Impact of atopic dermatitis on occupational contact dermatitis among young people: A retrospective cohort study

Occupational contact dermatitis (OCD) is a common occupational disease. Atopic dermatitis (AD) is a known risk factor for OCD.


| INTRODUCTION
Occupational contact dermatitis (OCD) is the most commonly recognized occupational disease in many Western countries including Denmark. 1 In Denmark, OCD is defined as contact dermatitis either caused or worsened by exposure to irritants and/or allergens in the workplace.Over 90% of all OCD cases affect the hands leading to hand eczema (HE). 2,34][15] AD has also been determined to be a strong predictor of persistent HE, both occupational and nonoccupational, 7,16 increasing the severity of HE, 7,17 and decreasing the quality of life in afflicted individuals. 17D often presents during training or within the first few months of employment in high-risk occupations, 2 and the mean age at notification of OCD to the Danish Labour Market Insurance (DLMI) has earlier been found to be $36 years.18 Furthermore, young workers have been found to be especially vulnerable toward the long-term consequences of OCD.19 To the best of our knowledge, no previous studies have examined the effect of AD on the disease course and long-term consequences of OCD among young workers.Understanding this is essential for effective preventive strategies.The aim of this study was to investigate the long-term effects of AD on the eczema status and quality of life of young individuals, after reporting OCD to the DLMI Register, as well as its influence on their work life.

| Study design
The study is a retrospective cohort study utilizing a selfadministered questionnaire.The study was conducted and reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines to ensure a robust methodological approach and transparent reporting of findings. 20

| Data from the Danish Labour Market Insurance Register
The Danish Labour Market Insurance Register (DLMI) evaluates and compensates recognized cases of occupational diseases in Denmark, including OCD.Before recognition, a dermatology specialist evaluates the case and conducts patch testing.To focus on young workers, data on individuals with recognized OCD between 2010 and 2019 under 35 years of age at notification were obtained from the DLMI, including sex, birthdate, dates of notification and recognition and diagnosis.The cut-off age of 35 years was chosen, as the mean age at notification has previously been found to be 36 years. 18cupational irritant contact dermatitis (OICD) was viewed as an exclusion diagnosis, and cases with both OICD and occupational allergic contact dermatitis (OACD) were coded as OACD.

| Questionnaire
A self-administered questionnaire, which included validated questions from other questionnaires and new questions specifically developed for the study, was sent electronically on 21 January 2021, with two reminders sent 1 month apart.Before the study, the questionnaire was evaluated and revised based on interviews with peers and hand eczema outpatients.The questionnaire covered various topics, including current status and severity of eczema, current occupational status and occupational consequences of OCD.A self-administered photographic guide 21 and a visual analogue scale were used to evaluate the severity of current hand eczema.Atopic dermatitis history was estimated by a question about doctor-diagnosed atopic dermatitis.3][24] We used the categorization of skindex-29 scores suggested by Nijsten et al. 25 Before engaging in the questionnaire, all participants were required to complete an informed consent process.Participants were provided with written material explaining the nature, purpose and procedures of the study, and explicitly consenting to participate in the research.

| Data analysis and statistics
Study data were collected and managed using REDCap electronic data capture tools hosted at the Capital Region of Denmark, 26,27  For comparisons of basic characteristics between workers with and without AD, Fisher's exact test and Wilcoxon rank-sum test were applied.To determine the association between explanatory variables and the investigated outcomes, we used binary logistic regression models, adjusting for age, sex and duration of time between notification and the time of the questionnaire.Before application, assumptions underlying the binary logistic regression models, including the linearity of the logit for continuous variables and lack of multicollinearity, were tested and found to be satisfied.
Linear regression models were used to examine the effect of AD adjusted for sex, age at notification and years since notification on severity evaluated on the VAS-scale and quality of life evaluated by DLQI and Skindex-29.Although the residuals were not normally distributed for all models, this was deemed acceptable given the large sample size.All tests were two-tailed, with the level of statistical significance set at 5%.

| Addressing bias
To minimize potential bias in this study, several measures were undertaken.Selection bias was addressed by identifying all eligible individuals with recognized OCD from the DLMI register and sending them a questionnaire, representing a comprehensive sample of the target population.Non-response bias could be a potential limitation, but we mitigated this by assessing the representativeness of the sample by comparing the baseline characteristics of respondents and nonrespondents (described in a previous study). 19call bias was mitigated by using a combination of registerbased data and self-reported data from the questionnaire.The register-based data provided objective information on OCD cases and dates, while the self-reported data allowed for insights into eczema severity, quality of life and occupational consequences.The use of validated questions in the questionnaire also helped reduce measurement bias.Furthermore, potential confounding factors, such as sex, age and duration between notification and the time of the questionnaire, were controlled for in the statistical analyses.

| Data permissions
The study was approved by the Knowledge Centre on Data Protection Compliance in the Capital Region of Denmark on behalf of the Danish Data Protection Board (P-2020-508).

| Basic characteristics
A total of 6251 workers met the inclusion criteria of being under the age of 35 years at notification and were sent a questionnaire.Out of these, 2942 workers responded to the questionnaire, resulting in a response rate of 47%.The study population's characteristics have been previously described and showed no significant differences between respondents and non-respondents. 19Of the respondents, 2392 answered the question about previous doctor-diagnosed AD and constituted the study population for the present study.
Table 1 presents the basic characteristics of the study population.The prevalence of AD was found to be statistically significantly higher in women than in men, 82.2% of workers with AD were women, compared with 69.1% of workers without AD (P < 0.001).Furthermore, workers with AD tended to report OCD at a younger age compared with those without AD (median age 25 years for the workers with AD and 27 years for the workers without AD, P < 0.001).The prevalence of AD was similar among workers with OICD and OACD.And the time between notification and questionnaire was comparable between workers with and without AD, with a median time of 6 years (interquartile range 4-9) for both groups.

| Socioeconomic characteristics
Table 2 presents the socioeconomic characteristics of the study population.Compared with workers without AD, workers with AD had a statistically significantly higher adjusted odds ratio (OR) for having a higher education (OR 1.4, 95% CI: 1.1-1.7,P < 0.01), while no significant differences were observed in their current occupational status when adjusting for age, sex and years since notification.Those with AD had a significantly higher adjusted OR of receiving their primary income from the public sector (OR 1.4, 95% CI: 1.2-1.7,P < 0.001), and a lower adjusted OR of being unemployed or on sickness benefits (OR 0.7, 95% CI: 0.5-0.9,P < 0.05 and 0.5, 95% CI: 0.3-1, P < 0.05, respectively).
Analysis of the occupational reasons for developing OCD revealed that workers with AD had a higher adjusted OR for having been healthcare workers when developing their OCD (OR 1.3, 95% CI: 1.1-1.6,P < 0.01), and a lower adjusted OR for having been cleaning personnel or agricultural workers/farmers (OR 0.5, 95% CI: 0.3-0.8,P < 0.01, and 0.5, 95% CI: 0.2-0.9,P < 0.05, respectively; Table 3).

| Severity of eczema
A significant difference was found between workers with and without AD in terms of self-reported eczema symptoms within the past 3 months (Table 4).Specifically, a higher proportion of workers with T A B L E 1 Basic characteristics of workers with atopic dermatitis and workers without atopic dermatitis (n = 2392).AD reported experiencing eczema during this period compared with workers without AD (82.5% vs. 72.6%),resulting in an adjusted OR of 1.7 (95% CI: 1.4-2.1,P < 0.001).Furthermore, workers with AD were more likely to report eczema symptoms for more than half of the time during the past 3 months than workers without AD, with a statistically significant adjusted OR of 1.3 (95% CI: 1.0-1.5,P < 0.05).
Workers with AD exhibited more severe all time worst eczema compared with workers without AD, as measured by the VAS scores (Table 5).The mean VAS score for the all-time worst eczema was 7.3 (standard deviation [SD] 2.1) for workers with AD and 7.0 (SD 2.3) for workers without AD, which was statistically significant (P < 0.05).No statistically significant difference in VAS score was found for current eczema or eczema within the last 3 months.

| Location of eczema
Among patients with AD, the most common location of eczema within the last 3 months was the hands, reported in 95% of cases.This was not significantly different from patients without AD (94%).Patients with AD had a significantly higher adjusted OR of face eczema within the last 3 months (adjusted OR 3.4, 95% CI: 2.6-4.4,P < 0.001), armpits (adjusted OR 3.0, 95% CI: 1.7-5.3,P < 0.001) and stomach, chest or back (adjusted OR 3.4, 95% CI: 2.4-4.9,P < 0.001) compared with patients without AD.
In addition, we also investigated the original location of the recognized OCD in workers with and without AD.As shown in Table 7, the most commonly affected location for OCD was the hands, followed by the face and elsewhere on the body, in both groups.Workers with AD had a higher OR for OCD on the face (adjusted OR 1.5, 95% CI: 1.1-1.9,P < 0.01) and elsewhere on the body (adjusted OR 1.6, 95% CI: 1.2-2.1,P < 0.01) compared with those without AD.There was no significant difference in the adjusted OR of OCD on the hands, feet, armpits or stomach, chest or back between the two groups.
T A B L E 2 Socioeconomic characteristics of workers with atopic dermatitis and workers without atopic dermatitis.

| Quality of life
The health-related quality of life of workers with OCD with and without AD was evaluated using three different questionnaires: the DLQI, Skindex-29 and the EQ-5D-5L.was comparable between the two groups.The impact interpretation for the Skindex-29 subscales ranged from moderate to extremely severe for workers with AD, while it varied from mild to severe for those without AD.
The results of the EQ-5D-5L questionnaire, which evaluates five dimensions of health: mobility, self-care, usual activities, pain/ discomfort and anxiety/depression can be found in Table S1.No statistically significant difference was found between the two groups for any of the five dimensions examined.

| DISCUSSION
In our study population of workers under the age of 35 years at notification with previously recognized OCD, we found the prevalence of previously diagnosed AD to be $42%.This is a substantially higher than the life-time prevalence of doctor-diagnosed AD among adults reported in a systematic review by Bylund et al.
which ranged from 17.6% to 20.2% in studies performed on data from the 21st century. 28This higher prevalence is not surprising, given that AD is a known risk factor for the development of OCD. 13 Our findings also reveal that the prevalence of AD was significantly higher in women than in men, which is consistent with previous studies that have reported higher rates of AD among females. 29,30rthermore, workers with AD tended to report developing OCD at a younger age compared with workers without AD.This could be attributed to the increased risk of eczema in individuals with AD, which may lead to the development of OCD at a faster rate compared with their colleagues without AD.
T A B L E 8 Occupational consequences.

| Socioeconomic effects of atopic dermatitis
Our study found that workers with AD tended to be better educated and were more likely to work in the public sector, particularly in healthcare than workers without AD.While we can only speculate on the reasons for these findings, one possible explanation for this pattern is that early experiences with eczema during childhood may have influenced individuals with AD to seek careers with less skin exposure (i.e., to avoid unskilled worker status), as was also found by Bandier et al. in a study from 2013. 31 Interestingly, our findings contrast with those of previous studies that have suggested a negative association between AD and educational attainment. 32,33This discrepancy may be partially attributed to our study's focus on workers, who may have a higher educational level overall than the general population.The observed higher proportion of healthcare workers among our study population with AD and OCD may be explained by the fact that individuals with a higher level of health literacy are more likely to seek out healthcare services and obtain a diagnosis of AD.
Our study also investigated the self-reported occupational consequences of OCD and AD and found somewhat conflicting results.
Specifically, workers with AD were more likely to report negative effects of OCD on their choice of jobs and occupations compared with workers without AD.However, workers with AD were less likely to report salary decrease or difficulties finding work compared with workers without AD.
One possible interpretation of these findings is that workers with AD are aware that their options for certain types of work are limited by their condition, and they may have adapted their job search accordingly.Additionally, it is possible that workers with AD have found ways to manage their symptoms and work effectively within their limitations, which may explain their lower likelihood of reporting difficulty finding work.

| Severity of eczema
In our study, which includes a study population with a median age at notification of 26 years, we found that AD can have a marked negative effect on the prognosis of OCD.Specifically, we observed a higher prevalence of eczema among workers with AD compared with those without AD, with workers reporting more frequent and severe eczema symptoms.Our results also support previous findings that suggest AD is strongly associated with increased severity of hand eczema among healthcare workers. 34

| Location of eczema
We found the hands to be the most commonly affected location for eczema in both groups, which is consistent with previous literature that the hands is the most affected area of OCD. 2,3However, workers with AD had a significantly higher risk of eczema on various other locations, including the face, armpits, stomach, chest or back, and other (not disclosed) locations.These findings are in line with the typical presentation of AD, which often involves the flexures, wrists, ankles and eyelids or the upper trunk, shoulders and scalp. 35garding current hand eczema, we found that workers with previously diagnosed AD had a significantly higher risk of eczema on the fingers, interdigital spaces, wrists and lower arms within the last 3 months compared with workers without AD.No significant difference was observed in eczema on the fingertips, dorsum manus and palm.
Moreover, our findings indicated that the most commonly affected first location of OCD was the hands, followed by the face and elsewhere on the body, in both groups.However, workers with AD had a slightly higher OR for OCD in the face and elsewhere on the body compared with workers without AD.This may suggest that workers with AD are at a slightly higher risk of developing OCD in different locations than workers without AD, or that their AD is exacerbated by their OCD.

| Quality of life
It is well known that both OCD and AD considerably impair the healthrelated quality of life of those affected. 10,36Our study found that the health-related quality of life was significantly more adversely affected in workers with AD than those without.We found an adjusted mean difference in DLQI between the two groups of 1 (95% CI: 0.6-1.4),which was statistically significant, albeit it did not reach the recommended minimal clinical important difference of 4, 37 suggesting that the observed difference in DLQi score might not be considered clinically meaningful.Nevertheless, we also found statistically significant higher mean scores in the Skindex-29 subscales 'emotions' and 'functioning', indicating that there is a valid difference in quality of life between the two groups.

| Strength, weaknesses and generalizability
Our study benefits from a large sample size, and the diverse occupational backgrounds of the included workers, which enhances the generalizability of our results to the broader working population.
Moreover, we employed a comprehensive approach to assess the association between AD and OCD, examining multiple aspects such as socioeconomic factors, severity, location of eczema, occupational consequences and quality of life.
Limitations include recall bias, which we attempted to mitigate using objective measures whenever possible and corroborated selfreported information with registry data from the DLMI when available, and attrition bias, as the response rate was 47%.We did however not find any alarming differences between responders and non-responders.Confounding is another concern; while we used statistical methods to control for known confounders, unmeasured or unknown confounders could still affect our findings.
The generalizability of our findings to the broader population of workers with OCD is likely high, as mandatory reporting ensures that a large proportion of workers with OCD are included in our study population.

| CONCLUSION
Our study demonstrates a significant association between AD and OCD, with AD being a notable risk factor for OCD development, and by negatively affecting the prognosis of OCD.By addressing both conditions, we can improve the prognosis, work life and health-related quality of life for affected individuals.These findings underscore the need for targeted prevention and treatment strategies to better manage the burden of AD and OCD in the workplace.
Odds ratio (OR) for the different educational and occupational groups in workers with atopic dermatitis compared with workers without atopic dermatitis adjusted for age at notification, years since notification and sex.Result in bold were statistically significant (P < 0.05).Abbreviation: CI, confidence interval.a Adjusted for age at notification, years since notification and sex.b Exact numbers cannot be disclosed due to discretion rules.*P < 0.05; **P < 0.01; ***P < 0.001.

T A B L E 5
Severity of eczema on a VAS-scale 0-10 for the workers all time worst eczema, their eczema within the last 3 months and their current eczema.10) of eczema within the last 3 months (n = 2340) Mean difference between the two groups adjusted for age at notification, sex and years since notification.Result in bold were statistically significant (P < 0.05).Abbreviation: CI, confidence interval.a Adjusted for age at notification, sex and years since notification.*P < 0.05; **P < 0.01; ***P < 0.001.

a
Adjusted for age at notification, sex and years since notification.b Exact numbers cannot be disclosed due to discretion rules.*P < 0.05; **P < 0.01; ***P < 0.001.T A B L E 7 Location of occupational contact dermatitis (OCD).

Table 8 )
Self-reported occupational reason for the development of occupational contact dermatitis.
. However, workers with AD were more likely to report negative occupational consequences of OCD in other areas.Specifically, workers with AD were more likely to report that their choice of jobs and occupations had been negatively affected byT A B L E 3Note: Odds ratio (OR) for the different occupational groups in workers with atopic dermatitis compared with workers without atopic dermatitis adjusted for age at notification and sex (n = 2228).Result in bold were statistically significant (P < 0.05).Abbreviation: CI, confidence interval.a Adjusted for age at notification and sex.*P < 0.05; **P < 0.01; ***P < 0.001.T A B L E 4 Current eczema status.a Adjusted for age at notification, sex and years since notification.*P < 0.05; **P < 0.01; ***P < 0.001.

Table 9
In the Skindex-29 assessment, workers with AD exhibited a significantly higher mean score than those without AD in the 'emotions' (adjusted mean difference 3.2 [95% CI: 1.3-5.2],P < 0.01) and 'symptoms' subscale (adjusted mean difference 6.6 [95% CI: 4.4-8.8],P < 0.001).However, the mean score in the 'functioning' subscale Note: OR for current location of eczema in workers with and without atopic dermatitis adjusted for age at notification, sex and years since notification.
T A B L E 6 Current location of eczema.a Adjusted for age at notification, sex and years since notification.*P < 0.05; **P < 0.01; ***P < 0.001.
Mean difference between workers with OCD with and without AD adjusted for age at notification, sex and years since notification.Result in bold were statistically significant (P < 0.05).