Incidence Rates, Costs, Severity, and Work-Related Factors of Occupational Dermatitis: A Workers’ Compensation Analysis of Oregon, 1990-1997

Objectives  To extend and update past research on occupational dermatitis by examining recent workers’ compensation claims data.

Design  Retrospective analysis of workers’ compensation claims from Oregon (1990-1997).

Setting  All dermatitis-related workers’ compensation claims were merged with US census data to estimate rates of dermatitis by age, sex, occupation, and industry. Associated claim costs and disability times were also calculated from these data.

Participants  All individuals with accepted dermatitis claims (N  =  611) were included in the analyses.

Main Outcome Measures  The overall claim rates of individuals by age, sex, industry, and occupation were estimated. Total costs and average disability time were computed. Monthly patterns of dermatitis claims were examined.

Results  The average claim rate of occupational dermatitis was estimated to be 5.73 per 100 000 workers (95% confidence interval, 5.66-5.80). Statistically significant differences (P<.001) in claim rates by age, sex, industry, and occupation were found. More than 47% of all claimants had 1 year of job tenure or less. Employees in the farming, forestry, and fishing occupations and industries had significantly higher claim rates compared with employees in other occupations. The average cost per claim was $3552, and the average disability time was 23.9 days. Some temporal trends in claim rates were observed.

Conclusions  Occupational dermatitis remains a significant problem in workplace settings. In addition, certain types of occupations and industries seem to be particularly affected by occupational dermatitis. Interventions may be particularly valuable for workers with little job tenure.

Dermatoses resulting from occupational activities and work site conditions has been found to represent almost 15% of total workplace injuries, thus making it among the most prevalent of all occupational disorders and a serious public health concern in need of continued investigation.¹ The surveillance of occupational dermatitis has been a major task of health researchers aiming to assess risk factors and ultimately develop interventions to reduce job-related injuries to the skin. In doing so, investigators have used various epidemiological methods including the use of national data from the US Bureau of Labor Statistics and National Health Interview Survey, the United Kingdom’s EPIDERM and Occupational Physicians Reporting Activity (OPRA), clinical testing reports and case studies, and specific information collected from industry sources or workplaces.^2-18 As a complementary method of surveillance, workers’ compensation data were initially used to analyze injury costs and indemnity duration from occupational dermatitis in South Carolina and to estimate the risks of specific occupations among agricultural workers in California.^19,20 A formal analysis of assessing the value of workers’ compensation information as a method of surveillance of occupational skin maladies was later conducted using Ohio data, which were determined to contribute unique information, relative to other surveillance methods, with respect to detailing specific occupations for which risk was highest and involved factors that are most likely to cause skin diseases among workers.²¹ Subsequently, workers’ compensation data were used to estimate incidence and claim rates by claimant demographics, industry, body part affected, and source of injury in Washington State.²²

To estimate and compare occupational dermatitis rates, as well as examine information on claim incidence, costs, disability duration, sources of injury, and the demographics of claimants, the present study analyzed detailed workers’ compensation data from Oregon from 1990 through 1997, combined with baseline demographic and industrial data derived from the Department of Labor’s Current Population Survey (CPS). Our study updates previous findings using workers’ compensation data and explores in much more detail information on claimant injuries compared with studies in the past. Because Oregon requires a 3-day indemnification period before such data are reported, the cases examined in this study generally comprised more serious conditions than those in prior workers’ compensation analyses of occupational dermatitis claims. The results from this study were then compared with previous investigations to ascertain consistency between findings and identify areas especially in need of further research.

This study used all workers’ compensation claim data as provided by the Oregon Department of Consumer and Business Information and Management Division for 1990 through 1997. Oregon requires the reporting of claims when the claimant injury or illness is believed to have resulted in 3 or more days of disability or when permanent disability resulted even if no time was lost. In some cases, there were claims that were expected by insurers to have involved 3 or more days off but recorded only medical treatment, hence resulting in claims with no associated disability time.

The data set included information on claimant occupation and industry, individual demographics (eg, age and sex), work schedules, nature of reported injury, body part affected, compensated days of lost work, and costs associated with indemnity and injury. Only claims with the nature of injury coded as dermatitis (US Bureau of Labor Statistics Occupational Injury and Illness Classification System’s nature codes 1820-1829) were analyzed. Claim costs were tracked through 1999, and the cost data reflect accumulated claim costs through this time. Thus, claims were followed for at least 2 years. By the end of the observation period, 583 (95.4%) of all accepted dermatitis claims were closed, and for these claims the cost data were complete. For the remaining 28 open claims, all but 1 involved no costs and are likely to have remained open because the employer/insurer failed to file a notice of closure. For the remaining open claims, costs accumulated through 1999 were used in the analysis.

Data from the US Bureau of the Census’ CPS were used to estimate Oregon employment levels for different demographic categories and periods. The CPS is a monthly survey of approximately 50 000 households that is used by the US government to assess, among other things, monthly unemployment rates. The CPS is a rotating survey with households first surveyed for 4 months, then not surveyed for the next 8 months, and then surveyed for an additional 4 months before permanently leaving the survey. Most of the employment estimates are based on the monthly outgoing rotation group (CPS-MORG) files for 1990 through 1997. These files contain data for all individuals participating in their fourth or eighth monthly survey. For individuals in CPS-MORG, additional questions pertaining to an individual’s employment are asked. This study restricted the sample to individuals who reported residing in Oregon and who were employed at the time of the interview and excluded self-employed workers because they are not recorded in Oregon’s workers’ compensation reports.

The dermatitis claim rates for specific categories of workers were calculated by dividing the reported number of dermatitis claims by the number of employed individuals for each particular category. Data for the numerator were obtained from the Oregon workers’ compensation administrative data, while the denominator was estimated using CPS employment data for Oregon. The claim rates were then converted to injuries per 100 000 workers. Because estimates were used in calculating claim rates, 95% confidence interval (CI) estimates were derived using the delta method. The delta method uses a first-order Taylor series expansion of the rate function around the estimated claim rate and the standard error of the employment estimate to compute the standard error of the claim rate.²³ To test for claim rate differences across different groups, χ² test statistics were computed. All calculations were performed using Stata statistical software (version 8.2; Stata Corp, College Station, Tex).

Between 1990 and 1997, there were 727 workers’ compensation claims filed citing occupationally induced dermatitis as cause of injury, of which 611 claims (84.0%) were determined to be compensable. Of the accepted claims, 362 (59.3%) were filed by men and 249 (40.8%) were filed by women. Workers who were 24 years or younger accounted for 155 (25.4%) of the accepted claims, while those aged between 25 and 39 years accounted for 240 (39.3%), those aged between 40 and 55 years accounted for 165 (27.0%), and those older than 55 years accounted for 51 (8.4%) of the accepted claims. Using the CPS to estimate the number of workers covered and working during the study period, an overall claim rate of 5.73 (95% CI, 5.66-5.80) per 100 000 was computed. Further analysis using the CPS employment estimates by demographic subgroup of the Oregon workforce during the 1990 through 1997 period was performed, and the dermatitis claim rates per 100 000 workers by age and sex were estimated. χ² Tests for equality demonstrated significant differences between claim rates based on age and sex at the P<.001 level. These results are presented along with 95% CIs in Table 1, and a breakdown of rate estimates is provided in Figure 1.

The most frequently affected body part was the hands, with 230 (37.6%) of the total accepted claims. Multiple body parts and body systems accounted for 220 claims (36.0%), followed by dermatitis to the upper extremities (75 claims [12.3%]), arms (23 claims [3.8%]), and forearms and face and head (11 claims [1.8%] each). Less common body parts cited for occupational dermatitis claims include those to foot or feet (8 claims [1.3%]); legs (6 claims [1.0%]); eyes (5 claims [0.8%]); wrists (4 claims [0.7%]); lower leg (3 claims [0.5%]); multiple lower extremities, ankles, and chest (2 claims [0.3%] each); and thighs, knees, and pelvis/groin (1 claim [0.2%] each). In 6 claims (1.0%), the body part affected was not recorded.

Claims were analyzed by both the nature and the source of injury, and it was found that unspecified dermatitis was the largest nature of injury cited (486 claims [79.5%]), followed by contact dermatitis and other eczema (65 claims [10.6%]), dermatitis nowhere else classified (27 claims [4.4%]), allergic dermatitis (26 claims [4.3%]), irritant dermatitis (6 claims [1.0%]), and other contact dermatitis (1 claim [0.2%]). The sources of injury listed as causes of dermatitis were, in order of frequency, chemicals (314 claims [51.4%]), plants and vegetation (134 claims [21.9%]), other sources (95 claims [15.5%]), and clothing (38 claims [6.2%]). In 30 claims (4.9%), the source of dermatitis was listed as unclassified.

The mean total cost per dermatitis claim, including costs of total temporary disability (TTD), medical treatment, partial permanent disability (PPD), and vocational rehabilitation was $3552.35. For the 8-year period examined, the total cost of all dermatitis claims amounted to $2 170 489.55, averaging $271 311.94 annually. The TTD indemnification duration per claim was 23.9 days, with the most days off taken for claims citing allergic dermatitis with 27.1 days and from chemicals with 28.1 days. Table 2 presents a complete breakdown of all indemnification time and costs associated with the natures and sources of injury.

Estimates of dermatitis claim rates were calculated for both the major occupation and industry categories according to US census data. The highest average claim rate per 100 000 workers was found for farming, fishing, and forestry workers (18.24 claims [95% CI, 16.43-20.04]) and likewise for the forestry and fishing industry (64.53 claims [95% CI, 18.79-110.27]). Cost data were also examined by major occupation and major industry, and the highest total cost was for claims filed by employees whose major occupation was in precision production and crafts, with an average of $8181.27 per claim, and for those in the wholesale trade industry, with an average of $6928.44 per claim. The average number of days for TTD indemnity per claim was longest for precision production and craft workers (48.8 days) and for those in the transportation industry (37.7 days). Notably, while farming, fishing, and forestry workers had the highest claim rate, they also had the least cost associated with TTD indemnification, averaging $424.86 per claim. χ² Tests examining differences between claim rates by both occupations and industries were conducted, and significant differences at the P<.001 level were found. The average estimated claim rates, CIs, costs, and TTD indemnity durations for both major occupation and industry are given in Table 3.

An analysis of employee job tenure data shows that at the time of injury, the largest group of claimants were those with a tenure of 1 year or less (288 [47.1%] of the accepted claims). Employees with 1 to 5 years of tenure filed 163 (26.7%) of the dermatitis claims, followed by employees with 5 to 10 years of tenure (44 claims [7.2%]), employees with 10 to 15 years of tenure (26 claims [4.3%]), employees with 15 to 20 years of tenure (17 claims [2.8%]), and employees with more than 20 years of tenure (12 claims [2.0%]). Tenure information was missing for 61 (10.0%) of the claimants. The most expensive claims were made by workers with 10 to 15 years of tenure (averaging $5275.77 per claim), whereas the least expensive claims were made by workers with 15 to 20 years of tenure (averaging $2219.71 per claim). Employees with more than 20 years of job tenure took the longest average indemnification time per claim (30.3 days), whereas those with 5 to 10 years of tenure had the lowest average indemnification periods of all groups (6.3 days).

Dermatitis claims were also examined by subdividing incidence by season, and it was found that the majority of claims (192 [31.4%]) occurred during the summer months of June, July, and August, followed by the spring months of March, April, and May (158 claims [25.9%]), the fall months of September, October, and November (140 claims [22.9%]), and the winter months of December, January, and February (121 claims [19.8%]) claims. July had the highest amount of claims of all months (69 claims [11.3%]), which was followed with a steady decline in claims through the month of February. Figure 2 provides a breakdown of dermatitis claims by month.

In comparison with previous surveillance studies on occupational dermatitis, the overall rate of 5.73 per 100 000 employees estimated in the present study are among the lowest. For example, workers’ compensation data analyzed for Ohio workers reported a rate equivalent to 24 occupational skin diseases per 10 000 workers, incident rates reported from workers’ compensation data from Washington State ranged between 530 to 3070 for 100 000 full-time equivalent employee years.^21,22 However, there is much variation in the reported rates of occupational dermatitis, and our rates were similar to the rate of 11.2 claims per 100 000 workers when Bureau of Labor Statistics data were used and restricted to cases that involved lost work time only and similar to the rate of 12.9 claims per 100 000 workers when data from the EPIDERM and OPRA surveillance report (comprising dermatologist and occupational physician reports from the United Kingdom) were used.^3,7 It is important to note that these differences may largely be attributable to variations in the data collection and reporting systems used. For example, while some state workers’ compensation systems require the reporting of all medical incidents, others only record incidents that necessitate disability leave.²⁴ Because Oregon only mandates the reporting of incidents in which 3 or more days of disability leave are necessary, the lower ratesin this study relative to the others may be explained by this higher reporting requirement. For this reason, the present study makes a particularly important contribution to the literature on occupational dermatitis. By using data that primarily consisted of the most serious incidents of dermal injuries, many cases that required only minor medical attention and typically included in other studies were filtered out. It should also be noted that in our analysis the restriction to cases of dermatitis as opposed to all forms of occupational skin disorders may have led to lower reported rates compared with previous studies.

The results of the claims analysis indicate a declining trend in occupational dermatitis claims. However, caution should be exercised in interpreting this finding because the cause of this decline may be due to many factors that alone or interactively explain the decrease in claims. One possible explanation is that with the continued dissemination of research on occupational dermatitis, employers have increasingly adopted safety measures, such as greater use of protective clothing and skin barriers to reduce risk of dermal injury. Another possibility is that employees who observe the claims of coworkers whose claims are later rejected in the adjudication process may become discouraged from filing their own occupational dermatitis claims; the rejected claim rate of 16% reported in the “Results” section provides some support for this potential explanation. The reluctance of workers to file claims when particular job markets becomes restricted and variability in the federal funding of occupational health programs may also have affected the observed trend in claims. Thus, while the declining claim rate appears to give rise to optimism, these potentially confounding factors necessitates a tempering of enthusiasm. To confirm such a trend is actually the result of increased workplace awareness and prevention interventions, more workers’ compensation studies from varying states over similar periods, as well as studies that are able to control for many of the aforementioned confounders, are necessary.

An important outcome of this investigation was the relative comparison of rates between industries and occupations, in which those with the highest incidence of dermatitis were identified. It was found that workers in the forestry and fisheries, agriculture, and personal services industries had the highest rates of occupational dermatitis, and workers specifically in farming, fishing, and forestry, machine operators and assemblers, handlers, helpers, and laborers, and service occupations were specifically identified as having the highest claim rates. These findings are consistent with those from earlier investigations showing higher risk among these groups and highlight the continued need for diligence in developing workplace interventions to target these high-risk employee populations. It should be noted that a limitation of the Oregon workers’ compensation data is that injuries among self-employed individuals are not recorded, and some occupations that are comprised primarily of self-employed individuals who are exposed to workplace conditions that have been demonstrated to increase their risk of dermatitis, such as hairdressers,⁷ were unable to be assessed. Future research using state workers’ compensation data that collects information on self-employed individuals would be valuable to extending this line of research.

The analysis of employee tenure found that more than 47% of all claimants had 1 year of job tenure or less, and almost three quarters of claimants had 5 years or less of job tenure at the time of their claim. To some extent, this finding may be influenced by a “healthy worker” effect, in which new employees who are affected with occupational injuries opt out of the job or occupation, thus resulting in a higher concentration of younger workers reporting injury. Evidence of this effect has been previously reported among workers experiencing occupational skin disease.⁹ Nevertheless, this finding strongly indicates that employee training should be given high priority as an intervention for the prevention and reduction of occupational dermatitis in the workplace. Moreover, future investigations of employee job tenure and incidence of occupational dermatitis should be designed in a manner to control for potential healthy worker effects so that tenure effects can be more accurately measured without the potential of confounding with occupational switching.

More than half of all dermatitis claims were attributable to chemicals, which also resulted in the highest medical cost of all sources of injury recorded, and the most frequently reported bodily area of injury was the hands. The overall average cost per claim of $3552 was higher than previously reported,^19,22,25 as was the average TTD indemnification time of 23.9 days per claim. While the aforementioned Oregon reporting requirement of a 3-day indemnification period may have led to the relatively higher average costs and time-off per claim, these results still illustrate that the severity of injury associated with many claims are high and that there is a significant need for preventative efforts on the parts of employers. In the past, recommendations have been made advocating the increased use of a combination of protective clothing, barrier cream, warnings, personal and environmental hygiene, and education to increase the safety of workers from occupational skin disease.²⁶ The results of this study may provide the basis for employers to consider such recommendations not only from an altruistic viewpoint of employee welfare but from a cost-benefit one as well.

This study examined workers’ compensation claims from Oregon over an 8-year period and found that while the trend of workers’ compensation claims from occupational dermatitis is decreasing, the reasons were not readily discernible. Thus, occupational dermatitis continues to remain an important problem in need of further attention by employers and researchers alike. This is especially true in workplaces where chemicals are used and where employees’ skin is exposed to potential irritants. The data indicate that increased job training and implementation of preventative measures, especially in high-risk industries and occupations, should be diligently practiced for reducing occupationally caused dermatitis. The high costs and average time associated with the dermatitis claims underscore the need for continued employer attention to reducing the sources of occupational dermatitis in their workplaces. Continued research using workers’ compensation claims from different states with both similar and unique data could provide an important means by which to compare and contrast the findings in this study and could be used for prioritizing interventions to workplaces where they would be most efficacious.

Correspondence: Irwin B. Horwitz, PhD, University of Texas, School of Public Health at Houston, 1200 Herman Pressler, W310, Houston, TX 77030 (ihorwitz@sph.uth.tmc.edu).

Accepted for Publication: December 22, 2004.

Financial Disclosure: None.