to identify acceptable or unacceptable levels of performance. The method has been used previously in firefighting research, specifically for purposes related to identification of acceptable work rates in setting physical test standards (IAFF, 1999; Sothmann et al., 2004). In this project, expert opinion identified acceptability based on criteria for safe and effective work rates in fire-rescue activities that were similar to the activities performed in the tests. Each SME rated each performance based on his/her personal criteria for safe and effective performance of similar tasks. Subsequently, our analysis identified the point of transition between what the SME judged to be acceptable and unacceptable work rates for each test. Typically, this transition point fell between the times associated with the mean time plus two and three standard deviations. The method of Gledhill and Jamnik (1992b) used the mean plus two standard deviations to set minimum standards for performance times on similar tests Determining Optimal Standards The scores consistent with the 90 th percentile for the group of 150 experienced firefighters were used to denote an optimal performance level for each jobrelated test where applicants are encouraged to complete the test as quickly as possible (Hose Drag, Rope Pull, Forcible Entry, Victim Rescue and Ladder Climb). Demographic characteristics for this group are shown in Table , and the performance times associated with the 90 th percentile for each test are shown in Table The 90 th percentile captures the very best scores from the sample of experienced firefighters. Using the 90 th percentile to identify an optimal level of performance

160 148 was somewhat arbitrary, however the decision was grounded in experience and common sense. The reader should bear in mind that the intent of this step was to allow the end-user of the testing protocol the ability to make distinctions between applicants based on physical fitness. Once the screening process was completed by identifying those applicants who either passed or failed based on achieving the minimum standard of performance, the next possibility would be to use fitness for the purpose of selection. It may be helpful to use the analogy of assigning grades to the fitness scores in the same manner as a teacher would assign grades to students based on academic achievement. Gledhill and Jamnik (1992b) used a similar method, however those authors selected the mean score from the group of incumbent firefighters as the optimal standard of performance. Our long experience with testing firefighter applicants reveals that most applicants can be expected to exceed the mean performance of incumbents, and consequently, the method of Gledhill and Jamnik (1992b) does not effectively discriminate between passing applicants to identify levels of fitness that might be associated with descriptors such as good, very good and excellent. The intent of our method was to create an adequate distribution of possible scores and categories (in our case, points awarded based on performance level) so as to use the entire range of performance to grade applicants. Grading practice at schools and universities generally is based on the expectation that approximately 10% of students in a class should earn excellent grades. Fitness categories and norms typically classify the top 10-15% of the population as excellent (CSEP, 1986). Therefore, while somewhat arbitrary, the use of the 90 th percentile is consistent with other applications where the intent is to classify performance in a group.

161 Interpretation of Test Scores The minimum standard for each test was determined based on the input from fire service SME. The optimal standard for each test was determined based on the performance of a reasonably large group of experienced firefighters. The range between the minimum acceptable level and the optimal level was divided into equal time periods which are analogous to categories of performance. Each category was assigned a number so that applicants could earn points based on their performance on each test. The distribution of points for each of the jobrelated tests is shown in Table A similar method was used to identify levels of excellence in VO 2max (see Table 2.5-2). For the purposes of screening, an applicant must meet the minimum acceptable level of performance on each test. For the purposes of selection, an applicant can be assigned points based on performance. The lowest number of points is six and the highest possible number is 30. This provides a wide range of possibilities for classifying physical fitness, and in our experience, the full range may be expected.

162 150 SECTION 4 EVALUATION OF THE APPLICANT TEST PROTOCOL 4.1 Introduction This test protocol was developed to evaluate the physical fitness of firefighter applicants. In the interest of fairness, the test should not be biased on any grounds other than physical fitness. Applicants with greater levels of fitness should do better on the tests and applicants with lower levels of physical fitness should do worse. Applicants should not pass or fail systematically on grounds other than fitness. Characteristics such as gender, age and experience should not have a systematic influence on the test outcome. Ultimately, a thorough analysis can only be done after the test has been implemented and a large database has been acquired. However, part of the research process leading to this report involved a trial implementation of the test protocol at the University of Alberta with local fire departments and training schools. The resulting data was used to provide a preliminary insight into whether the test was systematically biased on the grounds of fire training experience, gender and age. The University of Alberta has long-standing relationships with numerous fire departments in Alberta, as well as two fire-training schools. The Applicant Test protocol was implemented on a trial basis to evaluate firefighter applicants for the City of Edmonton, the City of St. Albert, the County of Strathcona, the City of Fort McMurray, and the City of Lethbridge. As well, applicants to the full-time NFPA 1001 program at Fire ETC (Vermilion, AB) and students graduating from the same program at Emergency Services Academy (Edmonton, AB) were tested. Data were acquired on 427 firefighter applicants who were in actual job competitions. The results of testing 399 male and 28 female applicants allowed a preliminary evaluation of the test with respect to gender. The test performances from younger and older applicants from the large group of males allowed a preliminary evaluation of the test with respect to age. The test results

163 151 from 75 students applying to the NFPA 1001 program were compared to the results of 75 students graduating from the NFPA 1001 program in order to evaluate the effect of fire-training on test performance. It should be stressed that these comparisons are based on test results from convenience samples. The results are preliminary and cannot be considered definitive at this time. On the other hand, it should be stressed that the 427 applicants were involved in actual job competitions for five separate fire departments. Therefore, the physical and performance characteristics of these individuals are consistent with the real world. The 150 fire-training students were either applying to or graduating from recognized training schools and again, these should be recognized as being real data. Ultimately, the evaluation of this type of test can really only be done after a large database has been acquired from testing the special individuals for whom it was designed in the first place. The results of the preliminary evaluations are encouraging, however it will be important to continue this process. 4.2 The Effect of Fire-Training on Test Performance In order to examine whether or not fire-training had a systematic effect on the outcome of a test that is intended to be fitness-based and not skill-based, two groups were studied. The members of the first group were applying to the NFPA 1001 program at Fire ETC (Emergency Training Center). These applicants have to provide evidence of passing a firefighter-related physical fitness test as a condition of acceptance into the training program. The second group consisted of students who were graduating from the same program at the Emergency Services Academy (ESA). In this case, as part of the program, ESA provides the students with exposure to the type of physical fitness test protocol that they will typically encounter when applying for employment. At both schools, the students undertake approximately 11 weeks of full-time study (theory and practical) in order to earn the NFPA 1001 credential.

164 152 Convenience samples of seventy-five male students were selected from each program. ESA typically runs three classes of students in a year. The vast majority of these students are males, and within the one-year time period allotted for this part of the research project, there were data available from 75 male students. The University of Alberta typically tests approximately applicants to Fire ETC each year. A convenience sample of 75 males was selected from the pool of data acquired during the same year for comparison with the graduating students. For the purposes of this evaluation, the two groups were designated as Trained (those graduating from the NFPA 1001 program) and Untrained (those applying to enter the NFPA 1001 program). The main question of interest was whether or not the skills acquired during the extensive training period would systematically influence the test results. If test performance was truly based on physical fitness, then there should be no systematic difference between the groups. On the other hand, if firefighting skills (e.g., advancing charged hose, climbing ladders) influenced test performance, then we reasoned that the Trained group should demonstrate systematically superior results. Descriptive and performance characteristics for the Trained and Untrained groups are shown in Table Unpaired t-tests revealed a small but statistically significant difference between the two groups in age. However, there were no significant differences in height, body mass, or in previous firefighting experience (typically volunteer). There was also a small but significant difference in aerobic fitness level as indicated by VO 2max. In this case, the adjustment of VO 2max relative to weight takes into account both body mass and the weight of the PPE. The difference of approximately 6% is the same when VO 2max is expressed relative to simply body mass. However, it seems unlikely that the difference in aerobic fitness can be attributed to fire-training, since there is no physical fitness component to the training course at ESA.

165 153 Table Mean (± SD) Physical Characteristics and Test Results for Trained (graduating NFPA 1001) and Untrained (applying to NFPA 1001) Fire-training Students Variable Group Combined (n=150) Trained (n=75) Untrained (n=75) Age (yr) 24.9 (4.8) 26.1 (5.2) 23.7* (4.1) Height (cm) (7.0) (6.5) (7.6) Body Mass (kg) 86.9 (12.3) 86.6 (10.9) 87.5 (14.0) Firefighting Experience (yr) 0.7 (1.7) 0.5 (1.6) 0.9 (1.8) VO 2max (L min -1 ) 3.8 (0.5) 3.9 (0.5) 3.7* (0.5) 1 VO 2max (ml kg -1 min -1 ) 35.2 (4.4) 36.3 (4.1) 34.2* (4.4) Treadmill Time (min) 18.2 (2.2) 18.6 (2.1) 17.8 (2.3) Hose Drag Time (s) 18.4 (4.2) 18.1 (5.5) 18.8* (2.3) Rope Pull Time (s) 60.5 (12.2) 62.6 (14.0) 58.4 (9.9) Forced Entry Time (s) 14.2 (6.4) 16.8 (7.1) 11.6* (3.0) Victim Rescue Time (s) 27.5 (6.1) 27.9 (7.1) 27.1 (5.0) Ladder Climb Time (s) 75.4 (14.8) 76.7 (16.7) 74.0 (12.5) Equipment Carry 2 (# Failures) Total Failures relative VO 2max is expressed per kg gear weight (weight of subject in PPE and wearing SCBA) 2 The Equipment Carry is graded as a pass/fail event, so time is not reported here * p<0.05 = significant difference between trained and untrained

166 154 The significant differences in the mean completion times for the Hose Drag and Forced Entry are more relevant to the question under review, however there was no systematic result that could be attributed to skills acquired in training. The Trained group performed slightly better on the Hose Drag test while the Untrained group performed slightly better on the Forcible Entry test. In both cases, the performance differences were small. The differences do not appear to be systematic nor can they be explained by training status. Overall, 9.3% of the 150 students failed the Applicant Test (Table 4.2-1). The failure rates were 10.7 and 8.0% for the Trained and Untrained groups, respectively. If skills acquired in fire-training influenced test performance, it would seem logical that the Untrained group would have a higher failure rate, however this does not appear to be the case. In summary, the very similar overall performance patterns and failure rates would indicate that test performance is not significantly influenced by skills acquired through fire training. The results of this preliminary analysis would seem to indicate that fire training does not provide any apparent advantage to individuals undertaking the Applicant Test protocol. Furthermore, the results suggest that test performance is explained by physical fitness rather than skill. While this result is encouraging, it is strongly recommended that further analysis be undertaken to explore this important question before definitive conclusions are drawn.

167 155 Table Failing test scores for Trained (graduating NFPA 1001) and Untrained (entering NFPA 1001) fire-training students Task Combined Trained Untrained n % n % n % Treadmill % % % Hose Drag % % % Rope Pull % % % Forced Entry % % % Victim Rescue % % % Ladder Climb % % % Equipment Carry % % % Overall % % % Number of Subjects The Effect of Gender on Test Performance There are very few women in the fire service and consequently, it is extremely difficult to address questions of gender bias. Of the 427 applicants tested, 28 (or approximately 7%) were female. This number is consistent with the experience of our laboratory in testing firefighter applicants in Alberta. While there has generally been a trend towards more female applicants over the past 20 years, there is no doubt that firefighting remains an occupation primarily dominated by males. For example, Edmonton Fire-Rescue is a large urban fire department, and approximately 1% of the members are female. Furthermore, relatively few females attend fire-training at the two schools (ESA and Fire ETC) that our laboratory interacts with. Typically, only about 5% of full-time students in fire-

168 156 training classes at these schools are female. The reasons behind the underrepresentation of females in the fire service are no doubt varied and potentially very complex. We can make no attempt to explain why more men than women are inclined to apply for work in firefighting, but offer these comments simply to point out the difficulties in addressing questions of gender and physical fitness. As a preliminary evaluation of this problem, we have examined the performance of the 28 female applicants and the 399 male applicants. It should be emphasized that these data were acquired from males and females that were actually tested as part of job competitions. As such, this analysis provides a view of the type of individual who is inclined to apply for work in the fire service. The reader should refrain from any inference that the data are representative of males and females in general, but rather always bear in mind that the data represent the available perspective on male and female applicants to professional fire services. Descriptive characteristics of the female and male applicants are shown in Tables and No attempt has been made to do statistical comparisons because of the extremely large differences in numbers within the gender groups. However, there appears to be little difference in age, while there are large differences in body size. This observation is borne out by the fact that the heaviest female is slightly lighter than the average male, while the average female is approximately 18 kg lighter than the average male. The tallest female is approximately 5 cm shorter than the average male, while the average female is approximately 11 cm shorter than the average male. The mean VO 2max scores for both genders are consistent with a classification of Good aerobic fitness for young adults (<29 years) based on established fitness norms (McArdle et al., 2001, page 163). However, considering the range in VO 2max scores, the actual classification for any single applicant might range from Poor to Excellent.

169 157 Table Selected characteristics of female firefighter applicants (n=28) Variable Mean SD Minimum Maximum Age (yr) Height (cm) Weight (kg) VO 2max (ml. kg -1. min -1 ) Table Selected characteristics of male firefighter applicants (n=399) Variable Mean SD Minimum Maximum Age (yr) Height (cm) Weight (kg) VO 2max (ml. kg -1. min -1 )

170 158 Table Distribution of passing and failing results on each individual test for female applicants (n=28) Test N Pass Fail DNA* or DNF Treadmill (93%) 2 (7%) 0 (0%) Hose Drag (64%) 7 (25%) 3 (11%) Rope Pull (64%) 6 (21%) 4 (14%) Forcible Entry (79%) 2 (7%) 4 (14%) Victim Drag (79%) 2 (7%) 4 (14%) Ladder Climb (61%) 6 (21%) 5 (18%) Equipment Carry/VE (61%) 4 (14%) 7 (25%) * DNA Did not attempt; DNF Did not finish Table shows the distribution of passing and failing results for each test for the group of 28 female applicants. For each test, the 28 potential results are separated into categories. First, the number of applicants who attempted each test is shown in the second column. Second, the number of passing scores and failing scores is shown in the third and fourth columns, respectively. Finally, the number of applicants who either did not attempt or did not finish each test is shown in the last column. No distinction is made between applicants who did not attempt or did not finish a test. Overall, of the 28 female applicants who

171 159 attempted the Applicant Test, 14 completed all aspects successfully. Therefore, the overall pass rate for this sample of female applicants was 50%. Table Distribution of passing and failing results on each individual test for male applicants (n=399) Test N Pass Fail DNA* or DNF Treadmill (92%) 30 (8%) 0 (0%) Hose Drag (97%) 5 (1%) 7 (2%) Rope Pull (97%) 2 (1%) 9 (2%) Forcible Entry (97%) 1 (1%) 11 (3%) Victim Drag (96%) 3 (1%) 13 (3%) Ladder Climb (93%) 3 (1%) 25 (6%) Equipment Carry/VE (91%) 7 (2%) 30 (8%) * DNA Did not attempt; DNF Did not finish Table shows the distribution of passing and failing results for each test for the group of 399 male applicants. For each test, the 399 potential results are separated into categories. First, the number of applicants who attempted each test is shown in the second column. Second, the number of passing scores and failing scores is shown in the third and fourth columns, respectively. Finally, the number of applicants who either did not attempt or did not finish each test is

172 160 shown in the last column. No distinction is made between applicants who did not attempt or did not finish a test. Overall, of the 399 male applicants who attempted the Applicant Test, 341 completed all aspects successfully. Therefore, the overall pass rate for this sample of male applicants was 85%. In summary, based on the preliminary analysis of the limited data available on female firefighter applicants, there is a higher failure rate among the female applicants. Fifty percent (14 of 28) female applicants failed to complete at least one of the tests at the required standard or alternately, withdrew from the testing protocol. Approximately 15% of male applicants (58 of 399) failed to complete at least one of the tests at the required standard or alternately, withdrew from the testing protocol. These observations suggest that a higher failure rate (or lower passing rate) should be expected among female applicants. Further analysis is essential as more data become available. As pointed out above, there are very substantial differences in body size between the male and female applicants studied, and this may begin to explain the majority of the difference in the failure rates. There appears to be no difference in failure rate on the treadmill test, however the failure rate on each of the jobrelated tests is higher for the female applicants. Each of the job-related tests requires that heavy objects be moved. Logically, larger individuals may be more successful than smaller individuals at such tasks. However, it should also be pointed out that 50% of the female applicants were successful at completing all tests at the required standard. Figures to display the distribution of test scores for the female applicants. These figures clearly show that female applicants are capable of performing these tasks at a high level of performance.

173 Count Treadmill Time (min) Figure Frequency distribution of treadmill test scores for female applicants (n=28). The minimum standard required to pass the test is 13 min. Twenty-six (93%) of the female applicants who completed the test passed and 2 (7%) failed to meet the minimum standard. Count Hose Drag (s) Figure Frequency distribution of Hose Drag scores for female applicants (n=25). The minimum standard required to pass the test is 31 s. Eighteen (72%) of the female applicants who completed this test passed and 7 (28%) failed to meet the minimum standard.

174 162 Count Rope Pull (s) Figure Frequency distribution of Rope Pull scores for female applicants (n=24). The minimum standard required to pass the test is 103 s. Eighteen (75%) of the female applicants who completed this test passed and 6 (25%) failed to meet the minimum standard Count Forcible Entry (s) Figure Frequency distribution of Forcible Entry scores for female applicants (n=24). The minimum standard required to pass the test is 45 s. Twenty-two (92%) of the female applicants who completed this test passed and 2 (8%) failed to meet the minimum standard.

175 163 Count Victim Rescue (s) Figure Frequency distribution of Victim Rescue scores for female applicants (n=24). The minimum standard required to pass the test is 49 s. Twenty-two (92%) of the female applicants who completed this test passed and 2 (8%) failed to meet the minimum standard. Count Ladder Climb (s) Figure Frequency distribution of Ladder Climb scores for female applicants (n=23). The minimum standard required to pass the test is 109 s. Seventeen (74%) of the female applicants who completed this test passed and 6 (26%) failed to meet the minimum standard.

176 Count Equipment Carry/VE (s) Figure Frequency distribution of Equipment Carry/Vehicle Extrication test scores for female applicants (n=22). The minimum standard required to pass the test is 270 s. Seventeen (77%) of the female applicants that completed this test passed and 5 (23%) failed to meet the minimum standard. In summary, the substantial difference in failure rates between male and female applicants must be acknowledged. Whether the difference can be attributed to gender per se cannot be determined at this time. We suggest that a large portion of the differences in performance can be explained by the difference in size. The salient physical and physiological differences between genders that influence work capacity in physically demanding occupations have been reviewed by Shephard and Bonneau (2002). These authors comment on expected differences in size among male and female police officers, referring to data reported by Rhodes and Farenholtz (1992). The mean height and mass for male and female officers in that study were 181 and 169 cm and 86 and 66 kg, respectively. The fitness scores on both job-related and laboratory tests were substantially better for the males, and the females had a significantly higher fail rate on the job-related test. The height and mass values for the police officers are very similar to the data for our male and female firefighter applicants (Tables and 4.3-2).

177 165 Sothmann et al., (2004) reported performance times for groups of male (n=138) and female (n=15) firefighters completing a simulated fire suppression evolution. The evolution was completed with fire protective equipment and involved a significant amount of work against heavy absolute loads (e.g., 75 kg dummy drag) On average, the women completed the evolution 35% slower than their male counterparts (p<0.01). A level of acceptable performance was determined based in input from a panel of experienced firefighters (n=41). While specific data on males and females not meeting this level of acceptable performance were not reported, one can infer from the available information that the failure rate was greater in the female group. Height, mass and fitness values were not reported for any of the subjects. The results provide further evidence that on average, even professional female firefighters do not perform as effectively as their male counterparts on a work simulation involving heavy absolute loads. In two studies presented earlier in this report, male and female subjects completed either fire-training scenarios (Section 2.3) or the CF/DND FF Test (Section 2.4). Both firefighting simulations required working against absolute loads while dressed in fire protective equipment. On average, the female subjects completed the fire-training scenarios approximately 26% slower than the male subjects. On average, the female subjects completed the FF Test approximately 22% slower than the male subjects. The height and mass values for these two groups of subjects reported in Tables and are very similar to the values from Rhodes and Farenholtz (1992) for police officers and also are very similar to the average values for our male and female firefighter applicants (Tables and 4.3-2). These results suggest that performance differences between male and female applicants should be expected. Shephard and Bonneau (2002) state that: A variety of biological differences predispose to discrepancies of functional capacity between women and men: height, body mass, body fat content, haemoglobin level, cardiac dimensions, muscle dimensions,

178 166 muscle composition, and vascular impedence. Nevertheless, the female disadvantage can be largely overcome by rigorous training (p. 265) In order to ensure the highest possible chances of success for female applicants who are likely to be smaller, on average, than their male counterparts, every effort should be made to provide complete information on the tests and high quality training guidelines for optimal preparation of all applicants. The analysis of male and female performance on the Applicant Test is very preliminary, however it is our opinion that the trends observed in these data are probably representative. Clearly, it is essential to continue to analyze gender differences as more data become available. Because of the numerous physical and physiological differences between males and females (Shephard and Bonneau, 2002), it is absolutely essential to place appropriate emphasis on the requirement for rigorous training in preparation for both the test and the job. 4.4 The Effect of Age on Test Performance Thirty-four (9%) of the 399 male applicants were 35 years or older. The demographic and performance characteristics of the older group were compared with 34 of the youngest applicants in the male group. These comparisons are shown in Table By design, there was a significant difference in age between the two groups. The youngest member of the older group was 14 years older than the oldest member of the younger group, while the difference between the group means was approximately 18.5 years. Many of the applicants to the fire service that present at our laboratory for fitness evaluation have some background in the fire service, and as shown in Table above, not surprisingly, the older group averaged approximately 4.5 more years of experience than their younger counterparts. With the exception of these characteristics, there were no other significant differences between the groups. The data shown in Table clearly show that the mean and range o f scores on each performance test were very similar.

179 167 Five of the younger group and three of the older group failed to complete the Applicant protocol successfully. Based on analysis of the limited data available, there does n ot appear to be any evidence to sug gest that older applicants are disadvantaged by the Applicant protocol by virtue of thei r age. Table Physical and performance characterist ics of older (n=34) and you nger (n=34) male firefighter applicants. Variable Younger (n=34) Range Older (n=34) Range Age (yr) * (0.85) (3.7) Height (cm) (4.8) (6.3) Body Mass (kg) (8.8) (11.0) Experience (yr) * 0-23 (1.0) (6.7) VO 2max (ml kg min ) (5.4) (5.8) Treadmill Time (min) (3.0) (2.9) Hose Drag (s) (3.7) (3.7) Rope Pull (s) (11.0) (10.6) Forcible Entry (s) (5.0) (4.3) Victim Drag (s) (6.0) (6.1) Ladder Climb (s) (9.8) (9.7) Equipment Carry/VE (s) (19) (26.3) * indicates a significant difference (p<0.05) between group means

180 Summary The preliminary evaluations of the Applicant Test described in this section have shown that performance does not appear to be affected by age or fire-training experience. Consequently, it appears that the test outcomes are not determined by skills acquired in fire-training, and that older applicants are not disadvantaged by their age. Furthermore, while the failure rate in the group of female applicants is substantially higher than in the group of male subjects, we believe that much of the difference can be explained by the size difference between the male and female applicants. There is good evidence to show that some females can perform at a very high level on this test, and consequently, differences in performance are most likely determined by size and training status rather than gender per se. The strength of the conclusions drawn from these analyses is limited by the small numbers of subjects. On the other hand, it is important to remember that these test data were derived from real individuals that in fact, are seriously interested in career firefighting. Consequently, the results have more meaning than might be inferred from a contrived experiment. While we recognize the limitations of the small sample sizes, the Applicant test appears to be relatively free of bias on the grounds of age, experience or gender per se. It is important to recognize the differences in performance between males and females are largely due to physical and physiological differences that can be significantly influenced by physical training.

181 169 SECTION 5 RECOMMENDATIONS 5.1 Introduction The overall goal of this research project was to develop a valid physical fitness testing program for firefighter applicants. When developing the tests and the standards, a significant effort was made to match the physical requirements of the Applicant Tests with the physical requirements of the CF/DND FF Test. Consequently, it is our belief that the physical standard for applicants to the CF/DND fire service is consistent with the physical standard for incumbents. Notwithstanding the efforts to maintain consistency between applicant and incumbent physical standards within the CF/DND fire service, we also recognize the very significant importance of maintaining consistency between the physical standards for applicants and the demands of firefighting per se. To this end, we have relied heavily on input from the scientific literature and other fire jurisdictions outside CF/DND. The final product can be used in concert with the CF/DND FF Test, or alternately, could be used independently as a valid method of evaluating the physical suitability of applicants to the fire service. 5.2 Recommendations We make the following recommendations regarding adoption, implementation and evaluation of the Applicant Test The Applicant Test should be adopted as a complete package. It would be inappropriate to alter the testing protocol without further research. For example, adding or deleting tests, changing the order of the tests, the rest periods, the test equipment or clothing ensemble could potentially affect the validity of the standards The Applicant Test was designed to be administered using essentially the same equipment and facilities as required for the FF Test. Therefore, in theory, the Applicant Test could be administered at most Canadian Forces bases.

182 The Applicant Test was designed to be used as a SCREENING tool as well as a SELECTION tool. If, after appropriate consideration by CFFM and CFPSA, the test is implemented for the purpose of screening applicants, then the minimum standard for each test defines the level of performance required to pass each individual test. An applicant who meets or exceeds the minimum required level of performance on all tests should obtain an overall PASS Individuals who pass the Applicant Test should be considered physically fit to undertake either fire training or probationary employment. In either case, there should be further follow-up to regulate safe and effective work. It is important to remember that physical fitness is not stable and is subject to change over time Individuals who fail to meet the standards outlined in this report should not be considered to be physically fit to undertake either training or probationary employment as firefighters. The minimum standards are based on our best understanding of the minimum requirements of firefighting Individuals who fail to meet the minimum standards should be fully debriefed and could be offered remedial physical training programs, followed by opportunities for retesting Opportunities for retesting should normally be allowed after an appropriate period of time that allows for physical training (e.g., three months) If so desired, the Applicant Test may also be used as a SELECTION tool. Such use will depend on organizational decisions on whether or not it is feasible or desirable to rank applicants on physical fitness. If applicants are ranked on physical fitness, then the overall score (total points) could be used in conjunction with other attributes (e.g., aptitude test scores, interview results) to make decisions on selecting the most qualified applicants for the CF/DND fire service.

183 If the test is to be used for SELECTION purposes, it is recommended that VO 2max be measured during the treadmill test. This may limit the locations where the test can be correctly administered CFFM and CFPSA should develop an implementation plan for the Applicant Test. This plan should address matters concerning test locations (e.g., all CF/DND fire halls, selected CF bases, outside contractors such as University laboratories) The implementation plan should also address matters related to ensuring validity and reliability of the test protocol between test sites The University of Alberta should be consulted regarding the qualifications and training of individuals involved in administering the protocol A comprehensive information package should be prepared and distributed to all prospective applicants well in advance of the testing date. The package should provide adequate information about the test procedures and standards so that applicants can prepare effectively for the tests. A sample of the Information Package used by the University of Alberta is shown in Appendix A It is recommended that information on effective training be provided to prospective applicants The University of Alberta should be consulted regarding preparation of all test materials and information for applicants A process for medical clearance of applicants prior to testing is strongly recommended. The requirement for medical clearance of applicants may depend on whether other medical information is available that may substitute for the type of clearance provided by the Medical Clearance Form used by the University of Alberta (see Appendix B) It is strongly recommended that a plan be developed to evaluate the suitability of the Applicant Test for the CF/DND fire service. While the Applicant Test has been used extensively at the University of Alberta for a number of other fire service jurisdictions, it is essential that an evaluation process be considered to ensure that the test meets the needs of DND.

184 It is strongly recommended that every effort be made to continue the evaluation of the Applicant Test with respect to bias on the grounds of age, gender and fire-related skills. The results of the preliminary analysis presented in Section 4 must be followed-up with ongoing analysis as data become available.

185 173 SECTION 6 REFERENCES American College of Sports Medicine. (2006). ACSM Guidelines for Exercise Testing and Prescription (7 th ed.). Lippincott Williams and Wilkins: New York. Astrand, P.O. and Rodahl, K. (1986) Textbook of Work Physiology (3 rd Edition). McGraw-Hill Book Company: New York. Bilzon, J.L., Scarpello, E.G., Smith, C.V., Ravenhill, N.A., Rayson, M.P. (2001) Characterization of the metabolic demands of simulated shipboard Royal Navy fire-fighting tasks. Ergonomics. 44: Borg, G.A. (1982). Psychological bases of perceived exertion. Med. Sci. Sports Exerc.14: Canadian Society for Exercise Physiology (1986). Canadian Standardized Test of Fitness (CSTF) 3 rd Edition. Fitness Canada. Davis, P. O., Dotson, C. O., and Santa Maria, D. L. (1982). Relationship between simulated fire fighting tasks and physical performance measures. Med. Sci. Sports Exerc. 14: Davis, S.C., Jankovitz, K.Z., Rein, S. (2002) Physical fitness and cardiac risk factors of professional firefighters across the career span. Res Q Exercise and Sport. 73: Deakin, J. M., Pelot, R. P., Smith, J. M., Stevenson, J. M., Wolfe, L. A., Lee, S. W., Jaenen, S. P., Hughes, S. A., Dwyer, J. W., and Hayes, A. D. (1996). Development of a bona fide physical maintenance standard for CF and DND fire fighters: 119. Kingston, Ontario: Queen's University. Dreger, RW and Petersen, SR. (2003a) Measurement of VO 2max for firefighters: a new approach. Can. J. Appl. Physiol. 28(Suppl): S48. Dreger, RW, Jones, RL, and Petersen, SR. (in press) Effects of the selfcontained breathing apparatus and fire protective clothing on maximal oxygen uptake. Ergonomics Duggan, A. (1988). Energy cost of stepping in protective clothing ensembles. Ergonomics. 31: 3-11.

186 174 Eves, N.D., Petersen, S.R., and Jones, R.L. (2002a). The effect of hyperoxia on submaximal exercise with the self-contained breathing apparatus. Ergonomics 45: Eves, N.D., Petersen, S.R., and Jones, R.L. (2002b). Hyperoxia improves maximal exercise with the self-contained breathing apparatus (SCBA). Ergonomics 45: Eves, N.D., Petersen, S.R., and Jones, R.L. (2003a). Effects of helium and 40% O 2 on graded exercise with self-contained breathing apparatus. Can. J. Appl. Physiol. 28: Eves, N.D., Petersen, S.R., and Jones, R.L. (2003b). Submaximal exercise with self-contained breathing apparatus: The effects of hyperoxia and inspired gas density. Aviat. Space Environ. Med. 74: Eves, N.D, Jones, R.L., and Petersen, S.R. (2005). The influence of selfcontained breathing apparatus (SCBA) on ventilatory function and maximal exercise.can. J. Appl. Physiol. 30: Fox, E.L. and Mathews, D.K. (1981). The physiological basis of physical education and athletics (3 rd ed.). Saunders College Publishing: Philadelphia. Gerkin, D. (1995). Firefighters: fitness for duty. Occup. Med. 10: Gledhill, N., and Jamnik, V. K. (1992a). Characterization of the physical demands of firefighting. Can. J. Sport Sci. 17: Gledhill, N., and Jamnik, V. K. (1992b). Development and validation of a fitness screening protocol for firefighter applicants. Can. J. Sport Sci. 17: Green, J. S., and Crouse, S. F. (1991). Mandatory exercise and heart disease risk in fire fighters. Int. J. Occup. Environ. Health. 63: Guidotti, T. L., and Clough, V. M. (1992). Occupational health concerns of firefighting. Annu. Rev. Public Health. 13: Ilmarinen, J. (1992a). Job design for the aged with regard to decline in their maximal aerobic capacity: Part I - Guidelines for the practitioner. Int. J. Ind. Ergonomics. 10: Ilmarinen, J. (1992b). Job design for the aged with regard to decline in their maximal aerobic capacity: Part II - The scientific basis for the guide. Int. J. Ind. Ergonomics. 10:

187 175 International Association of Fire Fighters. (1997) Fire Service Joint Labor Management Wellness Fitness Initiative. Washington, D.C.: IAFF. Kilbom, A. (1980). Physical work capacity of firemen. With special reference to demands during fire fighting. Scand. J. Work. Environ. Health. 6: Lemon, P. W., and Hermiston, R. T. (1977). The human energy cost of fire fighting. J. Occup. Med. 19: Louhevaara, V., Smolander, J., Tuomi, T., Korhonen, O., and Jaakkola, J. (1985). Effects of an SCBA on breathing pattern, gas exchange, and heart rate during exercise. J. Occup. Med. 27: Louhevaara, V., Ilmarinen, R., Griefahn, B., Kunemund, C., and Makinen, H. (1995). Maximal physical work performance with European standard based fireprotective clothing system and equipment in relation to individual characteristics. Eur. J. Appl. Physiol. 71: Louhevaara, V., Smolander, J., Korhonen, O., and Tuomi, T. (1986a). Effects of industrial respirators on breathing pattern at different work levels. Eur. J. Appl. Physiol. 55: Louhevaara, V., Smolander, J., Korhonen, O., and Tuomi, T. (1986b). Maximal working times with a self-contained breathing apparatus. Ergonomics. 29: Lusa, S., Louhevaara, V., and Kinnunen, K. (1994). Are the job demands on physical work capacity equal for young and aging firefighters? J. Occup. Med. 36: Lusa, S., Louhevaara, V., Smolander, J., Kivimaki, M., and Korhonen, O. (1993). Physiological responses of firefighting students during simulated smoke-diving in the heat. Am. Ind. Hyg. Assoc. J. 54: Manning, J. E., and Griggs, T. R. (1983). Heart rates in fire fighters using light and heavy breathing equipment: similar near-maximal exertion in response to multiple work load conditions. J. Occup. Med. 25: McArdle, W.D. Katch, F.I., Katch, V.L. (2001) Exercise Physiology (5 th Edition) Lippincott Williams and Wilkins: Philadelphia. Misner, J. E., Plowman, S. A., and Boileau, R. A. (1987). Performance differences between males and females on simulated firefighting tasks. J. Occup. Med. 29:

188 176 Myhre, L. G., Tucker, D. M., Bauer, D. H., Fischer, J. R., Grimm, W. H., Tattersfield, C. R., and Wells, W. T. (1997). Relationship Between Selected Measures Of Physical Fitness And Performance Of A Simulated Firefighting Emergency Task. Brooks AFB, Texas: Armstrong Laboratory Systems Research Branch. Report number AS/CF-TR NFPA. (1997). NFPA 1001, Standard for Fire Fighter Professional Qualifications. Quincy, MA: National Fire Protection Association. NFPA. (2001). NFPA 1710, Standard for the Organization and Deployment of Fire Suppression Operations, Emergency Medical Operations, and Special Operations to the Public by Career Fire Departments. Quincy, MA: National Fire Protection Association. NFPA. (2003). NFPA 1582, Standard on Comprehensive Occupational Medical Program for Fire Departments. Quincy, MA: National Fire Protection Association. O'Connell, E. R., Thomas, P. C., Cady, L. D., and Karwasky, R. J. (1986). Energy costs of simulated stair climbing as a job-related task in fire fighting. J. Occup. Med. 28: Peate, W. F., Lundergan, L. L. and Johnson, J. J. (2002). Fitness self-perception and VO 2max in firefighters. J. Occup. Environ. Med. 6: Pelot, R. P., Dwyer, J. W., Deakin, J. M., and McCabe, J. G. (1999). The design of a simulated forcible entry test for fire fighters. Appl. Ergonomics 30: Petersen, S. R., Dreger, R. W., Williams, B. E., and McGarvey, W. J. (2000). The effects of hyperoxia on performance during simulated firefighting work. Ergonomics. 43: Petersen, S.R and Dreger, R.W. (2004) Aerobic demands of fire rescue work in males and females. Can J. Appl. Physiol. 29 (Suppl): S72. Saupe, K., Sothmann, M., and Jasenof, D. (1991). Aging and the fitness of fire fighters: the complex issues involved in abolishing mandatory retirement ages. Am. J. Public Health. 81: Shephard, R. J. and Bonneau, J. (2002) Assuring gender equity in recruit standards for police officers. Can. J. Appl. Physiol. 27: Smith, D. L., and Petruzzello, S. J. (1998). Selected physiological and psychological responses to live-fire drills in different configurations of firefighting gear. Ergonomics. 41:

189 177 Sothmann, M., Saupe, K., Jansenof, D., Blaney, J., Fuhrman, S. D., Woulfe, T., Raven, P., Pawelczyk, J., Dotson, C., Landy, F., Smith, J. J., and Davis, P. (1990). Advancing age and the cardiorespiratory stress of fire suppression: determining a minimum standard for aerobic fitness. Human Performance. 3: Sothmann, M., Saupe, K., Raven, P., Pawelczyk, J., Davis, P., Dotson, C., Landy, F., and Siliunas, M. (1991). Oxygen consumption during fire suppression: error of heart rate estimation. Ergonomics. 34: Sothmann, M.S., Saupe, K., Jasenof, M., and Blaney, J. (1992a). Heart rate responses of fire-fighters to actual emergencies. J. Occup. Med. 34: Sothmann, M. S., Landy, F., and Saupe, K. (1992b). Age as a bona fide occupational qualification for firefighting. A review on the importance of measuring aerobic power. J. Occup. Med. 34: Sothmann, M.S., Gebhardt, D.L., Baker, T.A., Kastello, G.M., and Sheppard, V.A., (2004) Performance requirements of physically strenuous occupations: validating minimum standards for muscular strength and endurance. Ergonomics 47: Sutton, J.R. (1992) VO 2max new concepts on an old theme. Med. Sci. Sports and Exerc. 24: Teitlebaum, A. & Goldman, R. (1972). Increased energy cost with multiple clothing layers. J. Appl. Physiol. 32: U.S. Department of Labor (1993). Selected Characteristics of Occupations Defined in the Revised Dictionary of Occupational Titles. U.S. Dept. of Labor Employment and Training Administration. Washington, DC: U.S. Government Printing Office. White, M. K., Hodous, T. K., and Vercruyssen, M. (1991). Effects of thermal environment and chemical protective clothing on work tolerance, physiological responses, and subjective ratings. Ergonomics. 34: Williford, H. N., Duey, W. J., Olson, M. S., Howard, R., and Wang, N. (1999). Relationship between fire fighting suppression tasks and physical fitness. Ergonomics. 42:

190 178 Appendix A Sample Information Package for Applicants (previously used with the City of Lethbridge Fire and Emergency Services)

191 179 WORK PHYSIOLOGY LABORATORY FACULTY OF PHYSICAL EDUCATION AND RECREATION UNIVERSITY OF ALBERTA FIREFIGHTER APPLICANT PHYSICAL FITNESS EVALUATION APPOINTMENT CHECKLIST The information package provides a detailed overview of the physical evaluation process. More information can be obtained from Lethbridge Fire and Emergency Services by telephone at If you have any questions about the testing procedure, you can contact Tim Hartley at the Work Physiology Laboratory by telephone at or by at firetest@ualberta.ca. You must confirm your appointment by sending in Page 9 of the Information Package and a money order (payable to University of Alberta) for $ Complete the Checklist! To do: Action: Done: Complete page 9 Attach money order for $ Arrange doctor s appointment to complete Medical Clearance Form (pages 10-11) Read the Information Package carefully Mail to Lethbridge Fire and Emergency Services to arrive at least 5 days before your appointment Bring the completed Medical Clearance Form with you on your test day Follow the instructions to prepare as well as possible Questions? Problems? us for help

192 180 FACULTY OF PHYSICAL EDUCATION AND RECREATION UNIVERSITY OF ALBERTA FIREFIGHTER APPLICANT PHYSICAL FITNESS EVALUATION INFORMATION PACKAGE The physical evaluation program is administered by the Faculty of Physical Education and Recreation at the University of Alberta. Please read the following information carefully in order to prepare for the tests. GENERAL INFORMATION The tests will be completed at the University of Alberta in Edmonton. Your appointment time will be assigned to you by Lethbridge Fire and Emergency Services. The official results will be returned to Lethbridge as soon as possible after all applicants have been tested. We will give you a personal copy of your results after your test. The testing program runs on a strict schedule, so you must be on time. If you are not familiar with the University of Alberta campus, please allow yourself a little extra time since the nearest parking is at least a block away from the test location. You should expect to be at the University for about 3-4 hours on your test day. This amount of time permit s adequate rest between tests so that you will be able to perform as well as possible. TESTING LOCATION The tests are conducted at the Work Physiology Laboratory in the Universiade Pavilion (usually referred to as the "Butterdome") at the University of Alberta. This large yellow building is located on the northwest corner of the intersection of 87 Avenue and 114 Street (see the enclosed campus map on page 8). The main entrance is on the East side of the building. Enter the building here, and you will see a sign directing you to the Work Physiology Laboratory. There are male and female locker rooms on the lower level where you may change and shower. You should bring your own towel and a lock with you to put on a "day-use" locker, since we have no provision for securing your valuables.

193 181 Pay parking is available at the Education Carpark and the Jubilee Auditorium lot which are on the Northeast and Southwest corners of the 87 Avenue and 114 Street intersection respectively. Additional parking is available at Stadium Carpark, located on the North side of the Physical Education complex near the Student Union Building. For more information on building and parking locations, refer to the map of the University of Alberta Campus (page 8 of this package). COST OF TESTING The fee for the testing service is $ (including GST). You must pay the entire fee at least 5 days in advance of your appontment. You must pay your fee by money order. Personal cheques or credit card payments will not be accepted. Please make your money order payable to the University of Alberta. CONFIRMING YOUR APPOINTMENT Lethbridge Fire and Emergency Services will inform you of the date and time of your testing appointment. You must confirm your appointment by mailing the Appointment Confirmation Form (page 9 of this package) to Lethbridge, along with payment of your testing fee of $ Your appointment will not be confirmed until the payment is received. Lethbridge Fire and Emergency Services will collect the deposit and send it to the University of Alberta. After you have received your appointment time for physical testing from Lethbridge Fire and Emergency Services, complete and mail the Appointment Confirmation Form (page 9) to: Lethbridge Fire and Emergency Services Avenue North Lethbridge, Alberta T1H 0P2 You must include the $ testing fee to secure your test appointment. Your deposit must arrive at Lethbridge Fire and Emergency Services at least 5 days before your test day. MEDICAL CLEARANCE FOR TESTING The tests are very demanding and are designed to assess the physical capabilities of healthy individuals. In order to be tested, you must have a physician certify that you are medically fit to undertake the tests. The Medical Clearance for Testing form (pages 10 and 11 of this package) must be signed by your physician and you must bring it with you when you come to the University for testing. You will not be permitted to complete the tests unless this form has been completed by your physician.

194 182 DESCRIPTION OF THE PHYSICAL FITNESS TESTS This program is designed to evaluate the physical work capacities of healthy, physically active individuals. Each test requires a maximal effort. All of the tests are completed while wearing firefighting personal protective equipment (PPE) that weighs approximately 22 kg (50 lb). This ensemble includes: helmet, flashhood, gloves, pants, boots, jacket and self-contained breathing apparatus (SCBA). You will not breathe from the SCBA, but you must carry it. For safety during the treadmill test, running shoes are substituted for firefighting boots. After completing the treadmill test, you will rest for 60 minutes before starting an orientation to the job-related performance tests. The orientation to the jobrelated tests consists of a walk-through session to practice each of the tasks. This will take approximately 30 minutes and will familiarize you with testing procedures and provides a suitable warm-up for the demanding tests that follow. Each test is followed by a rest period of 3 minutes for recovery and hydration. You are not permitted to leave the testing area or remove the PPE during the rest periods. The tests are described briefly in the following sections: 1. Aerobic Endurance Maximum oxygen uptake (VO 2max ) will be measured during a progressive, incremental exercise test to exhaustion on a treadmill. After a standardized 5- minute warm-up, you will walk at 3.5 mph and 10% grade for 8 minutes. After this phase is completed, the grade (and if necessary, speed) will be increased every minute until you are too tired to continue. Depending on your fitness level and motivation, this test lasts between minutes. During the test, expired gases are monitored with an automated metabolic measurement system to calculate the rate of oxygen consumption. Heart rate is monitored continuously with a telemetry system. 2. Charged Hose Advance Test You will drag a charged (full of water) 38 mm (1.5 inch) hose a distance of 45 m (125 ). Three 15 m (50') lengths of hose are "snaked" behind the starting line. The nozzle is held over the shoulder and you advance to the finish line as quickly as possible (running is not permitted). This test assesses lower body strength and power and must be completed safely in less than 31 sec.

195 High Volume Hose Pull Test You will pull a bundle of hose weighing approximately 56 kg (123 lb) a distance of 15 m (50 ) over a smooth concrete floor using a rope. This task is repeated 3 times. During this test, you must stand still and pull the hose bundle towards you using 16 mm (5/8") rope. This test assesses upper body strength, power, and endurance and must be completed safely in less than 103 sec. 4. Forcible Entry Simulation Test You will use a 3.6 kg (8 lb) "dead blow" sledge hammer to move a weighted truck tire (102 kg or 225 lb) a distance of 30.5 cm (12 ) as rapidly as possible. This test assesses muscle strength, power and endurance, particularly in the upper body and must be completed safely in less than 45 sec. 5. Victim Drag Test You will drag a mannequin weighing 68.2 kg (150 lb) a total distance of 30 m (100'). The test starts with the mannequin lying "face- up" on the floor. You will lift the mannequin and walk backwards for 15 m, turn around a traffic cone and return to the start line as quickly as possible. This test assesses muscle strength and endurance and must be completed safely in less than 49 sec. 6. Ladder Climb Test You will climb a 7.3 m (24 ) ladder to the 10 th rung and return to the floor as quickly as possible. This task will be repeated five times. This test assesses muscle strength, endurance, and anaerobic capacity and must be completed safely in less than 109 sec.

196 Equipment Carry/Vehicle Extrication Test You will carry small (18 kg or 40 lb) and large (36 kg or 80 lb) vehicle extrication tools (the Jaws of Life ) a total distance of 30 m (100 ). In addition, you will lift and hold the 18 kg tool in specific positions that simulate the work required to remove a vehicle door. This test is designed to evaluate the strength and endurance required to lift, carry and use heavy tools in rescue situations. This test must be completed safely in less than 270 sec. EFFECTIVE PREPARATION FOR THE TESTS In order to do your best, you should come to the laboratory on your testing day well nourished and well rested. You should not do strenuous exercise on the days immediately before your tests. Sleep well the night before and try to be as relaxed as possible. Avoid alcoholic beverages the day before and definitely on the day of your test. Do not smoke or drink beverages with caffeine (tea, coffee, hot chocolate, cola, etc.) for at least two hours prior to your test. Do not eat for at least two hours before your test appointment. However, it is important to be well nourished and well hydrated. The tests are very demanding and most individuals are extremely tired at the end of each test. If your appointment is first thing in the morning, do not skip breakfast. You should eat a light meal (e.g., fruit, toast or cereal, and juice) about three hours before your test. WHAT DO I NEED TO BRING? Clothing Bring the following items of clothing with you: shorts, two T-shirts, running shoes, extra socks, gloves, and sweats. Your T-shirt will be wet from sweat after the treadmill test. You should change into a dry shirt and then put on sweats to keep warm during the 60-minute rest period. We have a good selection of firefighting boots, however getting an exact fit may not always be possible. In order to get the best fit, bring several pairs of socks (thin and thick sport socks).

197 185 Nutrition You should bring a water bottle or sports drink (e.g., Gatorade). You may want to eat a small snack (e.g., banana or Power Bar) during the rest period between the treadmill test and the job-related tests. Be careful to practice in advance so that you know how much to eat and drink during 3+ hours of intermittent, extremely strenuous exercise. If you eat or drink too much you will feel sick and do poorly. If you eat and drink too little, you will get dehydrated and do poorly. Optimal nutrition and hydration strategies tend to be very individual. Work this out for yourself. Don t follow someone else s advice unless you have had the chance to make sure it works for you under the kind of conditions you will experience during these tests. Identification Your drivers' license (with photograph) is required in order to register for the test and verify your identity. Medical Clearance You must bring the Medical Clearance for Testing (pages 10-11) document that has been completed by your physician. This document provides medical clearance for you to undertake the specific tests in this program. You will not be permitted to do any of the tests until the Medical Clearance for Testing document has been signed by your physician. Payment The testing fee must be paid before Lethbridge Fire and Emergency Services will confirm your test appointment. Remember, you may only pay by money order or certified cheque, payable to the University of Alberta. Sorry, no personal cheques or credit card payments. An official receipt will be issued for the full amount. Overnight Accommodation in Edmonton If you are traveling to Edmonton for your test, we suggest the following choices for reasonably priced accommodation at or near the University of Alberta: Campus Tower Hotel: Tower on the Park Hotel: Varscona Hotel: There are a limited number of guest rooms available at U of A Student Housing (residence):

198 UNIVERSITY OF ALBERTA CAMPUS MAP 186

199 187 FACULTY OF PHYSICAL EDUCATION AND RECREATION UNIVERSITY OF ALBERTA FIREFIGHTER APPLICANT PHYSICAL FITNESS EVALUATION CONFIRMATION OF APPOINTMENT WHEN YOU RECEIVE YOUR APPOINTMENT TIME FROM LETHBRIDGE FIRE AND EMERGENCY SERVICES, COMPLETE THE FOLLOWING INFORMATION AND MAIL TO: Lethbridge Fire and Emergency Services Avenue North Lethbridge, Alberta T1H 0P2 Name: Address: (Street Address) (City) (Province) (Postal Code) Telephone: I will attend the physical fitness testing for Lethbridge at the following time: Date Time If you have questions about your appointment, contact Tim Hartley at firetest@ualberta.ca or leave a telephone message for him at Enclose a MONEY ORDER OR CERTIFIED CHEQUE (payable to the University of Alberta). Payment is required to confirm your appointment. This page and testing fee must be received at least 5 days BEFORE your appointment.

200 188 APPENDI X B Sample Medical Clearance for Testing

201 189 Applicant name: FACULTY OF PHYSICAL EDUCATION AND RECREATION UNIVERSITY OF ALBERTA FIREFIGHTER APPLICANT PHYSICAL FITNESS EVALUATION MEDICAL CLEARANCE FOR TESTING This program is designed to evaluate the physical work capacities of healthy, physically active individuals. Each test requires a maximal effort. All of the tests are completed while wearing firefighting personal protective equipment (PPE) that weighs approximately 22 kg (50 lb). This ensemble includes: helmet, flash-hood, gloves, pants, boots, jacket and self-contained breathing apparatus (SCBA). The applicant is not required to breathe from the SCBA, but must carry it. For safety during the treadmill test, running shoes are substituted for firefighting boots. The tests are administered by the Faculty of Physical Education and Recreation at the University of Alberta, and are not medically supervised. The test procedures are described briefly below: AEROBIC ENDURANCE Maximum oxygen uptake (VO2max) will be measured during a progressive, incremental exercise test to exhaustion on a treadmill. During the test, expired gases are monitored with an automated metabolic measurement system to calculate the rate of oxygen consumption. Heart rate is monitored continuously with a telemetry system. Depending on fitness level and motivation, this test normally requires the individual to walk on the treadmill for between minutes. Regardless of the fitness level of the individual, the test normally involves a maximal effort and is terminated when the person is too fatigued to continue exercise. Combined with the exercise stress, the weight and heat retention properties of the PPE result in a significant level of fatigue. After completing the treadmill test, the applicant will rest for 60 minutes before moving on to the job-related performance tests. JOB-RELATED PERFORMANCE TESTS Prior to completing the job-related tests, the applicant will complete a walk-through session where they are allowed to practice each of the tasks. This takes approximately 30 minutes and serves to familiarize the applicant with testing procedures and provides a suitable warm-up for the demanding tests that follow. Each test is followed by a rest period of 3 minutes for recovery and hydration. Applicants are not permitted to leave the testing area or remove the PPE during the rest periods. Charged Hose Advance Test The applicant will drag a charged (full of water) 38 mm (1.5 inch) hose a distance of 45 m (125 ). Three 15 m (50') lengths of hose are "snaked" behind the starting line. The

202 190 nozzle is held over the shoulder and applicant advances to the finish line as quickly as possible. This test assesses lower body strength and anaerobic power. High Volume Hose Pull Test The applicant will pull a bundle of hose weighing approximately 56 kg (123 lb) a distance of 15 m (50 ) over a smooth concrete floor using a rope. This task is repeated 3 times. During this test, the applicant is stationary and must pull the hose bundle towards them using 16 mm (5/8") rope. This test assesses upper body strength, power, and endurance. Forcible Entry Simulation Test Using a 3.6 kg (8 lb) "dead blow" sledge hammer, the applicant moves a weighted truck tire (102 kg or 200 lb) a distance of 30.5 cm (12 ) as rapidly as possible. This test assesses muscle strength, power and endurance, particularly in the upper body. Victim Drag Test The applicant will drag a mannequin weighing 68.2 kg (150 lb) a total distance of 30 m (100'). The test starts with the mannequin lying "face-up" on the floor and the applicant standing. The applicant lifts the mannequin and walks backwards for 15 m, turns around a traffic cone and returns to the start line as quickly as possible. This test assesses strength, power, and agility. Ladder Climb Test The applicant will climb a 7.3 m (24 ) ladder to the 10 th rung and returns to the floor as quickly as possible. This task will be repeated five times. This test assesses muscle strength, endurance, and anaerobic capacity. Equipment Carry/Vehicle Extrication Test The applicant will carry small (18 kg or 40 lb) and large (36 kg or 80 lb) vehicle extrication tools (the Jaws of Life ) a total distance of 30 m (100 ). In addition, the applicant will lift and hold the 18 kg tool in specific positions that simulate the work required to remove a vehicle door. The tools will then be returned to the storage cabinet. This test is designed to evaluate the strength required to lift, carry and use heavy tools in rescue situations. Is this individual taking any medication that could affect normal physiological responses to exercise? No Yes If yes, please explain. Is there any medical reason that this individual should not undertake very strenuous exercise? No Yes If yes, please explain. I certify that this applicant has been given a medical examination and is medically fit to undertake the Physical Fitness Evaluation described above. Physician's name: Address: (or stamp) Date: Telephone: Signature:

203 191 APPENDIX C Sample Informed Consent for Testing

204 192 UNIVERSITY OF ALBERTA FIREFIGHTER APPLICANT PHYSICAL FITNESS EVALUATION INFORMED CONSENT NAME: The tests in this program involve very strenuous exercise and maximal effort. There may be some health risk with this type of exercise. During and after the tests it is possible to experience symptoms such as abnormal blood pressure, fainting, lightheadedness, muscle cramps or strain, nausea, and in very rare cases, heart rhythm disturbances or heart attack. There is also some risk of musculo-skeletal injury from falling or lifting heavy objects during the job-related tests. While serious risk to healthy individuals is unlikely, it is important to acknowledge that you have been informed of these possibilities and willfully assume the risks of participation. You will receive an official copy of your test results when you leave the lab today. A copy of your results will be sent to (Fire Department). Your test results will be maintained in confidence by the University of Alberta. Normally, your file will be kept in secure storage for one year and then will be destroyed. Before each test, full instructions on procedures and safety will be provided. You will also have the opportunity to practice and warm-up exercises before the tests. You may ask questions on test procedures at any time. The tests are described briefly below: 1. Aerobic Endurance Maximum oxygen uptake (VO 2max ) will be measured during a progressive, incremental exercise test to exhaustion on a treadmill. After a standard warm-up, you will walk at 3.5 mph and 10% grade for 8 minutes. After this phase is completed, the grade (and if necessary, speed) will be increased every minute until you are too tired to continue. Depending on your fitness level and motivation, the actual test lasts between minutes. During the test, expired gases are monitored with an automated metabolic measurement system to calculate the rate of oxygen consumption. Heart rate is monitored continuously with a telemetry system. 2. Charged Hose Advance Test You will drag a charged (full of water) 38 mm (1.5 inch) hose a distance of 45 m (125 ). Three 15 m (50') lengths of hose are "snaked" behind the starting line. The nozzle is held over the shoulder and you advance to the finish line as quickly as possible (running is not permitted). This test assesses lower body strength and power.