Courtney, Theodore K. MS; Matz, Simon MS; Webster, Barbara S. BSPT, PA-C
Well known for its high risk of occupational fatalities worldwide, 1–4 construction work is also known for its relatively high risk of nonfatal occupational morbidity. 2,5,6 In 1996, the construction industry, which comprised approximately 5.4% of the average annual US employment, accounted for 7.8% of all US nonfatal occupational injuries and illnesses and 9.7% of all cases involving a day or more away from work (DAFW). 7,8 Construction ranked second to manufacturing in total case incidence rate (9.9 per 100 worker-years) and a close second to transportation in DAFW case rate (3.7 per 100 worker-years). 7,8
Although a number of studies have examined the frequency distribution of nonfatal injuries within construction, 2,5 fewer have examined injury cost and disability duration. Recent articles by Glazner and colleagues 9,10 addressed cost and disability aspects of particular trades and phases of construction in detail using workers’ compensation (WC) data from the Denver International Airport construction project. However, their results reflect the experience of a single, large municipal project and may prove difficult to generalize on an industry-wide scale.
One workers’ compensation provider (WCP) in the US covers approximately 10% of the private insurance market with a wide distribution of coverage nationally. 11 Previous studies of WCP data have reported not only injury frequencies but also injury costs and durations of disability. 12–15 Two WCP claims data analyses have also been published that addressed characteristics of a specific injury type (eg, low back pain and slips, trips, and fall-related injuries) within the construction industry as part of larger studies of these injuries in US industry as a whole. 16,17
In the present study, 1996 WCP claims for injuries in the construction industry were examined by part of body (BP) and nature of injury (NOI) combinations. Cross-tabulated combinations were used to provide more specific injury descriptions. 18 The objectives of the study were to examine 1) the most frequently occurring construction injuries; 2) the injuries involving the highest percentage of cost; 3) the injuries that contributed the most to lost time disability in construction (count, median and mean number of disability days); and 4) the distribution of antecedents contributing to the most disabling of these injuries.
Construction industry claims data were selected from the total population of WC claims reported to WCP in 1996 using the approach published by Murphy and Courtney. 17 Basic manual classification codes established by the National Council on Compensation Insurance (NCCI) 19,20 were used to define a construction industry group subset for WCP claims.
In the United States, compensation for injuries and illnesses sustained on the job is organized and regulated at the state level. Generally, for nonfatal injuries, each state’s system provides reimbursement for medical treatment and partial reimbursement for wages lost as a result of the injury (for an overview of the US system of WC, see Mont et al 21). The claims selected into the study included construction WC claims from each state of the United States and the District of Columbia, except the six states maintaining monopolistic, state WC funds in 1996 (North Dakota, Nevada, Ohio, Washington, West Virginia, and Wyoming).
Affected body parts were classified based on the recorded narrative descriptions of injury to standard BP codes published by NCCI. Similarly, injuries were also classified using NCCI standard nature of injury codes. To maintain comparability between this study and prior reports of WCP data on low back pain (LBP) in the literature, the previously published definition for LBP involving an aggregation of BP and NOI codes was used. 17 The BP codes included in this definition were as follows: disc-trunk, lower back, multiple trunk, sacrum, and coccyx. This aggregation of BP codes is labeled “low back region” in the results. NOI codes included in this definition were as follows: contusion, fracture, hernia, inflammation, sprain, strain, and rupture. This aggregation of NOI codes is labeled LBP in the results.
Antecedent event categories were based on a “cause” code assigned by the WCP claims department. The cause codes were derived from narrative descriptions of the event from first reports of injury submitted by either the injured worker, employer, medical personnel, or claims personnel. 11,17
Cost and Disability Duration
Claims data were retrieved in August 2000, allowing a minimum 3.6-year claim cost development period. For closed cases, the total cost of a claim was equal to the sum of the paid indemnity, medical, and other expenses. For open cases, the total cost was equal to the sum of the indemnity, medical, and other expense amounts (including amounts already paid) estimated by insurance claim managers at the individual claim level according to established insurance industry practices.
For closed claims, disability duration was defined as the total number of disability days. Temporary total disability days and temporary partial disability days were summed to arrive at the number of disability days. To account for those days of disability that occur during state waiting periods (state-mandated number of disability days that must occur before the onset of indemnity benefits; up to 7 days), the days in each state-established waiting period were added to any indemnity claims that did not extend beyond the retroactive period (the point at which workers are retroactively compensated for the waiting period days) for that state.
For open claims and those involving lump-sum settlements, the method developed by Hashemi and colleagues 12–14 was used. Disability duration for these claims was estimated based on actual and estimated future indemnity payments. (A lump sum settlement is a one-time settlement that may be made when a claimant’s medical condition stabilizes but disability is expected to persist for a long period or becomes permanent.) Open and lump-sum claim disability durations were calculated by dividing the total indemnity amount by the average weekly payment rate. The average weekly rate was calculated as the lesser of two thirds of the claimant’s average weekly wage or the maximum rate allowed in the state of jurisdiction. For those claims without a reported average weekly wage, the median of all average weekly wages was used.
A 7-day week was used for calculating disability duration. Days were counted whether they involved total or partial disability and whether they involved temporary or permanent disability. This was done because it was not possible to assess how much time was lost if partial disability payments were made for part time work. All analyses were performed using SAS/STAT software version 8.0.
There were a total of 35,790 nonfatal construction WC claims reported to WCP in 1996. Of these, 784 (2.2%) remained open at the time of extraction. Table 1 presents those construction injuries (BP-NOI combinations) that accounted for more than 1% of the total number of WC claims. Collectively, the 15 injury types in Table 1 accounted for 51.2% of all construction claims. LBP occurred most frequently, accounting for 5301 claims (14.8%). LBP was followed by 3045 injuries that involved foreign bodies in the eye (8.5%) and 1720 that involved finger lacerations (4.8%). Lacerations of the fingers, thumb, hand, and lower arm accounted for approximately 11% of claims.
Table 2 presents the construction injuries that accounted for more than 1% of construction claim costs and involved at least 100 claims. LBP was again first leading all other injury combinations with 21.3% of construction claim costs. LBP was followed by all other specific injuries not otherwise classified (hereinafter NOC injuries) to multiple body parts (8.1% of cost), knee strain (3.4%), upper arm (including the shoulder) strain (3.1%), and wrist fracture (1.6%).
Table 3 ranks BP-NOI combinations by their total contribution to the number of disability days experienced by construction workers. The table contains those combinations accounting for more than 1% of the total number of disability days and more than 100 claims. LBP accounted for nearly one quarter of all disability days (25.5%) with 416,444 days, followed by NOC injuries to multiple body parts (5.1%), knee strain (4.4%), upper arm strain (3.1%), and wrist fracture (1.9%).
Table 4 presents BP-NOI combinations by the median and mean number of disability days resulting from the injury (case severity). The table is rank ordered by the median number of disability days and subsequently by the mean number of days. Table 4 also contains only those BP-NOI combinations accounting for more than 1% of the total number of disability days and more than 100 claims. The 12 combinations listed in Table 4 accounted for 49.2% of construction injury claims.
The overall median and mean disability duration experienced by injured construction workers was 0 days and 46 days, respectively. Fractures proved most disabling, with consistently higher median and mean disability durations than other injury types: ankle fracture, 55 days and 171 days, respectively; foot fracture, 42 days and 138 days; and wrist fracture, 38 days and 245 days. The medians for all other combinations were the same as the overall median, 0 days. Following fractures, those combinations with the highest mean durations of disability were NOC injuries of the lower back (118 days), multiple body part contusions (95 days), knee strains (91 days), upper arm contusions (80 days), and LBP (79 days).
Figure 1 describes the relative contribution of seven antecedent event categories to the top eight (most disabling) injuries from Table 4. (Hand tool-related injuries are those injuries primarily attributed to the use, including the malfunction or failure, of hand tools.) Slips, trips, and falls (STFs) collectively accounted for the largest proportion of injuries in six of the top eight injuries. Elevation-level and same-level STF were the leading two contributors to ankle fractures, wrist fractures, and knee strains. Manual materials handling (MMH) accounted for the largest share of injuries in the two low back-related categories. Struck by/against/caught between (SBAC) events were the most frequent or second most frequent contributors to upper arm contusions, foot fractures, and multiple body part contusions.
The nonfatal injury frequencies described here are similar to those previously published in the literature for the United States 5 and South Korea. 2 Kisner and Fosbroke, 5 using data from the US Bureau of Labor Statistics Supplementary Data System, reported that back sprains and strains, eye scratches, and finger lacerations/punctures accounted for 30% of nonfatal construction worker injuries during the period from 1981 to 1986. Table 1 indicates that a decade later, the three most frequent nonfatal construction injuries remain similarly distributed. Jeong 2 also reported high frequencies of back injuries and traumatic injuries of the foot, eye, and hand in Korean construction workers.
The frequency of LBP was similar, although slightly lower, than that reported for 1992 WCP construction claims (14.8% vs 15.7%) by Murphy and Courtney. 17 Hashemi et al 14 reported that LBP claims as a proportion of all WCP claims fell from 17.5% to 14.9% across all industry groups from 1992 to 1996.
As in previous studies of WC claims cost, 12–14,17,22 the distribution of cost was skewed. The average cost of a construction injury claim was 23.2 times higher than the median cost, indicating that a small number of claims accounted for a high percentage of cost.
Only five of the most frequent injury types from Table 1 remained in Table 2: LBP, NOC injuries to multiple body parts, upper arm strain, knee strain, and knee contusion. The proportion of construction claim costs related to LBP in 1996 (21.3%) was lower than that reported by Murphy and Courtney for LBP in construction in 1992 (28.3%). 17 Similar to that study, the contribution of LBP to 1996 construction claim costs (21.3%) was also disproportionate to its contribution to claim frequency (14.8%). These values are also similar to those reported by Hashemi et al 14 for LBP across all industries in 1996 (22.7% of cost and 14.9% of claims).
Table 3 and Table 2 provide a similar rank order suggesting that disability cost, rather than medical treatment cost, was the principal driver of cost for many of these claims. The overall median of 0 disability days (Table 4) indicated that at least half of construction workers injured in 1996 either did not lose time from work or did not lose enough time from work to qualify for wage replacement under the compensation system. The Bureau of Labor Statistics reported a median number of 7 DAFW for the construction industry in 1996. 8 The differences between the two medians reflect the influence of compensation system waiting periods because the BLS survey begins disability day accrual the day after the injury occurs.
Lowery et al 10 also reported injury disability duration in construction based on WC data from the Denver airport project. However, the article only reported disability days by the type or phase of construction activity and did not report disability duration for specific injuries. Due to methodological differences involving the restrictions on included cases used to calculate median disability duration in the Lowery et al. study, the results from the two studies cannot be compared.
The mean disability duration for LBP in the present study was higher (79 days) than that reported by Hashemi et al 14 for WCP LBP claims across all industry groups in 1996 (61 days).This variation may reflect the differential severity of LBP in the construction industry versus other industries, which is related to unique challenges in return to work—rapidly changing working conditions, difficulty in identifying alternative duty—confronted by the construction industry. 17 It is also possible that part of the difference reflects the dissimilar periods of cost development for 1996 claims (approximately 14 months minimum for Hashemi et al. versus 43 months for the present study).
Fractures had substantially higher disability duration means (138 to 245 days) and medians (38 to 55 days) than other injuries. Finger fractures, which apparently occur more frequently in construction, 23 did not contribute substantially to the total burden of disability days for the industry, whereas wrist and lower extremity fractures did. Intuitively, a finger fracture would not necessarily involve impairment of the entire hand like a wrist fracture nor result in prolonged disability like a fracture of the foot or ankle, which must bear a worker’s weight.
Fractures as a class of injury are particularly problematic because once they have occurred, the recuperation time is relatively fixed and somewhat less likely to be shortened by traditional return to work efforts than some other types of injuries. Here, too, the return to work challenges confronted by construction play a role. Construction work is physically demanding, requiring significantly more worker capacity for awkward postures and movements (eg, stooping, bending, climbing, balancing, kneeling, crouching, and crawling, strength, and tolerance of weather and environmental extremes—heat, cold, moisture, wind, etc.). Compared with the general industry, it is somewhat more difficult to materially reduce worker exposures to these sorts of conditions. 24 Given this lack of malleability in workplace conditions and the disability duration that results, primary prevention is particularly strategic for combating fracture-related disability.
Sixty-one percent of LBP was attributed to MMH with the two STF categories accounting for an additional 17% of LBP cases. These percentages are consistent with those reported for 1992 LBP claims in construction (63% and 18%) by Murphy and Courtney, 17 indicating no relative change in the antecedent contributors to construction’s most burdensome injury type.
Holmstrom and colleagues 25 identified specific factors associated with LBP in construction workers in the literature. Associated activities included heavy manual work, such as lifting, lowering, and carrying, and awkward postures. Holmstrom et al 26 also have previously reported significant relationships between manual materials handling indices and LBP symptoms reported by construction workers. Everett 27 concluded that the majority of construction activities involved moderate or high exposure to risk factors for overexertion injuries.
Although construction presents significantly higher physical work demands, 24,28 approaches to reducing MMH related stresses do exist. These largely involve the application of existing techniques, such as reducing loads handled by workers, presenting loads more consistently and at mechanically advantageous heights, and transferring extremely heavy loads to mechanical materials handling systems. Schneider and Susi 29 describe materials handling and ergonomics-related factors encountered in each phase of a typical construction project and suggest approaches for addressing them.
Although LBP comprised a substantial burden on the construction industry in the aggregate, the results also suggest that the most disabling injuries in terms of the disability arising for a typical case—lower and upper extremity fractures—could be addressed by increasing prevention efforts for slips and falls and struck by/against exposures. The construction industry has the highest incidence rate of fall injuries with DAFW 30 and the highest per capita cost for fall injuries 16 among major US industry sectors. In the current study, elevation-level and same-level STF made the greatest contributions to wrist fracture (33% and 32%, respectively), and knee strain (20% and 36%). Courtney et al 31 recently reported that STF were responsible for 46% of fractures and 48% of sprains, strains, and tears involving DAFW across private industry. Here, interventions might include improving walking/working surfaces and fall protection resources and procedures. 32,33
SBAC events were the primary or important secondary contributors to upper arm contusions (32%), foot fractures (29%), and multiple body part contusions (20%). This substantial contribution of SBAC events to disabling injuries suggests potential for improving personal protective equipment for the foot and investigating protective possibilities for the shoulder as well as better controlling worker exposure to heavy equipment and mechanically manipulated materials.
Strengths and Limitations
This study was the first to examine the frequency, cost, disability duration, and disability-related antecedents of the leading injuries in the US construction industry. The strengths of the study included the use of a large, nationally distributed population of construction worker injuries and the use of data cross-tabulated by BP and NOI permitting better resolution of frequency, cost and disability burden of particular injuries. The data source utilized has also been benchmarked against other national surveillance data systems. 11
BLS DAFW data are right-censored because of the effects of the calendar year sampling scheme, which prevents the accurate reporting of total count or mean number of DAFW due to a loss to follow-up. 11,18,34,35 In the current study, right-censoring was minimal with nearly 98% of cases closed after a minimum 3.6-year claim development period.
The study limitations include factors related to methods and systematic features of WC claims data. The disability duration calculations could not take into account disability from those workers who did not lose enough time from work to satisfy the various state waiting periods for wage replacement benefits. Therefore, the results likely underestimate the actual disability experienced by construction workers and industry as a whole. Systematic and nonsystematic misclassification of both injury outcome and antecedent factors are possible the result of errors in data reporting and limitations in coding scheme specificity. 17 Finally, administrative data systems in general have several notable limitations including limited sensitivity for exposure assessment, susceptibility to filtering effects (and consequently underreporting) at each stage of the reporting process, and differential sensitivity to various types of injuries. 11,36,37
The present study found that average disability duration for an injured construction worker based on WC payments was 46 days with a median of 0 days. Direct cost and disability varied by the type of injury sustained.
The most frequently occurring construction injuries in 1996 were related to LBP, foreign body eye injuries, and finger lacerations. LBP also accounted for the greatest percentage of construction claim costs (21.3%) and disability days (25.5%). The conditions with the longest disability durations were sudden-onset injuries including fractures of the ankle (median = 55 days), foot (42 days), and wrist (38 days). MMH activities were most often associated with LBP disability. Same level and elevated falls were the principal exposures for wrist and ankle fractures, whereas elevated falls and struck by incidents accounted for the majority of foot fractures. The results suggest that these most disabling injuries can be addressed by increasing primary prevention resources in slips and falls and other sudden-onset exposures as well as in reducing manual materials handling and other gradual-onset exposures.
The authors gratefully acknowledge the assistance of Mr John Cotnam and Ms Jennifer Haimson in the acquisition and preliminary reduction of the data for this study. The authors also wish to extend their appreciation for the assistance of Drs Rammohan V. Maikala and Gary S. Sorock in the review of early drafts of the manuscript.
1. Ore T, Stout NA. Traumatic occupational fatalities in the US and Australian construction industries. Am J Indust Med. 1996; 30: 202–206.
2. Jeong BY. Occupational deaths and injuries in the construction industry. App Ergon. 1998; 29: 355–360.
3. Xia Z, Courtney TK, Sorock GS, et al. Fatal occupational injuries in a new development area in the People’s Republic of China, J Occup Environ Med. 2000; 42: 917–922.
4. Xia Z, Sorock GS, Zhu J, et al. Fatal occupational injuries in the construction industry of a new development area in east China, 1991 to 1997. Am Ind Hyg Assoc J. 2000; 61: 733–737.
5. Kisner SM, Fosbroke DE. Injury hazards in the construction industry. J Occup Environ Med. 1994; 36: 137–143.
6. Ringen K. National conference on ergonomics, safety and health in construction summary report. Am J Indust Med. 1994; 25: 775–781.
7. Bureau of Labor Statistics. Workplace Injuries and Illnesses in 1996. USDL 97-453. Washington, DC: US Government Printing Office; 1997.
8. Bureau of Labor Statistics. Lost Worktime Injuries and Illnesses: Characteristics and Resulting Time Away from Work, 1996. USDL 98–157. Washington, DC: US Government Printing Office; 1998.
9. Glazner JE, Borgerding J, Lowery JT, Bondy J, Mueller KL, Kreiss K. Construction injury rates may exceed national estimates: evidence from the construction of Denver International Airport. Am J Indust Med. 1998; 34: 105–112.
10. Lowery JT, Glazner J, Borgerding, JA, Bondy J, Lezotte D, Kreiss K. Analysis of construction injury burden by type of work. Am J Indust Med. 2000; 37: 390–399.
11. Murphy PL, Sorock GS, Courtney TK, Webster BS, Leamon TB. Injury and illness in the American workplace: a comparison of data sources. Am J Indust Med. 1996; 30: 130–141.
12. Hashemi L, Webster BS, Clancy EA, Volinn E. Length of disability and cost of workers’ compensation low back pain claims. J Occup Environ Med. 1997; 39: 937–945.
13. Hashemi L, Webster BS, Clancy EA, Courtney TK. Length of disability and cost of work-related musculoskeletal disorders of the upper extremity. J Occup Environ Med. 1998; 40: 261–269.
14. Hashemi L, Webster BS, Clancy EA. Trends of disability duration and cost for workers’ compensation low back pain claims (1988–1996). J Occup Environ Med. 1998; 40: 1110–1119.
15. Mahmud MA, Webster BS, Courtney TK, Matz S, Tacci JA, Christiani DC. Clinical management and the duration of disability for work-related low back pain. J Occup Environ Med. 2000; 42: 1178–1187.
16. Leamon, TB, Murphy, PL. Occupational slips and falls: more than a trivial problem. Ergonomics 1995; 38: 487–98.
17. Murphy PL, Courtney TK. Low back pain disability: relative costs by antecedent and industry group. Am J Indust Med. 2000; 37: 558–71.
18. Courtney TK, Webster BS. Disabling occupational morbidity in the United States: an alternative way of seeing the Bureau of Labor Statistics’ data. J Occup Environ Med. 1999; 41: 60–69.
19. National Council on Compensation Insurance. Scopes of Basic Manual Classifications. Boca Raton, FL: NCCI; 1989.
20. National Council on Compensation Insurance. Scopes of Basic Manual Classifications. Boca Raton, FL: NCCI; 1996.
21. Mont D, Burton JF, Reno V, Thompson C. Workers’ Compensation: Benefits, Coverage and Costs, 1999 New Estimates and 1996–1998 Revisions.
Washington, DC: National Academy of Social Insurance; 2001:2–3. Available at: http://www.nasi.org/usr doc/NASI%20Workers%20Comp%20FINAL.pdf.
22. Cheadle A, Franklin G, Wolfhagen, C, et al. Factors influencing the duration of work-related disability: a population-based study of Washington state workers’ compensation. Am J Pub Health. 1994; 84: 190–196.
23. Islam S, Biswas R, Nambier A, et al. Incidence and risk of work-related fracture injuries: experience of a state-managed workers’ compensation system. J Occup Environ Med. 2001; 43: 140–146.
24. Schneider S, Griffin M, Chowdhury R. Ergonomic exposures of construction workers: an analysis of the US Department of Labor Employment and Training Administration database on job demands. Appl Occup Environ Hyg. 1998; 13: 238–241.
25. Holmstrom EB, Moritz U, Engholm G. Musculoskeletal disorders in construction workers. J Occup Environ Med. 1995; 10: 295–312.
26. Holmstrom EB, Lindell J, Moritz U. Low back and neck/shoulder pain in construction workers: occupational workload and psychosocial risk factors. Spine. 1992; 17: 663–671.
27. Everett JG. Overexertion injuries in construction. J Construct Eng Mgmt. 1999; 125: 109–114.
28. Lee GKL, Chan CCH, Hui-chan CWY. Work profile and functional capacity of formwork carpenters at construction sites. Dis Rehab. 2001; 23: 9–14.
29. Schneider S, Susi P. Ergonomics and construction: a review of potential hazards in new construction. Am Ind Hyg Assoc J. 1994; 55: 635–649.
30. Webster T. Workplace falls. Compensation Working Conditions. 2000; 502: 28–38.
31. Courtney TK, Sorock GS, Manning DP, Collins JW, Holbein-Jenny MA. Occupational slip, trip, and fall-related injuries—can the contribution of slipperiness be isolated? Ergonomics. 2001; 44: 1118–1137.
32. Hsiao H, Simeonov P. Preventing falls from roofs: a critical review. Ergonomics. 2001; 44: 537–561.
33. Becker P, Fullen M, Akladios M, Hobbs G. Prevention of construction falls by organizational intervention. Injury Prev. 2001; 7 (Supp I): i64–i67.
34. Oleinick A, Guire KE, Hawthorne VM, et al. Current methods of estimating severity for occupational injuries and illnesses: data from the 1986 Michigan comprehensive compensable injury and illness-database. Am J Ind Med. 1993; 23: 231–252.
35. Courtney TK, Webster BS. Antecedent factors and disabling occupational morbidity—insights from the new BLS data. Am Ind Hyg Assoc J. 2001; 62: 622–632.
36. Webb GR, Redman S, Wilkinson C, Sanson-Fisher RW. Filtering effects in reporting work injuries. Accid Anal Prev. 1989; 21: 115–123.
37. Hagberg M, Christiani D, Courtney TK, Halperin W, Leamon TB, Smith TJ. Conceptual and definitional issues in occupational injury epidemiology. Am J Ind Med. 1997; 32: 106–115.