Abstract

Study Objectives:

Firefighters’ schedules include extended shifts and long work weeks which cause sleep deficiency and circadian rhythm disruption. Many firefighters also suffer from undiagnosed sleep disorders, exacerbating fatigue. We tested the hypothesis that a workplace-based Sleep Health Program (SHP) incorporating sleep health education and sleep disorders screening would improve firefighter health and safety compared to standard practice.

Design:

Prospective station-level randomized, field-based intervention.

Setting:

US fire department.

Participants:

1189 firefighters.

Interventions:

Sleep health education, questionnaire-based sleep disorders screening, and sleep clinic referrals for respondents who screened positive for a sleep disorder.

Measurements and Results:

Firefighters were randomized by station. Using departmental records, in an intention-to-treat analysis, firefighters assigned to intervention stations which participated in education sessions and had the opportunity to complete sleep disorders screening reported 46% fewer disability days than those assigned to control stations (1.4 ± 5.9 vs. 2.6 ± 8.5 days/firefighter, respectively; p = .003). There were no significant differences in departmental injury or motor vehicle crash rates between the groups. In post hoc analysis accounting for intervention exposure, firefighters who attended education sessions were 24% less likely to file at least one injury report during the study than those who did not attend, regardless of randomization (OR [95% CI] 0.76 [0.60, 0.98]; χ2 = 4.56; p = .033). There were no significant changes pre- versus post-study in self-reported sleep or sleepiness in those who participated in the intervention.

Conclusions:

A firefighter workplace-based SHP providing sleep health education and sleep disorders screening opportunity can reduce injuries and work loss due to disability in firefighters.

Statement of Significance

Firefighters are exposed to many risk factors that can affect their sleepiness including long work shifts, arduous work and a high prevalence of undiagnosed sleep disorders. This randomized clinical trial shows that a Sleep Health Program, consisting of a mandatory educational session plus a voluntary sleep disorders screening, diagnosis and treatment program, can substantially reduce absences due to injury and disability and decrease reported injuries. A specific sleep health education and sleep disorders risk screening program should be incorporated into workplace health and wellness initiatives to reduce the physical, mental and economic burden of avoidable sleepiness-related workplace injuries in firefighters.

INTRODUCTION

Firefighters work demanding schedules under highly stressful and challenging conditions. The need to work frequent extended shifts and long work weeks leads to acute and chronic sleep deficiency as well as misalignment of circadian phase.1 In addition, it is likely that a significant proportion of firefighters suffer from undiagnosed sleep disorders, which may further disturb sleep and exacerbate fatigue.2 Sleep disorders are common, costly, and treatable, but often remain undiagnosed and untreated.3 Our national survey of nearly 7000 firefighters, conducted across 66 fire departments, found that 37% of active firefighters were at risk of a primary sleep disorder when assessed using validated questionnaires, particularly obstructive sleep apnea (OSA), most of whom (81%) were undiagnosed.4

Since firefighter injuries and disability costs exact a human toll and financial burden on both firefighters and the communities whom they serve, we investigated the efficacy of a Sleep Health Program (SHP) on the risk of injury and use of disability time in firefighters.

Specifically, we used a station-level, randomized experimental design to test the hypotheses that implementation of an SHP will: (1) improve firefighters’ health as assessed as fewer days of leave due to sickness/illness or injury/disability; (2) improve firefight safety assessed as (a) fewer on-the-job motor vehicle crashes and (b) fewer reported injuries.

MATERIALS AND METHODS

Fire Department Selection

We initially invited 73 departments to consider participating in our program based on department size and workload, and received expressions of interest from 32 departments. Interested departments were sent additional information and appointments were arranged with senior departmental, union, and wellness program representatives to discuss the program. Final selection was based on department cooperation, the practicalities of initiating the program, and the availability of departmental measures to use in analysis of program impact. We selected Columbus OH Division of Fire, a mid-sized fire department with 1211 active firefighters and 32 stations. The study was approved by the Partners Healthcare Human Research Committee and registered at ClinicalTrials.gov Identifier: NCT01988129. (Note, the ClinicalTrials.gov registration originally incorporated additional work [not directly related to this trial]. For clarity, we have now reverted to describing only this original study in ClinicalTrials.gov and updated the entry accordingly.)

Cohort Selection and Station-Level Randomization

Stations were paired according to the previous calendar years’ workload, number of employees, and general station type. Retrospective workload for the 54-week study period was captured by the number of equipment runs (ladder and engine) and medic runs per station. Following pairing, the number of firefighters and workload were compared between the two groups using paired Student’s t tests. One station from each pair was then randomly selected to receive the SHP by the investigators. All personnel from the intervention stations were instructed to attend scheduled education sessions. Any firefighter present at the station was permitted to attend. Nominal scheduling operations for the fire department resulted in some spillover between personnel from control and intervention stations. Participation in education sessions was captured via sign-in sheets.

Intervention

The intervention consisted of three parts: (1) Mandatory Educational Sessions, (2) Voluntary Sleep Disorders Screening, and (3) Sleep Disorders Diagnosis and Treatment for those who screened at risk for a sleep disorder.

Mandatory Educational Sessions

Firefighters were scheduled to attend a mandatory 30-minute education training presentation centrally at fire department headquarters as operations allowed. These sessions were provided to firefighters over the last 2 weeks of August, as documented by a sign-in attendance sheet. Nearly all sessions (98%) were conducted by one investigator (CSO) and the remaining 2% by another (SWL). The session provided information on firefighter mortality, fatigue-related health hazards, and the physiological importance of sleep. The sleep health education also included strategies to improve sleep hygiene and tips on how to use caffeine and naps effectively to promote alertness. Eye masks and ear plugs were distributed to aid with sleep hygiene. This information was supplemented by distribution of brochures on various sleep health topics from the American Academy of Sleep Medicine and referral to the Harvard Division of Sleep Medicine education site: www.understandingsleep.org.

Voluntary Sleep Disorders Screening

Following the education, firefighters were invited and encouraged to complete a voluntary sleep disorders screening questionnaire. Prior to administration of the questionnaire, volunteers signed a research consent form. Participation was kept confidential; neither the department nor union was informed as to who completed the voluntary sleep disorders screening. This questionnaire used validated, self-report screening tools for OSA (Berlin Questionnaire; sensitivity 0.86, specificity 0.77),5 moderate to severe insomnia (Athens Insomnia Scale; sensitivity 0.93, specificity 0.85),6 and restless legs syndrome (Epidemiology, Symptoms, and Treatment Questionnaire; sensitivity 0.82, specificity 0.90).7 For shift work disorder, we created a screening tool based on the International Classification of Sleep Disorders-2 diagnostic criteria.8

Sleep Disorders Diagnosis and Treatment

All of the respondents who screened positive for risk of any sleep disorder were notified by letter and provided with contact information for local American Academy of Sleep Medicine–certified, partnering sleep clinics. Participants were encouraged to use a sleep clinic partnering with the study but were free to seek medical follow-up elsewhere. Telephone calls were also made to all at-risk participants to ensure that they were aware of the results and to facilitate scheduling a sleep clinic appointment. Participants were asked to provide voluntary medical records release consent for tracking diagnoses.

Primary Outcome Measures

Health, safety, and performance outcomes were assessed cumulatively over the study period from departmental records.

Firefighter Health

The number of 24-hour pay periods coded as “sick” time per firefighter and the number of 24-hour pay periods coded as injury/disability per firefighter were assessed from payroll records. Fewer leave days lost to sickness/illness or injury/disability were considered indicative of better health. Only leave days occurring after study initiation were included in these analyses. Contiguous long-term disability days that began before study start were excluded.

Firefighter Safety—Motor Vehicle Crashes

Accidents were counted as any incident that resulted in the filing and review of a department Fleet Accident Report. Fewer accidents were considered indicative of better safety. Only Fleet Accident Reports made after study initiation were included in these analyses.

Firefighter Safety—Injuries

On-the-job injuries that triggered the filing of an official city government injury report as the result of following normal departmental procedures were included in this study. Fewer injury reports were considered indicative of better safety. Only injury reports made after study initiation were included in these analyses.

Firefighter Performance—Response Time

Following detailed review of departmental procedures and records, we determined that “turn-out time” was already very rapid and not considered an accurate measure of firefighters’ performance by the department. Clearance time (time from the start until the end of the event), which could last for many hours, was also not considered an appropriate measure of firefighters’ performance in relation to sleep and alertness given the multiple factors, many of which are not under the control of the firefighters, that could affect clearance times. Similarly, we intended to collect additional data on cognitive performance but found that it was impractical to do so in this operational environment. We therefore did not address firefighter performance.

Baseline and End-of-Year Surveys

All individuals in the intervention stations were invited to complete a voluntary survey at the start of the study in conjunction with the education and sleep disorders screening intervention. Firefighters in the control group were not asked to complete the survey in order to ensure that any engagement with the study did not influence their behavior. These surveys provide comparative data pre- versus post-intervention to examine whether the program significantly improved sleep and general health. The baseline education and survey data were collected in the last 2 weeks of August 2009. Year-end surveys were collected from August through October 2010. Database measures encompass data from mid-August 2009 to September 2010. Sleep clinic follow-up records collection was completed in August 2013. We had intended to collect additional survey data to assess changes in firefighters’ and families’ job satisfaction but, due to unanticipated additional work in gathering the primary outcomes, we were unable to do so.

Data Analysis

Data are reported as mean ± SD unless noted otherwise. Comparisons between the control and intervention group for primary outcomes used unpaired Student’s t tests assuming equal variances. Disability days also used a hierarchical mixed model to allow for possible station-level pairing correlation. In addition, a negative binomial regression was used to evaluate disability days due to overdispersion and high percentage of firefighters reporting zero disability use. Chi-square Fisher’s exact tests were used to compare the gender and race between the control and intervention groups. Post hoc odds ratios were calculated to compare risk of reported injuries across those who did or did not attend the educational sessions. For survey outcomes, paired Student’s t tests were used to compare pre- versus post-intervention outcomes. Statistical analyses were conducted using SAS for Windows (ver. 9.4, SAS Institute Inc., Cary, NC).

Our study was powered assuming 500 firefighters per group, 90% power, and a two-sided alpha of 0.05. For our primary departmental database outcomes (injury and motor vehicle crash rates), we had the power to detect a 25% reduction in crash rates or injury rates. For sleep duration, we had the power to detect a 10% difference in weekly sleep hours.

RESULTS

Based on retrospective departmental data for 2008 prior to study start, the control and intervention stations showed no significant difference in the number of firefighters (mean ± SD; 41.5 ± 15.9 vs. 38.4 ± 10.5 firefighters/station for control and intervention stations, respectively; p = .15), the number of equipment (ladder and engine) runs (3193.1 ± 1602.9 vs. 3361.1 ± 1610.2 equipment runs/station for control and intervention stations, respectively; p = .25), or medic runs (3254.8 ± 1369.0 vs. 3478.9 ± 1341.0 medic runs/station for control and intervention stations, respectively; p = .15). Similarly, during the months comprising the 54-week study period, there were no significant differences in the number of equipment (ladder and engine) runs (mean ± SD; 3288.8 ± 1557.2 vs. 3889.5 ± 1583.5 equipment runs/station for control and intervention stations, respectively; p = .49) or medic runs (3506.6 ± 1437.9 vs. 3726.3 ± 1426.0 medic runs/station for control and intervention stations, respectively; p = .18; all paired Student’s t test).

The final cohort of 1189 firefighters represents 98% (1189/1211) of the active firefighters in the participating fire department at the time that the study was implemented. Of the 601 firefighters assigned to stations in the intervention group, 542 (90.2%) participated in one of 52 sleep education sessions provided to firefighters over a 2-week interval, as documented by sign-in attendance sheets (Figure 1). The cohort consisted predominately of white males with a mean age of 43.6 years (Table 1). The intervention group was slightly younger (42.7 vs. 44.4 years; p < .0001) on average as compared to the control group. There were no differences in gender (Fisher’s exact p = .208) or race (Fisher’s exact p = .367).

Figure 1

Participant disposition diagram.

Figure 1

Participant disposition diagram.

Table 1

Demographic Information on the Firefighter Department and Study Cohorts.

 Cohort Control Intervention Screened 
Firefighters, n 1189 588 601 431 
Agea, years mean ± SD (range) 43.6 ± 7.4 (22–72) 44.4 ± 7.4 (25–72) 42.7 ± 7.3 (22–65) 42.7 ± 7.4 (22–62) 
Genderb, n (%) 
 Women 16 (1.3) 5 (0.8) 11 (1.8) 9 (2.1) 
 Men 1173 (98.7) 583 (99.2) 590 (98.2) 422 (97.9) 
Raceb, n (%) 
 White 1081 (90.9) 540 (91.8) 541 (90.0) 398 (92.3) 
 Black 85 (7.1) 40 (6.8) 45 (7.5) 24 (5.6) 
 Hispanic 6 (0.5) 2 (0.3) 4 (0.7) 3 (0.7) 
 Asian 4 (0.3) 3 (0.5) 1 (0.2) 1 (0.2) 
 Native American 4 (0.3) 1 (0.2) 3 (0.5) 1 (0.2) 
 Bi-racial 9 (0.8) 2 (0.3) 7 (1.2) 4 (0.9) 
 Cohort Control Intervention Screened 
Firefighters, n 1189 588 601 431 
Agea, years mean ± SD (range) 43.6 ± 7.4 (22–72) 44.4 ± 7.4 (25–72) 42.7 ± 7.3 (22–65) 42.7 ± 7.4 (22–62) 
Genderb, n (%) 
 Women 16 (1.3) 5 (0.8) 11 (1.8) 9 (2.1) 
 Men 1173 (98.7) 583 (99.2) 590 (98.2) 422 (97.9) 
Raceb, n (%) 
 White 1081 (90.9) 540 (91.8) 541 (90.0) 398 (92.3) 
 Black 85 (7.1) 40 (6.8) 45 (7.5) 24 (5.6) 
 Hispanic 6 (0.5) 2 (0.3) 4 (0.7) 3 (0.7) 
 Asian 4 (0.3) 3 (0.5) 1 (0.2) 1 (0.2) 
 Native American 4 (0.3) 1 (0.2) 3 (0.5) 1 (0.2) 
 Bi-racial 9 (0.8) 2 (0.3) 7 (1.2) 4 (0.9) 

These were the categories provided by the fire department.

aControl versus intervention unpaired t test p < .0001.

bControl versus intervention Fisher’s exact gender p = .207 and race p = .367.

Departmental Database Measures

In terms of exposure to possible injury, firefighters stationed at intervention stations had one or two more 24-hour work shifts on average over the year as compared to firefighters in control stations (101.1 ± 15.2 vs. 99.7 ± 17.3, respectively; p = .120). Intervention firefighters had 4% more run/call deployments over the year on average than the non-intervention group (552.3 ± 286.2 vs. 531.9 ± 269.8, respectively; p = .208).

At the start of the study (baseline; mid-August–September 2009), there were no significant differences in the 24-hour pay periods coded as injury disability per firefighter between the firefighters assigned to the control and intervention stations (control 0.03 ± 0.4 vs. intervention 0.04 ± 0.5; p = .820). Over the entire 12-month study period, however, these departmental payroll records showed that non-intervention station firefighters averaged nearly double the number of 24-hour pay periods coded as injury and disability per firefighter than the intervention group (2.6 ± 8.5 vs. 1.4 ± 5.9 days/firefighter; F(1,1157) = 8.79; p = .003; Figure 2). On average there was a small significant difference in age, however, adding age as a covariate did not alter the outcomes. Seasonal variations in disability were not different between intervention and control stations (Figure 3A) and the benefits of the intervention were sustained over the study period (Figure 3B). The distribution of disability days across firefighters (Figure 3C) illustrates that a small number of firefighters in one group were not biasing the results. The intervention stations had 532/601 (89%) of their firefighters without any recorded disability days and the control stations had 497/588 (85%). After excluding those firefighters without disability days, the result remained significant; mean 16.96 ± 15.14 versus 11.91 ± 13.60 days/firefighter; p = .031, Student’s t test. The medians were 12 and 6, respectively. Furthermore, a general linear model regression to account for overdispersion, assuming a negative binomial distribution, showed a significant difference between groups (Wald χ2 = 4.61; p = .032). There was no significant difference in the average number of sick days per firefighter reported between the two groups (3.2 ± 4.5 vs. 3.1 ± 4.3 days/firefighter; p = .660).

Figure 2

Departmental payroll records revealed that non-intervention firefighters (n = 588) averaged double the number of 24-hour pay periods per firefighter coded as injury/disability than the intervention group (n = 601) (mean ± SE, *p = .003).

Figure 2

Departmental payroll records revealed that non-intervention firefighters (n = 588) averaged double the number of 24-hour pay periods per firefighter coded as injury/disability than the intervention group (n = 601) (mean ± SE, *p = .003).

Figure 3

Panel A shows the average daily number of firefighters on disability per month of the study to illustrate that seasonal variations were not different between intervention and control stations. Panel B shows the cumulative number of total disability days between the intervention and control stations. Panel C shows the distribution of disability days across firefighters for the intervention and control stations.

Figure 3

Panel A shows the average daily number of firefighters on disability per month of the study to illustrate that seasonal variations were not different between intervention and control stations. Panel B shows the cumulative number of total disability days between the intervention and control stations. Panel C shows the distribution of disability days across firefighters for the intervention and control stations.

City government injury reports filed during the study period showed a marginally higher average number of injuries per firefighter in the control group (0.40 ± 0.63 vs. 0.37 ± 0.63 injury reports/firefighter; p = .31). In post hoc analysis accounting for exposure to the intervention, however, firefighters who attended education sessions (160/400) were 24% less likely to file at least one official injury report during the study duration than those firefighters who did not attend (216/413) regardless of randomization (OR [95% CI] 0.76 [0.60, 0.98]; χ2 = 4.56; p = .033). Departmental records of fleet vehicle accidents showed that there were a total of 125 incidents reported during the entire study and there was no significant difference in the rate of incidents between the intervention and control groups (0.11 ± 0.35 incidents/firefighter vs. 0.10 ± 0.31 incidents/firefighter, respectively; p = .87).

Baseline and End-of-Year Surveys

In within-subject paired t test analyses of firefighters from intervention stations who participated in the program, completed both the pre- and post-study survey, and had at least 1 week of work scheduled in the 4 weeks prior to the survey, we did not find any significant differences in self-reported sleep duration or attentional failures while on the telephone, while driving, or while stopped in traffic between the pre- and post-study assessments (Table 2). Similarly, self-reported general health did not change significantly between the pre- and post-study survey (Table 2).

Table 2

Summary of All Outcomes.

Primary department level measures   
Between subjects Intervention Control pb pd 
n 601 588   
Health “sick” days 3.1 (4.3) 3.2 (4.5) 0.659 0.683 
Health “injury/disability” days 1.4 (5.9) 2.6 (8.5) 0.003 0.003 
Safety motor vehicle crashes 0.11 (0.35) 0.10 (0.31) 0.886 0.863 
Safety injuries 0.37 (0.63) 0.40 (0.63) 0.312 0.434 
Performance—response time 
Secondary survey measures 
Within subjects—intervention only Study start Post-study pc n 
Mean total sleep time (h/week)a 44.84 (7.05) 45.98 (6.48) 0.218 62 
Alertness—sleep during meetings (incidents/month)a 0.41 (1.39) 0.30 (0.78) 0.648 27 
Alertness—sleep on phone (incidents/month)a 0.06 (0.38) 0.05 (0.26) 0.708 88 
Alertness—sleep while driving (incidents/month)a 0.22 (0.63) 0.14 (0.65) 0.310 81 
Alertness—sleep while stopped in traffic (incidents/month)a 0.21 (0.72) 0.09 (0.36) 0.159 82 
Health—general health 3.73 (0.82) 3.68 (0.81) 0.458 97 
Family—job satisfaction 
Primary department level measures   
Between subjects Intervention Control pb pd 
n 601 588   
Health “sick” days 3.1 (4.3) 3.2 (4.5) 0.659 0.683 
Health “injury/disability” days 1.4 (5.9) 2.6 (8.5) 0.003 0.003 
Safety motor vehicle crashes 0.11 (0.35) 0.10 (0.31) 0.886 0.863 
Safety injuries 0.37 (0.63) 0.40 (0.63) 0.312 0.434 
Performance—response time 
Secondary survey measures 
Within subjects—intervention only Study start Post-study pc n 
Mean total sleep time (h/week)a 44.84 (7.05) 45.98 (6.48) 0.218 62 
Alertness—sleep during meetings (incidents/month)a 0.41 (1.39) 0.30 (0.78) 0.648 27 
Alertness—sleep on phone (incidents/month)a 0.06 (0.38) 0.05 (0.26) 0.708 88 
Alertness—sleep while driving (incidents/month)a 0.22 (0.63) 0.14 (0.65) 0.310 81 
Alertness—sleep while stopped in traffic (incidents/month)a 0.21 (0.72) 0.09 (0.36) 0.159 82 
Health—general health 3.73 (0.82) 3.68 (0.81) 0.458 97 
Family—job satisfaction 

All values are mean (SD).

aPer firefighter.

bPaired t test.

cPaired t test.

dMixed model adjusted for station pairs.

Sleep Disorders Screening, Diagnosis, and Treatment

In addition to sleep health education, the SHP included the opportunity for firefighters to be screened for sleep disorders. Most firefighters who attended the education sessions chose to complete the voluntary sleep disorders screening questionnaire (431/560; 77%) which may represent a measure of the feasibility and acceptability of the occupational SHP. 179/431 (42%) screened positive for one or more sleep disorders and were referred for clinical follow-up (Table 3). Of the 179 referred, 40 (22%) of these firefighters went on to seek medical evaluation at the partnering sleep disorders clinic and 37/40 (93%) were diagnosed with at least one sleep disorder. Records from firefighters who sought treatment at clinics not partnering with the study were not available. 104/431 firefighters who completed the sleep disorders screening also completed a follow-up end-of-year survey. Thirty-seven of these end-of-year respondents had screened positive for one or more sleep disorders 12 months earlier. Sixteen of the 179 positive screen respondents for whom we have an end-of-year survey indicated they sought clinical evaluation for a sleep disorder, 12 at the partnering clinic and 4 elsewhere.

Table 3

Sleep Disorders Screening Outcomes and Self-reported Demographics.

No. of respondents 431a    
Any sleep disorder 
 Positive 179 (41.5)  Any sleep disorder Obstructive sleep apnea 
 Negative 251 (58.2)  Positive Negative Positive Negative 
 Not known 1 (0.2) n 179 251 135 283 
Obstructive sleep apnea  Ageb 43.2 ± 7.7 (26–63) 42.9 ± 7.2 (22–61) 44.1 ± 7.7 (28–63) 42.5 ± 7.2 (22–61) 
 Positive 135 (31.3) 
 Negative 283 (65.7) Employedb 15.9 ± 8.0 (1–37.5) 15.4 ± 7.4 (1–36.7) 16.8 ± 7.9 (1–37.5) 14.9 ± 7.4 (1–36.7) 
 Not known 13 (3) 
Insomnia  Sexa     
 Positive 33 (7.7) Women 4 (2.2) 4 (1.6) 2 (1.5) 6 (2.1) 
 Negative 391 (90.7) Men 170 (95.0) 239 (95.2) 129 (95.6) 268 (94.7) 
 Not known 7 (1.6) Not known 5 (2.8) 8 (3.2) 4 (3.0) 9 (3.2) 
Restless legs syndrome 
 Positive 15 (3.5) BMIa     
 Negative 400 (92.8) <25 11 (6.1) 45 (25.1) 4 (3.0) 50 (17.7) 
 Not known 16 (3.7) ≥25 and <30 67 (37.4) 150 (83.8) 35 (25.9) 177 (62.5) 
Shift work disorder  ≥20 and <35 76 (42.5) 48 (26.8) 71 (52.6) 49 (17.3) 
 Positive 40 (9.3) ≥35 24 (13.4) 8 (4.5) 24 (17.8) 7 (2.5) 
 Negative 337 (78.2) Not known 1 (0.6) 0 (0.0) 1 (0.7) 0 (0.0) 
 Not known 54 (12.5)      
No. of respondents 431a    
Any sleep disorder 
 Positive 179 (41.5)  Any sleep disorder Obstructive sleep apnea 
 Negative 251 (58.2)  Positive Negative Positive Negative 
 Not known 1 (0.2) n 179 251 135 283 
Obstructive sleep apnea  Ageb 43.2 ± 7.7 (26–63) 42.9 ± 7.2 (22–61) 44.1 ± 7.7 (28–63) 42.5 ± 7.2 (22–61) 
 Positive 135 (31.3) 
 Negative 283 (65.7) Employedb 15.9 ± 8.0 (1–37.5) 15.4 ± 7.4 (1–36.7) 16.8 ± 7.9 (1–37.5) 14.9 ± 7.4 (1–36.7) 
 Not known 13 (3) 
Insomnia  Sexa     
 Positive 33 (7.7) Women 4 (2.2) 4 (1.6) 2 (1.5) 6 (2.1) 
 Negative 391 (90.7) Men 170 (95.0) 239 (95.2) 129 (95.6) 268 (94.7) 
 Not known 7 (1.6) Not known 5 (2.8) 8 (3.2) 4 (3.0) 9 (3.2) 
Restless legs syndrome 
 Positive 15 (3.5) BMIa     
 Negative 400 (92.8) <25 11 (6.1) 45 (25.1) 4 (3.0) 50 (17.7) 
 Not known 16 (3.7) ≥25 and <30 67 (37.4) 150 (83.8) 35 (25.9) 177 (62.5) 
Shift work disorder  ≥20 and <35 76 (42.5) 48 (26.8) 71 (52.6) 49 (17.3) 
 Positive 40 (9.3) ≥35 24 (13.4) 8 (4.5) 24 (17.8) 7 (2.5) 
 Negative 337 (78.2) Not known 1 (0.6) 0 (0.0) 1 (0.7) 0 (0.0) 
 Not known 54 (12.5)      

BMI = body mass index.

an (%).

bYears, mean ± SD (range).

DISCUSSION

This study is the first randomized prospective trial to demonstrate a reduction in risk of injury and disability through implementation of an SHP. In an intention-to-treat analysis, firefighters assigned to intervention stations which participated in education sessions and had the opportunity to complete sleep disorders screening reported 46% fewer disability days on average than those assigned to control stations. Furthermore, when analyzed with respect to intervention exposure, firefighters who attended education sessions were 24% less likely to file at least one official injury report during the study duration than those firefighters who did not attend. Our findings therefore suggest that a firefighter workplace-based wellness program providing sleep health education and sleep disorders screening opportunity can reduce injuries and subsequent work loss due to disability in firefighters.

The finding that 42% (179/431) of the firefighters in this cohort screened positive for one or more sleep disorders is comparable to that reported in a larger nationwide cohort that included these firefighters, in which 37% of firefighters screened positive for a sleep disorder.4 The reasons for the high prevalence include standard risk factors, such as age and body mass index,9,10 but also other factors more highly associated with firefighters, such as hypertension.11 Of concern, in this cohort, we found that at baseline, only 16.3% (22/135) of those who had screened positive for OSA reported being treated, consistent with the low rate of diagnosis of sleep disorders in the general population. It is well established that untreated sleep disorders carry a high risk of cardiovascular disease and metabolic disorders,12 as well as a high risk of drowsy driving crashes13 and therefore programs to identify and treat these disorders are likely to be beneficial to both short-term safety and long-term health, and occupational groups provide a convenient avenue through which diagnosis, treatment, and health can be promoted.

In terms of risk factors for fatigue, in addition to sleepiness associated with untreated sleep disorders, firefighters in this cohort were routinely scheduled to work 24-hour shifts in a recurring sequence of 24 hours on duty, followed by 48-hours off duty. While 24- or even 48-hour duty shifts are common among US fire departments,14 such extended duration shifts have been shown to increase the risk of sleepiness, burnout, injuries, and errors when compared to shorter shifts (≤16 hours) in resident physicians15–19 and first responders such as emergency medical technicians.20 Notwithstanding the high risk of fatigue under such circumstances, we found that implementation of an education and sleep disorders screening program significantly reduced the percentage of firefighters who filed an injury report, such that 12% fewer firefighters who participated in the education intervention reported an injury than those who did not in the subsequent 12 months. Consistent with the proposed health and safety benefits of our program, it also succeeded in reducing reported injuries by 10% per annum, reducing the percentage of firefighters injured by 17% and was accompanied by a 46% decrease in the utilization of paid disability time among firefighters, a substantial impact.

In addition to the considerable human toll of injuries, there are also financial costs. Firefighter injuries in the United States cost an estimated $4.70 to $11.73 billion annually21 (cost estimates are expressed in 2013 dollars, adjusted from original sources for inflation using the Bureau of Labor Statistics Consumer Price Index Inflation Calculator [www.bls.gov/data/inflation_calculator.htm]), which adds substantially to the $43.81 billion in direct expenditures paid each year for fire protection.22 In this department of approximately 1200 active firefighters alone, the reduction in disability day usage observed in the sleep health intervention cohort would translate into annual direct payroll savings of $2.1 million (cost estimates associated with missed disability days were calculated using a national hourly rate times the missed number of hours plus the cost of replacing absentee firefighters at time and a half23) based on the 2012 firefighter national median pay of $21.75 per hour,24 and not including the additional and likely substantial direct medical costs or indirect costs associated with the disability.

We found no other published study that evaluated the impact of a health-related intervention on firefighter injuries using a randomized, prospective, parallel group clinical trial design (ClinicalTrials.gov Identifier: NCT01988129), and none that evaluated a sleep health–related intervention. Our results compare favorably, however, with those from the retrospective Promoting Healthy Lifestyles: Alternative Models’ Effects (PHLAME) study on behavioral change models focusing on nutrition, physical activity, and cardiorespiratory fitness in firefighters.25 In this retrospective analysis, a sustained reduction in both workers’ compensation claims and medical costs was observed in PHLAME-participating departments, compared with matched departments that did not participate.26 Similarly, a prospective, but not randomized, study of the implementation of a Physician-Organized Wellness Regime reported a 40% reduction in recordable injuries during the first year.27 Not all workplace-based intervention programs are successful in reducing injuries, however. Poston et al. recently reported on data collected from 10 purposively sampled fire departments that implemented health promotion measures consistent with the Wellness Fitness Initiative developed by the International Association of Fire Firefighters (IAFF) and the International Association of Fire Chiefs (IAFC) and compared them with 10 fire departments that did not implement such programs. While firefighters in the intervention departments were healthier in a number of dimensions as compared with those in the standard departments, Poston et al. reported that firefighters in departments that implemented the IAFF/IAFC–compliant wellness programs were 63% more likely to have an injury of sufficient severity to report to workers’ compensation.28

While fatigue risk management programs are commonly implemented in other occupational settings, there is limited empirical evidence on their impact. A fatigue management program for nurses, including sleep health education and encouragement of napping during breaks or meal times, resulted in significant improvement in self-reported sleep duration, sleep quality, and error prevention.29 Most recently, a large-scale occupational program in truckers showed that truck drivers with untreated OSA had a five times higher rate of preventable crashes than those truck drivers who were not diagnosed with OSA,30 suggesting that a sleep disorders screening and treatment program31 could have a high impact on driver and public safety.

There were several limitations to our study. Although randomized, a double-blind design was not possible in this field study; however, as the outcomes were derived from standard departmental records rather than study participants, we believe it unlikely that knowledge of group assignment influenced the incidence of long-term disability between the groups. While there were no significant differences in outcomes at study initiation, a longer period of baseline data collection prior to study initiation may have alleviated concerns about any inherent differences between groups in this regard. Second, as expected due to the practice of trading shifts and routine temporary re- assignments based on operational needs, some mixing occurred between firefighters assigned to the randomized intervention and control stations. As a result, 59 firefighters assigned to intervention stations did not attend an education session and 19 firefighters assigned to control stations did attend an education session. We do not believe such mixing impacted our conclusions, as (1) the 78 affected firefighters represent only a small fraction (6.4%) of the cohort and (2) we subsequently accounted for such temporary station reassignments by basing our analyses on those firefighters who attended an education session versus those who did not, as shown in Figure 1. An additional limitation is the low response rate to the end-of-year survey which reduced our power to detect changes in self-reported sleep and behavior in those who participated in the program. Given that none of these outcomes were significant, however, any bias due to the low numbers may have resulted in underestimating the impact of the program. It would have been desirable, in retrospect, to collect more detailed data on the potential mechanism by which the program impacted the injury rate, for example through changes in sleep behavior tracked by sleep logs or actigraphy. We were also unable to measure the impact of the program on firefighters’ and families’ job satisfaction, or firefighter performance, due to reallocation of resources or finding that our approach was impractical in an operational setting. Further work is therefore needed to examine the broader impact of such a program beyond disability days and injuries. Finally, we have limited knowledge of sleep disorder diagnosis, treatment, and compliance after an individual firefighter was advised of screening positive for a sleep disorder. Many, but not all, firefighters provided permission to access medical records at the accredited sleep clinic partnering with our study, and some firefighters chose to follow-up at clinics not connected with our study. Further research is also required to replicate and expand these results in other fire departments to assess the generalizability of the findings, including efforts to determine the relative contribution and mechanism of action of each component of the program to observed safety improvements and how to increase adherence to sleep disorders treatment among firefighters after screening positive for risk of a sleep disorder. One possible way to increase adherence would be to include sleep disorders screening in mandatory occupational health physical exams where they already exist. We have also recently shown that an SHP is feasible and effective in increasing firefighter knowledge of sleep health and improving sleep health behaviors, whether delivered by experts, in a “train-the-trainer” format or online.32

Fatigue-related accidents and injuries are often underestimated yet contribute substantially to workplace health and safety risk. Sleep and fatigue-related factors are therefore a promising avenue to reduce avoidable workplace injuries. Our study demonstrates that a workplace sleep health education and sleep disorders risk screening can reduce injuries and disability in firefighters and has important public policy implications for the fire service and other occupational groups. A targeted sleep health intervention program thus holds promise for reducing the substantial morbidity and mortality associated with firefighter injury and should be considered as an addition to the Wellness Fitness Initiative standards promulgated by the IAFF/IAFC in order to reduce firefighter injury and associated disability costs.

FUNDING

This work was supported by Federal Emergency Management Agency (FEMA) Assistance for Firefighters grants EMW-2007-FP-02197 and EMW-2008-FP-02566; National Institute of Occupational Safety and Health grant 1R01OH010300-02; National Institutes of Health National Heart, Lung, and Blood Institute grant R01NS054277; and an endowed professorship provided to Harvard Medical School by Cephalon, Inc. The contents of this manuscript are solely the responsibility of the authors and do not necessarily represent the official views of FEMA or the Assistance to Firefighters Grant Program, NIOSH, NIH/NHBLI, or Cephalon, Inc. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

DISCLOSURE STATEMENT

JPS and CSB reports no conflicts of interest. LKB has previously received research support from Cephalon, NFL charities, Sysco, and San Francisco Bar Pilots. She has received consulting/lecture fees or served as a board member for Alertness Solution, Ceridian, Davis Joint Unified School Board, San Jose State University Foundation, Puget Sound Pilots, Sygma, and Torvec. SMWR reports that he has served as a consultant through his institution to Vanda Pharmaceuticals, Philips Respironics, EdanSafe, National Transport Commission, Rail, Bus and Train Union, Australian Workers’ Union, Tontine Group, Transport Accident Commission, Meda Consumer Healthcare, and New South Wales Department of Education & Communities, and has through his institution received research grants and from Vanda Pharmaceuticals, Philips Respironics, and Rio Tinto and reimbursements for conference travel expenses from Vanda Pharmaceuticals. He serves as a consultant to, and is a Program Leader for, the Cooperative Research Centre for Alertness, Safety and Productivity. His institution has received equipment donations or other support from Optalert, Compumedics, Philips Lighting, and Tyco Healthcare. He is a Director of the Sleep Health Foundation and is a Past President of the Australasian Sleep Association. He has also served as an expert witness and/or consultant to shift work organizations and in legal cases involving sleep loss. In the past 3 years, CAC has received consulting fees from or served as a paid member of scientific advisory boards for Amazon.com, Inc.; A2Z Development Center, Inc.; Bose Corporation; Boston Celtics; Boston Red Sox; Cleveland Browns; Columbia River Bar Pilots; Institute of Digital Media and Child Development; Jazz Pharmaceuticals, Inc.; Merck Sharpe and Dohme; NBA Coaches Association; Purdue Pharma; Quest Diagnostics; RMEI, LLC; Samsung Electronics; Teva Pharmaceutical Industries Ltd; Novartis; Synchrony; and Vanda Pharmaceuticals, Inc. He owns an equity interest in Vanda Pharmaceuticals, Inc. and Zurich Insurance. He has received royalties from McGraw Hill, Penguin Press/Houghton Mifflin Harcourt, and from Philips Respironics, Inc., for the Actiwatch 2 and Actiwatch Spectrum devices; his interests were reviewed and are managed by Brigham & Women’s Hospital and Partners HealthCare in accordance with their conflict of interest policies. The Harvard Medical School Division of Sleep Medicine and/or the Brigham and Women’s Hospital Division of Sleep and Circadian Disorders, which CAC directs, have received unrestricted research and educational gifts and endowment funds from Boehringer Ingelheim Pharmaceuticals, Inc.; Bose; Cephalon, Inc.; GlaxoSmithKline; Jazz Pharmaceuticals; National Football League Charities; Mary Ann and Stanley Snider through Combined Jewish Philanthropies; Merck & Co., Inc.; Optum; Pfizer; Proctor and Gamble; Philips Respironics, Inc.; ResMed; ResMed Foundation; the San Francisco Bar Pilots; Sanofi-Aventis, Inc.; Sealy, Inc.; Sepracor, Inc.; Simmons, Spring Aire; Sysco; Takeda Pharmaceuticals; Tempur-Pedic; Walmart; and Vanda Pharmaceuticals, Inc.; and has received Educational Grant funding from Apria Healthcare; Cephalon, Inc.; Jazz Pharmaceuticals; Philips Respironics; Takeda Pharmaceuticals; ResMed Foundation; Sanofi-Aventis, Inc.; Sepracor, Inc.; Teva Pharmaceutical Industries, Ltd; and Wake Up Narcolepsy. CAC collaborates on research that was supported by Schneider, Inc., the MacArthur Foundation, and the Sloan Foundation (through the University of Minnesota, Morris, MN); by the Liberty Mutual Insurance Company (through the Liberty Mutual Research Institute for Safety, Hopkinton, MA); by RANNIS—the Icelandic Research Fund, the Icelandic Cancer Society, the Icelandic Heart Association, and the Icelandic Parliament (through the University of Iceland, Reykjavik, Iceland); the Prostate Cancer Foundation (through the Brigham and Women’s Hospital); by the American College of Cardiology Foundation (through the Mid America Heart Institute, Kansas City, MO); by the Leakey Foundation (through Yale University, New Haven, CT); by NEC, Samsung, and more than 70 other commercial sponsors of the MIT Media Laboratory (through the Massachusetts Institute of Technology, Cambridge, MA); by Réseau Vision du Québec, Réseau de Bioimagerie du Québec, and Fonds de Recherche en Santé du Québec (through the University of Montréal and Hôpital du Sacré-Coeur de Montréal, both in Québec, Canada); by the Belgian Fonds de Recherche en Santé (through the University of Liège, Belgium); by the Wellcome Trust (through the University of Surrey, UK); by the La-Roche Foundation and the Novartis Foundation (through the Swiss Federal Institute of Technology, Lausanne, Switzerland); by the government of Australia and multiple academic and commercial sponsors (through the Cooperative Research Centre for Alertness, Safety and Productivity, Monash University, Melbourne, Australia); and by the National Sleep Foundation, which has multiple commercial sponsors. He holds a number of process patents in the field of sleep/circadian rhythms (eg, photic resetting of the human circadian pacemaker). Since 1985, he has also served as an expert witness on various legal cases related to sleep and circadian rhythms, including matters related to Bombardier, Citgo, HG Energy, Michael Jackson’s mother and children, Purdue Pharma, StricLan LLC, Valero, and matters related to commercial drivers employed by Celadon, Crete Carrier Corporation, FedEx, United Parcel Service, and other commercial carriers. In the past 3 years, SWL has received consulting fees from the Atlanta Hawks; Blackrock; Carbon Limiting Technologies Ltd (on behalf of PhotonStar LED); Cowen & Co; Endurant Capital Management; Far West Capital Management; Fidelity; Frankel Group; Impax Laboratories; Kearney Venture Partners; Lazard Capital Markets; Naturebright; New Horizon Capital; Perceptive Advisors; Polar Capital; ResearchWorks Inc.; Serrado Capital; and Wyvern Funds; has current consulting contracts with Akili Interactive; Delos Living LLC; Environmental Light Sciences, LLC; Focal Point LLC; Headwaters Inc.; Hintsa Performance AG; Light Cognitive; OpTerra Energy Services Inc.; Pegasus Capital Advisors LP; PlanLED; and Wyle Integrated Science and Engineering; has received unrestricted equipment gifts from Bioilluminations LLC; Bionetics Corporation; and Philips Lighting; advance author payment and royalties from Oxford University Press; payment for editing a textbook section from Elsevier; honoraria from the National Sleep Foundation; and for an article in the Wall Street Journal; honoraria plus travel, accommodation, or meals for invited seminars, conference presentations, or teaching from Brookline Adult Education; Brown University; Estee Lauder; Harvard University (CME); MediCom Worldwide, Inc. (CME); travel, accommodation, and/or meals only (no honoraria) for invited seminars, conference presentations, or teaching from 14th Annual Tennessee Perfusion Conference; American Society for Photobiology; Cantifix; Connecticut Business & Industry Association Health and Safety Conference; Emergency Services Steering Committee; FASEB; Harvard University; Hintsa Performance AG; Illuminating Engineering Society; Lightfair; Massachusetts General Hospital; Midwest Lighting Institute; New England College of Occupational and Environmental Medicine; Ontario Association of Fire Chiefs; Rio Tinto; UMass Memorial; University of Manchester; ongoing investigator-initiated research grants from Biological Illuminations LLC; and Respironics Inc.; completed service agreements with Rio Tinto Iron Ore and Vanda Pharmaceuticals Inc.; three completed sponsored initiated clinical research contracts with Vanda Pharmaceuticals Inc.; and one completed investigator-initiated research grant from Vanda Pharmaceuticals Inc. SWL also holds a process patent for the use of short-wavelength light for resetting the human circadian pacemaker and improving alertness and performance which is assigned to the Brigham and Women’s Hospital per Hospital policy. SWL has also served as a paid expert on behalf of five public bodies and groups in cases related to sleep, circadian rhythms, and work hours. SWL also serves as a Program Leader in the Cooperative Research Centre for Alertness, Safety and Productivity.

ACKNOWLEDGMENTS

We would like to thank Lauren Brogan, Amy Hallal, KC Root, Salim Qadri, and Michael Shreeve from the Division of Sleep and Circadian Disorders, Brigham and Women’s Hospital for their help in conducting the study. We are grateful to the Columbus OH Division of Fire Administration and the IAFF Local 67 and all of the firefighters who participated in this research program. We are especially grateful to Jack Reall, Ned Petttus Jr., Tommie Page, and Michael Sachs for agreeing to be videoed as part of our training program. Special thanks for acquisition and interpretation of departmental records to Jack Levy, Holly Hanf, Vickie Atkins Loeser, and Janet Hedges. Thank you also to the International Association of Fire Chiefs Safety, Health and Survival Section for providing feedback on the baseline health screening survey questions and to David Whiting, Mike Wajda, Bill Ehrgood, Barbara Salas, and Paul Parrish, who served as members of the Harvard Firefighter Research Advisory Committee.

REFERENCES

1.
Barger
LK
Lockley
SW
Rajaratnam
SM
Landrigan
CP
.
Neurobehavioral, health, and safety consequences associated with shift work in safety-sensitive professions
.
Curr Neurol Neurosci Rep
 .
2009
;
9
(
2
):
155
164
.
2.
Rajaratnam
SMW
Barger
LK
Lockley
SW
et al
Sleep disorders, health, and safety in police officers
.
JAMA
 .
2011
;
306
(
23
):
2567
2578
.
3.
Colten
HR
Alteveogt
BM
;
Institute of Medicine (US) Committee on Sleep Medicine and Research
.
Sleep Disorders and Sleep Deprivation: An Unmet Public Health Problem
 .
Washington, DC
:
Institute of Medicine of the National Academies, The National Academies Press
;
2006
.
4.
Barger
LK
Rajaratnam
SM
Wang
W
et al
Common sleep disorders increase risk of motor vehicle crashes and adverse health outcomes in firefighters
.
J Clin Sleep Med
 .
2015
;
11
(
3
):
233
240
.
5.
Netzer
NC
Stoohs
RA
Netzer
CM
Clark
K
Strohl
KP
.
Using the Berlin Questionnaire to identify patients at risk for the sleep apnea syndrome
.
Ann Intern Med
 .
1999
;
131
(
7
):
485
491
.
6.
Soldatos
CR
Dikeos
DG
Paparrigopoulos
TJ
.
The diagnostic validity of the Athens Insomnia Scale
.
J Psychosom Res
 .
2003
;
55
(
3
):
263
267
.
7.
Allen
RP
Walters
AlS
Montplaisir
J
et al
Restless legs syndrome prevalence and impact: REST general population study
.
Arch Intern Med
 .
2005
;
165
(
11
):
1286
1292
.
8.
American Academy of Sleep Medicine
.
International Classification of Sleep Disorders, Revised: Diagnostic and Coding Manual
 .
2
nd ed.
Westchester, IL
:
American Academy of Sleep Medicine
;
2005
.
9.
Bixler
EO
Vgontzas
AN
Ten Have
T
Tyson
K
Kales
A
.
Effects of age on sleep apnea in men: I. Prevalence and severity
.
Am J Respir Crit Care Med
 .
1998
;
157
(
1
):
144
148
.
10.
Young
T
Palta
M
Dempsey
J
Skatrud
J
Weber
S
Badr
S
.
The occurrence of sleep-disordered breathing among middle-aged adults
.
N Engl J Med
 .
1993
;
328
(
17
):
1230
1235
.
11.
Banes
CJ
.
Firefighters’ cardiovascular risk behaviors
.
Workplace Health Saf
 .
2014
;
62
(
1
):
27
34
.
12.
Rajagopalan
N
.
Obstructive sleep apnea: not just a sleep disorder
.
J Postgrad Med
 .
2011
;
57
(
2
):
168
175
.
13.
Tregear
S
Reston
J
Schoelles
K
Phillips
B
.
Obstructive sleep apnea and risk of motor vehicle crash: systematic review and meta-analysis
.
J Clin Sleep Med
 .
2009
;
5
(
6
):
573
581
.
14.
Elliot
DL
Kuehl
KS.
Effects of Sleep Deprivation on Fire Fighters and EMS Responders
 .
International Association of Fire Chiefs and the United States Fire Administration. Portland, OR: Division of Health Promotion & Sports Medicine, Oregon Health & Science University;
2007
.
15.
Fahrenkopf
AM
Sectish
TC
Barger
LK
et al
Rates of medication errors among depressed and burnt out residents: prospective cohort study
.
BMJ
 .
2008
;
336
(
7642
):
488
491
.
16.
Ayas
NT
Barger
LK
Cade
BE
et al
Extended work duration and the risk of self-reported percutaneous injuries in interns
.
JAMA
 .
2006
;
296
(
9
):
1055
1062
.
17.
Barger
LK
Ayas
NT
Cade
BE
et al
Impact of extended-duration shifts on medical errors, adverse events, and attentional failures
.
PLoS Med
 .
2006
;
3
(
12
):
e487
.
18.
Barger
LK
Cade
BE
Ayas
NT
et al
Extended work shifts and the risk of motor vehicle crashes among interns
.
N Engl J Med
 .
2005
;
352
(
2
):
125
134
.
19.
Landrigan
CP
Rothschild
JM
Cronin
JW
et al
Effect of reducing interns’ work hours on serious medical errors in intensive care units
.
N Engl J Med
 .
2004
;
351
(
18
):
1838
1848
.
20.
Boudreaux
E
Mandry
C
Brantley
PJ
.
Emergency medical technician schedule modification: impact and implications during short- and long-term follow-up
.
Acad Emerg Med
 .
1998
;
5
(
2
):
128
133
.
21.
TriData Corporation
.
Economic consequences of firefighter injuries and their prevention. Final Report (NIST GCR 05–874)
 .
Gaithersburg, MD
:
For U.S, Department of Commerce, Building and Fire Research Laboratory, National Institute of Standards and Technology
. http://www.fire.nist.gov/bfrlpubs/NIST_GCR_05_874.pdf. Accessed September 4, 2014.
22.
Karter
MJ
Jr
Stein
GP.
US Fire Department Profile 2012
 . In:
National Fire Protection Association
, ed.
Quincy, MA
:
National Fire Protection Association
;
2013
.
23.
Poston
WS
Jitnarin
N
Haddock
CK
Jahnke
SA
Tuley
BC
.
Obesity and injury-related absenteeism in a population-based firefighter cohort
.
Obesity
 .
2011
;
19
(
10
):
2076
2081
.
24.
Bureau of Labor Statistics US Department of Labor
.
Occupational Outlook Handbook, 2014–15 Edition, Firefighters Summary
. http://www.bls.gov/ooh/protective-service/firefighters.htm. Accessed
September 4, 2014
.
25.
Moe
EL
Elliot
DL
Goldberg
L
et al
Promoting healthy lifestyles: alternative models’ effects (PHLAME)
.
Health Educ Res
 .
2002
;
17
(
5
):
586
596
.
26.
Kuehl
KS
Elliot
DL
Goldberg
L
Moe
EL
Perrier
E
Smith
J
.
Economic benefit of the PHLAME wellness programme on firefighter injury
.
Occup Med (Lond)
 .
2013
;
63
(
3
):
203
209
.
27.
Leffer
M
Grizzell
T
.
Implementation of a physician-organized wellness regime (POWR) enforcing the 2007 NFPA standard 1582: injury rate reduction and associated cost savings
.
J Occup Environ Med
 .
2010
;
52
(
3
):
336
339
.
28.
Poston
WS
Haddock
CK
Jahnke
SA
Jitnarin
N
Day
RS
.
An examination of the benefits of health promotion programs for the national fire service
.
BMC Public Health
 .
2013
;
13
:
805
.
29.
Scott
LD
Hofmeister
N
Rogness
N
Rogers
AE
.
An interventional approach for patient and nurse safety: a fatigue countermeasures feasibility study
.
Nurs Res
 .
2010
;
59
(
4
):
250
258
.
30.
Burks
SV
Anderson
JE
Bombyk
M
et al
Nonadherence with employer-mandated sleep apnea treatment and increased risk of serious truck crashes
.
Sleep
 .
2016
;
39
(
5
):
967
975
.
31.
Berger
M
Varvarigou
V
Rielly
A
et al
Employer-mandated sleep apnea screening and diagnosis in commercial drivers
.
J Occup Environ Med
 .
2012
;
54
(
8
):
1017
1025
.
32.
Barger
LK
O’Brien
CS
Rajaratnam
SM
et al
Implementing a sleep health education and sleep disorders screening program in fire departments: a comparison of methodology
.
J Occup Environ Med
 .
2016
;
58
(
6
):
601
609
.

Author notes

Address correspondence to: Steven W. Lockley, PhD, Division of Sleep and Circadian Disorders, Brigham and Women’s Hospital, 221 Longwood Avenue, BLI 438, Boston, MA 02115, USA. Telephone: +617-732-4013; Fax: +617-732-4015; Email: slockley@hms.harvard.edu