INTRODUCTION
Work-related injuries impose a significant health and economic burden and contribute to lost productivity.
1 In 2013, according to Statistics Canada, 16% of Canadians aged 15 to 74 years (4.2 million people) sustained injuries in the previous 12 months that limited their normal activities.
2 For 14.5% of those who were injured, their most serious injury occurred in the workplace. The three most common types of injuries sustained at work in 2013 were sprains or strains (49.9%); cuts, punctures, or bites (19.2%); and fractures (8.7%).
2 Firefighters have high rates of work-related injuries because of the high physical demands of their job tasks.
3,4 In 2015, the Ontario Workplace Safety and Insurance Board (WSIB) reported that police officers and firefighters had the province's second-highest rate of allowed lost time claims (9.4%).
5 Soft-tissue injuries due to overexertion were the most frequently reported injury type and mechanism.
5–8Firefighters perform high-risk, physically demanding tasks and are simultaneously exposed to environmental factors that contribute to higher prevalence of injury and disease.
3,4,8,9 Tasks associated with firefighting, such as heavy lifting, unsafe work postures,
4,6 and body motion,
9,10 have been associated with higher injury rates in firefighters. Another study concluded that the high physical load associated with firefighting tasks predisposes firefighters to higher risk of injury.
8The role of firefighters in providing community safety service requires intermittent periods of peak physical activity, a pattern which places them at increased risk for musculoskeletal disorders (MSDs) and cardiovascular injury compared to other occupations.
3,4,11 Understanding the prevalence, distribution, and severity of MSDs among firefighters is necessary to identify those at high risk of MSDs and to develop specific and targeted injury prevention strategies for them. Previous studies have shown the effectiveness of context-specific, targeted approaches in preventing work injuries.
12–14Personal factors (e.g., age, gender, height, weight) and length of service (LOS) are important in identifying firefighters at risk of MSDs.
15,16 For example, a previous study that examined work injury claims filed by a sample of 171 US firefighters over a 12-year period found that personal factors (age, gender, and marital status) significantly predicted injury severity.
3 Injuries lasted longer in firefighters who were older or had shorter LOS.
3 Married female firefighters had shorter injury duration than unmarried male, unmarried female, and married male firefighters.
3Therefore, our primary objective in this study is to describe the prevalence of self-reported MSDs in active-duty firefighters from a southwestern Ontario fire service and how the prevalence and severity of MSDs vary by age, length of service, height, and weight. The secondary objective is to determine whether age, length of service, height, and weight vary across different regional MSDs (neck, upper limb, lower limb, back) and concurrent MSDs (>1 of the previous regional MSDs or no MSDs).
METHODS
Study design, setting, and population
This is a cross-sectional study of 294 active-duty firefighters recruited from the City of Hamilton Fire Service and enrolled in the Firefighter Injury Reduction Enterprise: Wellness Enabled Life & Livelihood (FIREWELL) study. The total number of firefighters on roster in Hamilton was 471. Those not evaluated were not on active duty during the data collection window, were not responsive to contact, or did not consent to participate. Upon enrolment into the FIREWELL study (January 2012–February 2013), all participants completed a series of self-report measures about their health. The study protocol was approved by the McMaster Research Ethics Board. All participants provided written informed consent before being enrolled in the study.
Data collection
Participants completed a series of self-report measures, including a questionnaire we developed for the study to elicit personal factors, demographics and anthropometry (age, gender, height, and weight), duration of professional firefighting service, and comorbidities. The majority of participants completed paper-based questionnaires administered by an experienced clinician and research staff. A small number completed web-based questionnaires administered via an open-source survey tool, LimeSurvey (LimeSurvey, Hamburg, Germany).
To identify the prevalence of self-reported MSDs, all 294 participants completed a pain body diagram on the Iconic Pain Assessment Tool (IPAT),
17 a self-report instrument that combines word descriptors and representative images (icons) to assess pain quality, intensity, and location. Participants were instructed to mark the location(s) of the pain they felt on their body within the past week. If the participant's body diagram showed neck, back, lower-limb, or upper-limb pain within the past week, the marking of a painful site was used to define the presence and location of MSDs. Participants who reported pain at one or more of these sites within the past week completed the relevant region-specific survey(s): Neck Disability Index (NDI),
18,19 Roland Morris Disability Questionnaire (RMDQ),
20,21 Lower Extremity Functional Scale (LEFS),
22 or Short Form of Disabilities of the Arm, Shoulder and Hand (QuickDASH).
23The NDI
18 measures disability related to neck pain. A systematic review found that the majority of studies on the NDI reported high reliability (ICC=0.50–0.98) and validity (correlations with other scales reported >0.70).
19 A higher NDI score indicates more neck-related disability.
The RMDQ is a self-report tool designed to assess physical disability caused by low back pain
20 and has excellent reliability (internal consistency=0.87; ICC = 0.79).
21 A higher RMDQ score indicates more back-related disability.
The LEFS is a self-report scale that measures lower-extremity function and has excellent reliability (internal consistency=0.96; ICC=0.94).
22 A lower LEFS score indicates more disability related to the lower extremity.
The QuickDASH is a reliable (ICC≥0.94) and valid (correlations with pain and function measures ≥ 0.64) measure of physical function and symptoms related to upper-limb MSDs.
23 A higher QuickDASH score indicate more disability related to the upper extremity.
We triangulated data using the pain body diagram (IPAT) and the region-specific self-report disability measures (NDI, RMDQ, LEFS, and/or QuickDASH).
Statistical analysis
The survey data were entered by a research assistant, and data quality was cross-checked by the study coordinator. There were no missing data. We then determined descriptive statistics for personal factors (age, gender, height, and weight) and LOS for the study sample. We confirmed assumptions of statistical analysis, including normality of distribution, using the Shapiro–Wilk test, then calculated central tendencies and measures of dispersion calculated for the region-specific surveys (NDI, RMDQ, LEFS, and QuickDASH). To enable a stratified analysis of MSD prevalence related to age, weight, height, and LOS, we dichotomized the study sample around the mean of each of these four variables. The scores of each region-specific survey were then analyzed separately among each of the four variables between two groups using Mann–Whitney tests. Using logistic regression, we compared the frequencies of regional MSDs among the two groups for each of the four variables. We then further divided the study sample, based on reported MSDs, into six groups of firefighters with MSDs in (1) neck, (2) upper limb, (3) lower limb, or (4) back; (5) more than one MSD; and (6) no MSDs. Differences in personal factors (age, height, weight), and LOS across the 6 groups of self-reported MSDs were determined using ANOVA. All analyses were conducted using Stata 13.1 software (StataCorp, College Station, TX).
RESULTS
Demographic characteristics of the firefighter sample and the frequencies of self-reported MSDs are shown in
Table 1. The highest prevalence of self-reported MSDs was in the upper (44%) and lower (45%) limb regions; 20% of firefighters reported experiencing neck MSDs, and 33% reported experiencing back MSDs.
Table 2 shows medians and inter-quartile ranges (IQRs) of the region-specific self-report questionnaires (NDI, RMDQ, LEFS, and QuickDASH), stratified between firefighters ≥42 years old (older group) and those <42 years old (younger group). Older firefighters reported significantly more severe lower-extremity disability (LEFS: 71 (65, 77) vs. 75 (69.5, 78.5),
p=0.03) and more severe back disability (RMDQ: 2 (1, 3) vs. 1 (0, 2),
p=0.04) than the younger group.
Medians and IQRs of the regional self-report questionnaires stratified based on LOS (≥15 and <15 years of firefighting service) are reported in
Table 3. Firefighters with ≥15 years of firefighting service had significantly more lower-extremity disability (LEFS: 71 (64, 77) vs. 76 (70, 79),
p=0.0005) than those with shorter LOS.
Analysis of self-report MSD questionnaire results in terms of height and weight showed no difference in severity of regional disabilities between the upper and lower 50th percentiles.
Table 4 shows the frequencies of MSDs reported by anatomic region and the 50th percentiles of age, LOS, height, and weight. Although the frequency of back MSDs did not differ between age groups, it did differ between longer and shorter LOS groups (more frequent with greater LOS). The frequencies of MSDs reported in all other regions (neck, lower limb, and upper limb) were significantly different among older and younger firefighters, and between those with shorter and longer LOS. However, frequencies of regional MSDs were not significantly different between the 50th percentile height and weight groups, except for a higher frequency of lower-limb disorders among the higher weight groups.
As
Table 5 shows, firefighters with more than one region affected by a MSD were significantly older than those who reported no MSDs (
F(5,285)=3.3,
p=0.002). Firefighters with more than one regional MSD had significantly longer firefighting service than those with an upper-limb only (
F(5,282)=6.8,
p=0.021) and those reporting no MSDs (
F(5,282)=5.4,
p=0.001). There was no difference between firefighters with no, one, or more than one regional MSD in terms of either weight (
F(5,287)=1.7,
p=0.14) or height (
F(5,249)=0.9,
p=0.48).
Table 5 also shows the distribution of MSDs across genders.
DISCUSSION
Our findings identify a high prevalence of MSDs among firefighters and reveal that these are most common in the upper (44%) and lower limbs (45%); 20% of reported MSDs were in the neck and 33% in the back. These findings are similar to those of previous studies that have also identified a high prevalence of MSDs among firefighters. For example, Bos et al.
8 studied the prevalence of health complaints among Dutch firefighters and reported similar prevalence of neck (16%) and back (32%) complaints, although they found lower reported prevalence of upper-limb (24%) and lower-limb (25%) complaints.
8 This difference may be explained by differences between their study sample and ours, such as younger mean age (39.2 vs. 42.6) and shorter LOS (13.2 vs. 15.1).
8 Poplin et al.
24 also examined the distribution of injury type among firefighters in the Tucson Fire Department in the southwestern United States. Similar to our findings, the most prevalent MSDs were in the lower limb (37%), followed by upper-limb (23%), back (22%), and neck injuries (11%).
24 A recent Canadian study by Frost et al.
25 found that the back (32%), knees (23%), and shoulders (15%) were the areas most affected by musculoskeletal injuries reported by the Calgary Fire Department in western Canada. Data collection in our study was based on a body diagram, whereas the Calgary study was based on work injury claims. It would be reasonable to expect higher rates from our approach, as not all injuries involve a claim, thus our data may have included less severe MSK problems that did not lead to work loss. However, we may also have missed injuries so severe that the firefighter was unable to participate in the research as a result. Our estimates reflect the prevalence of painful sites perceived by firefighters as being MSK-related at one point in time, whereas claims data represent incident claims over a specific period – one year, in the Calgary case. The data from Calgary show that the distribution of injury for strains/sprains is different and higher than that of traumatic injuries such as burns, fractures, and cuts. We did not define the diagnosis or mechanism of injury, and therefore cannot know why specific sites were reported as painful.
Women are often excluded or absent from firefighter studies because of the low numbers of female firefighters. Although our sample included only 8 women, 5 of them (63%) reported multiple painful MSK sites; by comparison, of 283 male participants, 120 (42%) reported multiple painful sites. Although our female prevalence estimate must be considered unstable because of the small sample size, our data do suggest a large MSK burden in women.
We did not find any association between height or weight and MSK complaints. Our models used height and weight separately because they provide more info than body mass index (BMI) – which can have the same value for many combinations of height and weight – and because they have independent effects on biomechanics/kinematics and, hence, on MSK risks such as joint loading and task reach. We assume that firefighters may be more homogeneous on these factors than the general population, since people in more extreme height and weight categories would be unlikely to apply for or pass firefighter screening.
Similar to our results, previous studies have also indicated that age affects aspects of injury or task performance in firefighters. Liao et al. concluded that age is a significant predictor of the duration of injuries in 171 firefighters.
3 Sinden et al. found that performance on the hose drag task was adversely affected by greater age (
r=0.34), and that greater LOS was moderately associated with slower task performance (
r=0.32).
26 Michaelides et al. also identified a significant, moderate correlation between increasing age and slower performance of a battery of tasks, including stair climbing, rolled hose lift and move, Keiser sled, hose pull and hydrant hookup, mannequin (82 kg) drag, and charged hose advance tasks.
27 The fact that firefighters acquire a greater burden of MSDs with increased age and LOS makes it difficult to distinguish age-related changes from the effects of repetitive injuries or cumulative strains on the MSK system. The slower performance observed among older firefighters during strenuous activities must be at least partially due to the cumulative effects of the MSK problems identified in this study and others.
Our study describes the prevalence of MSDs in a group of Canadian firefighters and compares the severity and prevalence of MSDs among younger and older firefighters. The results confirm the need for primary and secondary injury-prevention programs for firefighters, since rates of MSK problems are high and appear to increase over a firefighter's career. It is difficult to determine the impact of these problems on firefighters based on the data we collected, since generic patient-reported outcome measures are generally designed for patient populations, not for workers in highly physically demanding roles. Our participants' scores on self-reported outcome measures suggest relatively low levels of disability, consistent with the fact that all participants were fully active in a physically demanding occupation at the time of the study. Because our sample was drawn from firefighters on active duty, we may have missed counting those with severe MSK injuries who had gone on to long-term disability, which could mean that we underestimated mean severity. A firefighter-specific work limitations questionnaire might have captured more information about work-relevant disability, but to date no measure has been validated for this purpose.
Our findings suggest that upper-limb MSDs may occur early in firefighters' careers, since this subgroup had the lowest mean age and one of the shortest LOS. Earlier onset may indicate more traumatic types of injuries, such as traumatic rotator-cuff tears. Understanding how firefighters' tasks are performed with respect to upper-limb loading and movement may provide insights into how tasks can be modified to prevent upper-extremity injury. Screening for upper-limb MSDs may assist with early intervention and treatment. It is possible that upper-extremity impairments place firefighters at risk for future MSK injuries – an issue that should be explored in longitudinal studies.
In our study, the mean age of firefighters with more than one regional MSD was 45.4 (8.9) years, and frequencies of neck, upper-limb, and lower-limb MSDs in firefighters age 42 years or older were higher than among those younger than 42 years. This finding suggests that screening of MSDs should start early to detect earlier traumatic injuries of the upper extremity, such as rotator-cuff problems, and more gradual onset or cumulative injuries that occur over time. As expected, age and LOS are correlated (r=0.84). However, rates of reported back MSDs were not significantly different between age groups but were significantly different when LOS was considered. This implies that work exposure time (cumulative injury or system stresses), rather than age-related changes in the MSK system, may be the more direct driving force in back pain.
The way in which firefighting tasks are performed, equipment, and external environment may be areas where risk reduction could be explored to reduce the prevalence of MSDs. A study by Kim et al.
28 implemented a multi-faceted back injury prevention program, including education about epidemiology of low back pain, anatomy and biomechanics, back safety, correct lifting and handling techniques, correct posture, nutritional advice, stress management, exercises, and pain management. The program was implemented with 92 firefighters and showed a significant decrease in the number of days lost due to back injuries in the year following (from 59 to 0 days). The number of days lost per worker dropped from 0.64 to 0.13 two years after program implementation.
28 Others have suggested that short-term improvements in physical fitness may not necessarily translate into reduced risk of low back injury.
29Our findings confirm a high prevalence of MSK complaints among firefighters and highlight the need for primary and secondary prevention. However, our study has some inherent limitations. First, a cross-sectional study can only measure associations and provides limited confidence as to whether predictors are causative. Second, the female firefighters subgroup was underpowered because of the small number of women in the workforce, leading to imprecision in our estimates. Female firefighters may have different LOS, exposure, or risk profiles from their male colleagues; therefore, our results cannot be considered valid for all female firefighters. Third, our method of recording MSD sites was not diagnostic and cannot be used to target specific ergonomic changes. Finally, our sample was drawn from a single fire service; there may be variations in job requirements, personal factors, and training across different contexts that our results do not take into account. While we did try to reduce selection bias by enrolling as many of the roster firefighters as possible, we do not know the reasons why the rest chose not to participate, nor can we ascertain whether non-participants, as a group, differed from participants. In addition, our prevalence rates may be underestimated, as disabled firefighters and those on sick leave were not sampled.
Future steps toward understanding the MSK problems that occur among firefighters should include better sampling of women and larger sample sizes across multiple fire associations to allow for calculation of sex- and age-specific rates. Longitudinal studies of firefighters over time will determine which factors contribute to changes in MSK health problems. Such cohorts should consider personal, physical, psychosocial, task demand, sex/gender, age, injury exposure, and environmental factors in multivariate models, since work injury and recovery is a complex biopsychosocial problem.