Journal Logo


Physiological Demands of the Firefighter Candidate Physical Ability Test


Author Information
Medicine & Science in Sports & Exercise: March 2009 - Volume 41 - Issue 3 - p 653-662
doi: 10.1249/MSS.0b013e31818ad117
  • Free


Preemployment screening of firefighter candidates to select individuals capable of meeting the high physical demands of the profession has been supported in judgments by courts of law and human rights tribunals. Various means to screen suitable candidates have been suggested during the years using a "construct" approach of testing individual components of physical fitness as well as a "content" approach in which typical tasks are incorporated in a circuit that simulates the demands of firefighting (2). Recently, the International Association of Fire Chiefs (IAFC) and the International Association of Fire Fighters (IAFF) have collaborated in the development of the Candidate Physical Ability Test (CPAT), which is a content-based circuit of activities designed for screening of potential firefighter candidate recruits (11).

The basis for the CPAT is well established in an extensive literature on the physical demands of firefighting. During actual and simulated firefighting tasks, researchers have measured high heart rate (HR) (1,3,17,21). Further during simulated tasks of firefighting, direct measurement of oxygen uptake (V˙O2) or HR (4,14,19,21,22) as well as elevated posttask blood lactate concentrations (4,14,19,21,22) have confirmed the strenuous nature of the tasks. On the basis of these measurements, previous researchers have proposed that there should be lower limits for maximal oxygen uptake (V˙O2max) for firefighters such as 33.5 to 42.0 mL·kg−1·min−1 (22), 40 mL·kg−1·min−1 (14), 42 mL·kg−1·min−1 (1), 45 mL·kg−1·min−1 (4), 2.8 to 3.0 L·min−1 (12), 3.5 L·min−1 (15), and 4.0 L·min−1 (25).

The CPAT does not incorporate the direct measurement of V˙O2max. Rather, it uses an exclusively content-based approach to testing and bases its validity on the evaluation by experts of videotapes of the test being performed at different speeds by incumbent male and female firefighters (11). The CPAT is accepted as a bona fide occupational qualification or requirement (BFOQ or BFOR). To date, the energy demands of the CPAT and its relationship to other standard laboratory measures of physical fitness have not been evaluated. It was the purpose of this study to measure the energy requirements of the CPAT in subjects typical of candidate recruits and to compare the performance time with other standard laboratory measurements of fitness. It was hypothesized that direct measurement of respiratory gas exchange would reveal not only the high aerobic demands of the tasks but also an important anaerobic component. Further, we anticipated that individuals capable of completing the CPAT within the criterion time would possess both high aerobic fitness and high muscle strength.


Subjects and experimental design

Fifty-seven (23 women) healthy, physically active subjects, with age ranging from 19 to 46 yr (mean 23.7 ± 4.6 yr), volunteered for this study (Table 1). This research was approved by the Office of Research Ethics at the University of Waterloo. Full written and verbal details were provided, and informed written consent was obtained before participating in the study.

Group means for selected anthropometric measurements (age, height, body mass) and physiological variables at peak exercise during the treadmill test to exhaustion (E peak, V˙O2max, V˙CO2, RER, HRmax) for those who completed irrespective of time (a) and those who did not complete (b) the CPAT.

To simulate as closely as possible the actual conditions of the CPAT for candidate recruits, all subjects were given the instructions and opportunities for practice similar to those prescribed by the IAFF/IAFC Wellness-Fitness Task Force. An exception to this was that testing occurred within 2-3 wk of initial familiarization rather than the 8 wk that could be used for specific training and practice timed trials (although candidates can opt of this longer period); however, each subject could practice events until he or she was comfortable with the demands. All testing was conducted on a certified course at an accredited testing facility (24). The tests for data acquisition were obtained on two separate days separated by at least 48 h. On one test day, subjects completed the CPAT, and on the other, they completed fitness testing. Tests of V˙O2max during maximal treadmill running, muscle strength and endurance, and Wingate anaerobic tests were conducted (in that order) on the same day with a minimum of 20-min rest breaks between testing procedures.

Materials and methods

Ventilation (E) and gas exchange (V˙O2, carbon dioxide output, V˙CO2, and RER) were measured breath-by-breath during the maximal treadmill tests and the CPAT with the Cosmed K4b2 portable metabolic system (Cosmed, Italy). Before each test, the O2 and CO2 gas analyzers were calibrated according to the manufacturer's specifications using room air and a precision-analyzed gas mixture (approximately 5% CO2, 16% O2, balance N2). The volume turbine was calibrated using a 3.0-L hand-pumped syringe with flow rates approximating those during heavy exercise. HR was recorded during the tests using the Polar monitoring system incorporated with the portable unit.

An incremental exercise test was conducted on a motorized treadmill (Quinton, WA) to determine V˙O2max. After a 4-min warm-up at a brisk walking pace, speed was increased 1.6 km·h−1 every 2 min until reaching a comfortable running speed, followed by 2% increases in grade every 2 min thereafter. The test was terminated when subjects were unable to continue and reached volitional fatigue. V˙O2max was taken as the highest 20-s average during the final minute of the test.

Muscular strength measures were obtained using a predictive one-repetition maximum (1-RM) formula, as previously described (13):

Before each of the predicted 1-RM strength tests, participants completed a five-repetition warm-up using ∼40% of their 1-RM (13). After a 1-min rest period, a load was determined, which would fatigue the participant with fewer than 10 repetitions. After a predictive 1-RM test, a 3- to 5-min rest period was implemented before proceeding to the next muscle group. The muscular strength tests that were used (in order of testing protocol) were maximal handgrip using a hand dynamometer (Takei Co. Ltd, Tokyo, Japan), flat bench press using a 20-kg Olympic bar, seated 45° incline leg press, military shoulder press using a seated weight machine, and seated bicep curls using a 7-kg bent curl bar with elbows placed on a preacher stand.

Muscular endurance was evaluated for upper and lower body using the flat bench press and seated 45° incline leg press. For the upper body endurance, an absolute load of 30 kg was raised and lowered at 30 repetitions per min. For lower body endurance, subjects lifted and lowered an absolute load of 123 kg at a cadence of 50 repetitions per min. For both muscular endurance tests, subjects were required to lift the load until no further repetitions could be completed or until there was an inability to maintain cadence.


The CPAT was conducted in accordance with the actual administration of the test for firefighter candidate recruits with the exception of wearing the Cosmed K4b2 throughout the test. Subjects wore a 22.68-kg vest to simulate the weight of the self-contained breathing apparatus and firefighter protective clothing ensemble as well as safety gloves and protective hard helmet. The CPAT protocol consists of eight events in a continuous circuit with a 22.9-m walk between each of the events (11). The pass/fail cutoff criterion for the CPAT circuit has been established at 10 min 20 s, but in accordance with the CPAT timed trials that are allowed before the formal test, subjects were permitted to continue to completion or until they decided they were unable to complete the circuit.

  1. Stair Climb. Subjects wore two additional shoulder weights (5.67 kg each) to simulate the weight of a high-rise pack and climbed on the step mill (Stair Master, Stepmill 7000 PT, StairMaster Sports, Kirkland, WA) for 3 min at a stepping rate of 60 steps per min.
  2. Hose Drag. Subjects dragged a 44-mm hose equipped with a 3-kg automatic nozzle at a distance of 22.86 m to a pre-positioned drum then made a 90° turn and dragged the hose to an additional 7.62 m. The subject then dropped to one knee and pulled the hose for 15.24 m.
  3. Equipment Carry. Subjects removed two saws (14.5 and 12.7 kg, respectively), one at a time, from a tool cabinet and placed them on the floor. They picked up both saws and walked 22.86 m around a pre-positioned drum and back to the starting point. At the tool cabinet, subjects placed the saws on the ground and then put them one at a time back into the cabinet.
  4. Ladder Raise and Extension. Subjects grabbed a 7.32-m ladder by the top rung and walked it up, using each rung on the ladder, until it was placed against the wall. Subjects then proceeded to a secured ladder and extended it 7.32 m to its stopper before lowering it in a controlled manner back to the ground.
  5. Forcible Entry. Subjects picked up a 4.54-kg sledgehammer and struck the forcible entry device until a buzzer was activated. The forcible entry device simulated hitting a door and had a resistance between 700 and 800 psi (4826-5516 kN·m−2) requiring approximately four to five solid hits by most men. After activating the buzzer, subjects placed the sledgehammer on the ground.
  6. Search. Subjects crawled on their hands and knees through a blackout tunnel for a total length of 19.51 m with two 90° right turns. Subjects had to maneuver over, under, and around obstacles before reaching the exit of the maze.
  7. Rescue. Subjects had to grasp a 74.84-kg manikin by the shoulder harness, using one or both hands, and drag it 10.67 m around a pre-positioned drum and back to an additional 10.67 m to the finish line.
  8. Ceiling Breach and Pull. Subjects picked up a pike pole then used it to push the door of the ceiling completely open three times followed by hooking an adjacent ceiling device and pulling down completely five times. Subjects completed four sets of three pushes and five pulls with the time stopped upon completion of this event.

Data analysis

Anthropometric data and results of fitness testing were compared between men and women and between finishers and nonfinishers with one-way ANOVA. Breath-by-breath values of E, V˙O2, V˙CO2, and RER along with the HR during each breath were recorded. Average values were then calculated from the last 20-30 s of each event. Group data were used to obtain the mean and SD for the men and the women separately at each event. Comparisons were made at each event by one-way ANOVA with significance accepted for P < 0.05. Comparisons were also made between women who passed by the criterion time and those who finished but did not meet the criterion time by one-way ANOVA. Because of the smaller sample size in comparisons of pass versus fail, trends are reported for P < 0.1. In addition, coefficients of determination (r2) were calculated for pairs of variables, and stepwise backward regression (SigmaStat V. 3.5, Systat Software Inc., Point Richmond, CA) was used to estimate predictive models of performance on the CPAT.


Of the 34 men and 23 women who participated in this study, 2 men (5.9%) and 9 women (39.1%) were unable to complete the CPAT circuit. The data of the nonfinishers have been displayed separately in Tables 1 and 2; they are contrasted with finishers but they were not included in any other analyses related to CPAT completion time.

Group means for predicted 1-RM muscular strength measurements (handgrip, bench press, shoulder press, biceps curls, and leg press), muscular endurance to fatigue (bench press and leg press), and Wingate performance for those who completed irrespective of time (a) and those who did not complete (b) the CPAT.

The women who completed the CPAT circuit were significantly shorter, weighed less (Table 1), and had higher body mass index (BMI; 25.2 ± 3.6 vs 22.9 ± 2.8 kg·m−2) than the men in this study (P < 0.05). Women who did not complete the CPAT circuit did not differ from the women who did complete the circuit in terms of height or body mass (Table 1). Likewise, there were no differences between women who passed and those who failed the criterion time (pass: 167.8 ± 6.5 cm, 64.7 ± 3.7 kg; fail: 167.9 ± 4.5 cm, 64.6 ± 12.3 kg).

The men who completed the CPAT circuit had greater values than women during the maximal treadmill testing for each of peak E, V˙O2max (mL·min−1), and peak V˙CO2 (P < 0.05; Table 1). There was no difference between men and women in V˙O2max expressed as milliliters per kilogram per minute and in HRmax and RER (Table 1; P > 0.05). Women who did not complete the CPAT circuit when compared to those who did complete the circuit had significantly lower V˙O2max (expressed in absolute or relative terms) as well as peak V˙CO2 and a higher HRmax (Table 1). Women who passed the criterion time had V˙O2max only marginally greater than those who failed (53.4 ± 5.6 vs 51.23 ± 6.7 mL·kg−1·min−1, P > 0.1). Statistical comparisons were not made for the men who did not complete the CPAT circuit because of the very small sample size (n = 2), but it can be seen that they were slightly smaller but had markedly lower values of V˙O2max.

Muscular strength and endurance tests

Men who completed the circuit were significantly stronger than women in terms of maximal bench press, maximal biceps curl, maximal leg press, maximal shoulder press, and maximal handgrip strength (Table 2; P < 0.05). There were strong correlations when comparing men and women together between different measures of upper body strength (bench press to shoulder press, r2 = 0.80; bench press to bicep curl, r2 = 0.71); therefore, further comparisons will consider only bench press. Men also had significantly greater muscular endurance than the women in terms of total number of repetitions to fatigue for bench press and leg press endurance (Table 2; P < 0.05). The women who did not complete the CPAT had strength and endurance values that were remarkably lower than the women who did complete the circuit (Table 2). Likewise, the two men who did not finish the circuit were considerably less strong and had lower muscular endurance than the finishers. Women who passed the criterion time had slightly greater endurance (bench press: 19.0 ± 14.1 vs 10.9 ± 6.6 repetitions; leg press: 26.5 ± 16.0 vs 17.6 ± 16.0 repetitions) but the differences were not significant (P > 0.1).

The Wingate test revealed that the men had significantly greater peak power than the women (Table 2, P < 0.05) but there was no difference in the fatigue index. The women who did not finish the circuit had lower peak power than the finishers. As with muscle strength and endurance, women who passed by the criterion time had a fatigue index that was slightly less than women who completed the circuit but failed the criterion time (36.0% ± 3.3% vs 43.2% ± 9.9%), but the difference was not significant.

CPAT-simulated firefighting circuit

The energy demands of the circuit are shown for a single male subject in Figure 1 by the time course of V˙O2, V˙CO2, and HR across the eight circuit items. The mean completion time for the circuit was significantly faster for men (8 min 32 s ± 51 s) than for women (11 min 16 s ± 1 min 28 s), resulting in a pass/fail rate of 91% (31/34) and 15% (4/23), respectively (P < 0.05). Time to complete the individual events was significantly slower for women than men for each of the seven events after the stair climb, especially for last two events, namely, the victim rescue and ceiling breach and pull, which took approximately twice as long for the women (Fig. 2). The women who passed the criterion time had faster times for all events than those who failed to meet the criterion time with the largest differences in the search (14.8 ± 3.6 vs 21.2 ± 6.2 s, P = 0.07), the rescue (46.0 ± 6.6 vs 58.3 ± 22.6 s, P > 0.1) and the breach and pull (64.3 ± 6.6 vs 117.7 ± 52.7 s, P = 0.07).

The oxygen uptake (V˙O2; black), carbon dioxide output (V˙CO2; gray), and HR (dashed) are displayed for a male subject during the eight-event CPAT. The individual events (1 = 3-min step mill exercise, 2 = hose drag, 3 = equipment carry, 4 = ladder raise, 5 = forcible entry, 6 = maze search, 7 = victim rescue, and 8 = ceiling breach and pull) were separated by 22.9-m walking.
Time for each of the eight events in the CPAT circuit is shown for men (solid bars) and women (open bars). Values are mean ± SD. * Significant difference between men and women (P < 0.05).

Compared to the women, the men had significantly greater values during the entire CPAT circuit for each of absolute V˙O2 (3128 ± 484 vs 2357 ± 368 mL·min−1), V˙CO2 (3209 ± 368 vs 2270 ± 318 mL·min−1), and expired ventilation (103.9 ± 10.6 vs 85.0 ± 7.0 L·min−1; Fig. 3; P < 0.05). There were no differences during the circuit between men and women for relative V˙O2 (38.5 ± 5.3 vs 36.6 ± 5.5 mL·kg−1·min−1), percent of V˙O2max (73.1% ± 8.0% vs 70.6% ± 7.3%), HR (169 ± 10 vs 171 ± 5 bpm), percent of HRmax (90.1% ± 5.3% vs 90.8% ± 3.2%), and RER (1.03 ± 0.11 vs 0.97 ± 0.12; Figs. 3 and 4; P > 0.05). The women who passed the criterion time had a relative V˙O2 during the circuit that was higher than those who did not meet the criterion time (78.6% ± 4.8% vs 67.5% ± 5.5%, P < 0.02).

Expired ventilation (E), RER, HR, and oxygen uptake (relative to body mass and absolute V˙O2) were measured for men (solid circles) and women (open circles) during each of the eight events of the CPAT circuit. See Methods for complete description of the events. Equip. Carry indicates equipment carry; Ladder, ladder raise; Forcible, forcible entry; Ceiling, ceiling breach and pull. Values are mean ± SD. *Significant difference between men and women (P < 0.05).
HR and V˙O2 response represented in percent of HRmax and percent of V˙O2max for men (closed circles) and women (open circles). Values are mean ± SD. Note different ranges on y axes. *Significant difference between men and women (P < 0.05).

Laboratory exercise tests versus CPAT-simulated firefighting circuit

To test the hypotheses that faster completion times in the CPAT circuit would be associated with high levels of aerobic fitness and muscle strength, individual and stepwise linear regressions were performed. Linear regression of completion time on individual variables revealed that absolute V˙O2max had a strong relationship for men, women, and the combined men plus women observations (Table 3 and Fig. 5). On the other hand, relative V˙O2max had a poorer relationship to completion time except in the women (Table 3). With the exception of the relationship between completion time and relative V˙O2max, the coefficients of determination increased when men and women were combined rather than for the men or the women alone (Table 3).

Correlation coefficients from simple linear regression for time to completion of the CPAT circuit with respect to key physiological variables in men, women, and men and women combined.
The relationship is shown between completion time on the CPAT circuit and relative V˙O2max (mL·kg−1·min−1; A) absolute V˙O2max (mL·min−1; B) for men (solid circles) and women (open circles). Regression lines were calculated for combined men and women. The regression equation for the relationship in B is as follows: Completion time = 16.67 - 0.0018V˙O2max. P < 0.001, SEE = 1.09 min. Horizontal dashed lines indicate the pass/fail criterion time of theCPAT.

Backward stepwise regression to predict completion time for all men and women was performed by including different sets of independent variables that accounted for the collinearity of body mass and absolute V˙O2max compared to relative V˙O2max. Two different sets of regression analyses were performed, namely, one included relative V˙O2max and the other included absolute V˙O2max, and each model included body mass, maximal bench press, bench press endurance, maximal leg press, leg press endurance, anaerobic capacity, and anaerobic peak power. In the backward stepwise regression with absolute V˙O2max, each of these variables except absolute V˙O2max was removed from the final model. The equation that predicts completion times for all finishers was as follows:

The SEE was 1.30, and the estimates of variability for the coefficients given by the SE were 1.132 and 0.00028, respectively, with r2 = 0.65 (P < 0.001, power with α = 0.05: 1.0).

When the backward stepwise regression was performed with relative V˙O2max, the equation predicting completion times was as follows:

The SEE was 1.25, and the SE for the coefficients were 2.512, 0.0373, 0.0259, and 0.0313, respectively, with r2 = 0.71 (P < 0.001, P = 0.052, P = 0.023, respectively, power with α = 0.05: 1.0).


This is the first study to describe the energy requirements for completing the Candidate Physical Ability Test (CPAT) within the guidelines as prescribed by the International Association of Fire Chiefs and the International Association of Fire Fighters. Confirming the first part of our hypothesis, the energy requirements were high with V˙O2 averaging approximately 37-39 mL·kg−1·min−1 or 70%-73% V˙O2max and with HR approximately 90% of maximum across the entire circuit in men and women. The second component of the hypothesis that anaerobic energy contribution would be relatively high was also confirmed by the elevated RER that was higher throughout the circuit than anticipated for this percentage of V˙O2max. The final component of the hypothesis was partially supported by the multiple linear regression analysis as we found that performance time in the circuit could be predicted by absolute V˙O2max or by relative V˙O2max in combination with handgrip and body mass. However, upper and lower body strength measurements were not included within the prediction models. Overall, the results in this circuit designed to test candidate recruits showed energy requirements that were at least into the low to mid range of values of V˙O2 or HR measured during testing of incumbent firefighters doing simulations of work tasks (5,8,14,16,20,22,25).

The CPAT is a bona fide occupational qualification developed by the IAFF/IAFC Joint Labor Management Wellness/Fitness Task Force to serve as a screening tool for determining, which candidate recruits were able versus not able to perform the physically demanding job of firefighting (11). The eight-event circuit was developed by an expert panel that first surveyed approximately 100 firefighters in each of 10 different jurisdictions in North America to determine representative tasks. The criterion time was established by filming completion of the circuit at 13 different rates with finishing times between 7:30 and 12:10 then having 33 different training officers rate the performance in each videotape as acceptable, marginally acceptable, marginally unacceptable, or unacceptable. The time of 10:20 corresponded to the boundary between the marginally acceptable and the marginally unacceptable rating. The results from our study allow characterization of the demands of the tasks and provide insight into the types of individuals capable of meeting the criterion time.

We found a success rate for our subjects to complete the CPAT within the criterion time of 10 min 20 s to be 91% of men and 15% of women. Although these rates differ somewhat from the success rates for firefighter candidates on the same circuit (98% and 48%, respectively), the distribution of the time to completion of our population was representative of testing under actual conditions (Fig. 6). Therefore, we believe the current results to be representative of the demands of this bona fide occupational qualification and to reveal the characteristics of those individuals most likely to be successful in meeting the standard established by the CPAT.

The 896 men who successfully passed the CPAT as part of their job application requirements are used as the reference framework for describing the relative distribution of the successful female candidates as well as the male and female research subjects. The individual men's data are presented as "Time Remaining for Pass," which was calculated as the difference between the criterion time (10 min 20 s) and the individual's completion time then plotted (small circles) in a cumulative fashion for all 896 successful male candidate recruits tested in the facility. Superimposed at the appropriate point on the distribution of times for the men are the time remaining for the 34 successful female candidate recruits (inverted triangles in A) and 4 successful female research subjects (large open circles in A) and for the 31 successful male research subjects (upright triangles in B).

Cardiorespiratory response to CPAT

The first event of the CPAT is the 3-min stair climb while wearing the 22.68-kg weighted vest plus the two 5.67-kg shoulder weights. As anticipated, this caused a marked rise in V˙O2 and HR. In the final 30 s of stair climbing, V˙O2 was markedly elevated, reaching values of approximately 38 mL·kg−1·min−1 or more than 3000 mL·min−1 in the men (73.6% ± 8.0% V˙O2max) and 40 mL·kg−1·min−1 or more than 2500 mL·min−1 in the women (76.9% ± 7.6% V˙O2max). The %V˙O2max for women who completed the CPAT under the criterion time was 81.7% ± 8.1%, whereas for those unable to complete the circuit within the criterion time, it was 74.9% ± 6.9%. HR, which was observed to be 166-173 bpm and was already close to 90% HRmax by the end of the stair climb. The kinetics of the V˙O2 and HR responses are known to be very fast during heavy exercise (10,23), so observation of high levels by the end of 3 min is not surprising. The high values of V˙O2 and HR are also consistent with previous research of stair climbing while wearing a firefighter gear. Manning and Griggs (17) found that within the first minute of a firefighting task, HR increases 70% to 80% of maximum with a subsequent plateau between 90% and 100% until the task is completed. O'Connell et al. (19) observed the V˙O2 in excess of 80% V˙O2max and HR more than 90% maximum after 5 min of stair climbing. Firefighters in the study of von Heimburg et al. (25) had V˙O2 of 34 ± 4 mL·kg−1·min−1 or 64% ± 7% V˙O2max and HR of 167 ± 13 bpm or 88% ± 4% maximum after climbing six flights of steps (20.5 m vertical) in 90 ± 31 s. The CPAT protocol mandates walking between events, and there is slight recovery of V˙O2 and HR (Fig. 1); however, overall energy demands remain high. Throughout the circuit, men had significantly higher absolute values of E, V˙O2, and V˙CO2. During the stair climb activity, women had slightly higher values for V˙O2 when expressed as milliliters per kilogram per minute or percent V˙O2max, and they had significantly higher HR (absolute and percent maximum) during the stair climbing compared to the men. However, in the subsequent activities (hose drag, equipment carry, and ladder raise), the women's V˙O2 and HR declined until the women had significantly lower values for V˙O2 than the men during the ladder raise and forcible entry. The women's V˙O2 and HR both increased during the search and the victim rescue so that they were not different from the men. At this point, it is worthy to note that the four women who completed the circuit in less than 10 min 20 s had an average V˙O2 during the circuit of 41.8 ± 3.3 versus 34.5 ± 4.8 mL·kg−1·min−1 for the women who failed to meet the criterion time (78.6% ± 4.8% vs 67.5% ± 5.5% V˙O2max), so the slower women did bias the overall mean for some variables.

The RER reflects the dynamic relationship between V˙CO2 and V˙O2 as dictated by gas storage and metabolism including both oxidative phosphorylation and anaerobic glycolysis. At the onset of the stair climbing exercise, V˙CO2 initially lags behind V˙O2 (Fig. 1) owing to CO2 storage (9), but it can be seen from the RER taken as the average of the last 30 s of the 3-min stair climbing that V˙CO2 was already exceeding V˙O2 during the first activity of the CPAT. The RER for the men, as in a previous study (6), was consistently above 1.0 and exceeded 1.1 during the hose drag and the ceiling breach and pull. For the women, RER was above 1.0 during the stair climbing but then progressively decreased and was significantly less than the men at several activities (with no difference between women who met and did not meet the criterion time). As the metabolic respiratory quotient for the men was probably close to 1.0 during the CPAT, reflecting predominantly carbohydrate metabolism, excess CO2 could be estimated from the difference between V˙CO2 and V˙O2. For the typical subject shown in Figure 1, the excess volume of CO2 produced between 1 and 10 min of exercise was approximately 1570 mL of CO2 (70 mmol CO2). It can be assumed that this 70-mmol output of CO2 reflects a total of 70-mmol reduction in predominately plasma bicarbonate and an approximately 70-mmol increase in lactate. It is difficult to know the distribution of this lactate, but if distributed in 5-6 L of total blood volume, the numbers are consistent with the 13-mmol·L−1 blood lactate reported by von Heimburg et al. (25) after stair climbing and victim rescue simulations.

The cause of the marked increase in RER and blood lactate, although the V˙O2 is only approximately 75%-80% V˙O2max throughout the whole CPAT circuit, is likely the inadequate O2 delivery owing to high muscle tensions required to perform the tasks including carrying the extra weight of the vest simulating the firefighter's protective equipment. Muscle blood flow can be impeded during relatively low-force isometric contractions (7). Even the stair climbing had an important anaerobic contribution to metabolism because of the very high forces generated in the leg muscles and in trunk support muscles to raise body mass plus an additional 34 kg. Unfortunately, technology is not available in humans to measure blood flow under the conditions of this testing. However, the technology to assess continuous V˙CO2 in combination with V˙O2 as in the current study and in our previous report (6) provides considerable insight into the true metabolic demands of firefighting that are hidden by previous studies that reported only average V˙O2 values of approximately 70% V˙O2max without data for V˙CO2. Moreover, these findings suggest that tests that rely on exercise capacity at a predicted submaximal HR (18) also fail to represent the demands of firefighting.

Predictors of CPAT performance

Simple linear relationships between CPAT completion time and physical characteristics of the subjects revealed some modest correlations. For most characteristics, the correlation was greater when men and women were combined (Table 3), and this is not surprising because there was a greater range of values for attributes that are largely dependent on body mass. Individual analysis of the men and women actually revealed a very low correlation between completion time and body mass, whereas other factors such as absolute and relative V˙O2max, bench press strength and endurance, and handgrip had modest predictive ability within the sexes.

Backward stepwise multiple linear regression analyses for the total sample of 32 men and 14 women who finished the test were performed to determine whether combinations of physical attributes might predict CPAT completion time. Because of the relationship between absolute V˙O2max and the combination of body mass with relative V˙O2max, these two components were examined independently within the analyses. When absolute V˙O2max was included in the model, all other variables (body mass, strength, endurance, and power indicators) were dropped from the predictive equation. However, when relative V˙O2max was entered in the model, it emerged in combination with body mass and handgrip as a significant predictor of completion time. Each of these models was capable of explaining 65%-71% of the variance in CPAT completion time. However, it is important to note that the SEE were more than 1 min 15 s. This suggests that V˙O2max alone is not an adequate predictor of CPAT performance. Indeed, it can be seen in Figure 5A and B that men who completed the CPAT between 8 and 9 min could have a V˙O2max between 38 and 67 mL·kg−1·min−1 or approximately between 3800 and 5400 mL·min−1. This latter observation is similar to that of von Heimburg et al. (25) who found an error of estimate for a linear regression of 65 s for performance time during a high-rise rescue simulation predicted from absolute V˙O2max.

Practical considerations

In comparison with the actual applicants for positions as firefighters, our population of male subjects had slightly faster than average times for completion of the CPAT circuit, whereas our female subjects were slightly slower than the female candidates. The fastest time recorded was 6 min 50 s by a male candidate recruit (who completed the circuit without the gas collection system) compared to our fastest male subject whose fastest time recorded was 7 min 10 s. The distribution of candidate times is shown in Figure 6, with each male candidate represented by a single dot in a cumulative fashion (i.e., of the 896 successful male candidates, approximately 100 men completed the CPAT with less than 1 min remaining to the criterion time, male candidates numbered from 100 to 800 had between 1 min and 2 min 15 s remaining, and the final 96 had more than 2 min 15 s remaining). The fastest female recruit was 8 min 10 s, whereas our fastest female subject was 9 min 1 s. Our male subjects were distributed along the spectrum of times observed in the candidate recruits with some bias toward faster times (Fig. 6B), but the female subjects were toward the lower end of the spectrum of successful candidate recruits (Fig. 6A). As noted, the success rates for the actual candidates tested in this same facility were 98% and 48% for men and women, respectively, compared to our subjects at 91% and 15%. A probable reason for the relatively high success rate of the female candidates is that women who elect to become firefighters have recognized the need to be physically fit (aerobic fitness, strength, and muscle endurance) to enter their chosen profession, whereas our subjects were selected as fit and healthy young women with no specific training. Unfortunately, we do not have data on physical dimensions or physical fitness in the female candidates; however, based on success rates, they might have been fitter than our subjects and had relatively lower energy expenditure and HR. Our subjects tended to be lean (BMI: men = 25.2 ± 3.6 kg·m−2 and women = 22.9 ± 2.8 kg·m−2) and might not be fully representative of the firefighter candidates or incumbents. Another factor that can explain the relatively high success rate of male and female candidate recruits is that several municipalities that use the CPAT for screening require completion of a community college program that does include a course in physical fitness. It is not possible to differentiate for the actual candidates who were enrolled in the college program and who were not. Information is available on the stage of the CPAT at which individuals failed the test. Of the 34 men who failed, 6 were unable to complete the step mill phase, 3 stopped during the test for different reasons, 8 reached the 10 min 20 s time limit, and all of these were at the final event. Of the 37 women who failed the CPAT, 14 were unable to complete the step mill test, 1 timed out during the search, 6 timed out during the victim rescue, and 16 timed out during the final breach and pull event.


We studied the energy demands of the CPAT in a group of young men and women who had some characteristics similar to actual firefighter candidate recruits but who might not be fully representative of that population. We found the energy demands to be high, although perhaps at the lower end of values recorded in previous studies of incumbent firefighters performing task simulations (4,8,14,16,20,22,25). In our population of male and female test subjects, more than 65% of the variance in CPAT completion time was predicted by maximal aerobic power, both absolute V˙O2max alone or relative V˙O2max in combination with body mass and handgrip strength. However, the predictive ability was quite low as evidenced in an SEE that exceeded 1 min 15 s. Therefore, high physical fitness is required to successfully complete the CPAT, but the ability to predict performance from individual elements such as V˙O2max is not sufficiently accurate. The CPAT evaluates physical fitness of firefighter candidate recruits at a specific time point, but it does not indicate the likelihood of maintaining long-term fitness.

This research was supported by the Ontario Workplace Safety Insurance Board of the Province of Ontario (No. 04024). The authors thank Caryl Russell and Lori Kraemer of UW Fitness for their assistance with the CPAT testing site and for providing candidate recruit testing data.

No financial conflicts. The results of the present study do not constitute endorsement by ACSM.


1. Davis PO, Dotson CO, Santa Maria DL. Relationship between simulated fire fighting tasks and physical performance measures. Med Sci Sports Exerc. 1982;14(1):65-71.
2. Deakin JM, Smith JT, Pelot R, Weber CL. Methodological considerations in the development of physical maintenance standards. In: Gledhill N, Bonneau J, Salmon A. editors. Proceedings of the National Forum on Bona Fide Occupational Requirements. Toronto, ON, Canada, September 13-16, 2000. pp. 17-21
3. Gilman WD, Davis PO. Fire fighting demands aerobic fitness. NFPA Journal. 1993;2:68-73.
4. Gledhill N, Jamnik VK. Characterization of the physical demands of firefighting. Can J Sport Sci. 1992;17:207-13.
5. Gledhill N, Jamnik VK. Development and validation of a fitness screening protocol for firefighter applicants. Can J Sport Sci. 1992;17:199-206.
6. Harvey DG, Kraemer JL, Sharratt MT, Hughson RL. Respiratory gas exchange and physiological demands during a fire fighter evaluation circuit in men and women. Eur J Appl Physiol. 2008;103:89-98.
7. Hicks A, McGill SM, Hughson RL. Tissue oxygenation by near infrared spectroscopy and muscle blood flow during isometric contractions of the forearm. Can J Appl Physiol. 1999;24:216-30.
8. Holmer I, Gavhed D. Classification of metabolic and respiratory demands in fire fighting activity with extreme workloads. Appl Ergon. 2007;38:45-52.
9. Hughson RL, Inman MD. Gas exchange analysis of immediate CO2 storage at onset of exercise. Resp Physiol. 1985;59:265-78.
10. Hughson RL, O'Leary DD, Betik AC, Hebestreit H. Kinetics of oxygen uptake at the onset of exercise near or above peak oxygen uptake. J Appl Physiol. 2000;88:1812-9.
11. International Association of Fire Chiefs (IAFC) [Internet]. The IAFF/IAFC Fire Service Joint Labor Management Candidate Physical Ability Test Program Summary. [cited 2008 May 8]. Available from: =1&subarticlenbr=389.
12. Kilbom A. Physical work capacity of firemen. With special reference to demands during fire fighting. Scand J Work Environ Health. 1980;6:48-57.
13. Kraemer WJ, Fry AC. Strength testing: development and evaluation of methodology. In: Maud PJ, Foster C. editors. Physiological Assessment and Human Fitness. Champaign (IL): Human Kinetics; 2006. pp. 115-38.
14. Lemon PW, Hermiston RT. The human energy cost of fire fighting. J Occup Med. 1977;19:558-62.
15. Louhevaara V, Smolander J, Korhonen O, Tuomi T. Maximal working times with a self-contained breathing apparatus. Ergonomics. 1986;29:77-85.
16. Louhevaara V, Soukainen J, Lusa S, Tulppo M, Tuomi P, Kajaste T. Development and evaluation of a test drill for assessing physical work capacity of fire-fighters. Int J Ind Ergon. 1994;13:139-46.
17. Manning JE, Griggs TR. 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. 1983;25:215-8.
18. Mier CM, Gibson AL. Evaluation of a treadmill test for predicting the aerobic capacity of firefighters. Occup Med (Lond). 2004;54:373-8.
19. O'Connell ER, Thomas PC, Cady LD, Karwasky RJ. Energy costs of simulated stair climbing as a job-related task in fire fighting. J Occup Med. 1986;28:282-4.
20. Romet TT, Frim J. Physiological responses to fire fighting activities. Eur J Appl Physiol Occup Physiol. 1987;56:633-8.
21. Sothmann MS, Saupe K, Jasenof D, Blaney J. Heart rate response of firefighters to actual emergencies. Implications for cardiorespiratory fitness. J Occup Med. 1992;34:797-800.
22. Sothmann MS, Saupe KW, Jasenof D, et al. Advancing age and the cardiorespiratory stress of fire supression: determining a minimum standard for aerobic fitness. Hum Perform. 1990;3:217-36.
23. Tordi N, Perrey S, Harvey A, Hughson RL. Oxygen uptake kinetics during two bouts of heavy cycling separated by fatiguing sprint exercise in humans. J Appl Physiol. 2003;94:533-41.
24. UW Fitness [Internet]. Fire Fighter Testing. [cited 2008 May 8]. Available from:
25. von Heimburg ED, Rasmussen AK, Medbo JI. Physiological responses of firefighters and performance predictors during a simulated rescue of hospital patients. Ergonomics. 2006;49:111-26.


©2009The American College of Sports Medicine