Hand eczema and lifestyle factors in the Dutch general population: Evidence for smoking, chronic stress, and obesity

Abstract Background Several risk factors, among other lifestyle factors, have been suggested for hand eczema (HE). Objectives To investigate a possible association between HE and lifestyle factors, including smoking, alcohol consumption, stress, body mass index (BMI), waist circumference, physical activity, diet, and amount of sleep in the Dutch general population. Methods Data from the large population‐based LifeLines Cohort Study was used. Individuals with HE in the past year were identified by a cross‐sectional questionnaire in 2020. At baseline, information on lifestyle factors was collected. Results In total 57 046 individuals were included in the present analysis. Smoking ≥8 cigarettes/day, and smoking ≥15 pack years showed a positive association with HE in the past year. In addition, chronic stress, a BMI >30 kg/m2, and a waist circumference of >90 cm were positively associated with HE in the past year. Conclusions The current study indicates that lifestyle factors are associated with HE. Advice regarding lifestyle factors might contribute to enhance overall health, of which HE might possibly benefit in conjunction. Further studies should also focus on the association between lifestyle factors and the severity and prognosis of HE rather than on occurrence alone.


| INTRODUCTION
Hand eczema (HE) is an inflammatory skin disease with a 1-year prevalence of 9.1% in the general population. 1,2 It may cause both farreaching personal consequences, with an impaired quality of life of those affected, and socio-economic consequences in terms of sick leave, job-loss and change, and high health care costs. [3][4][5] The pathogenesis of HE has not yet been fully elucidated, but both endogenous and exogenous factors are assumed to play a role. 6 Several risk factors, such as atopic dermatitis (AD), 7 contact allergy, 8 and wet work, 9,10 are known to cause or contribute to HE.
The association between dermatological diseases and lifestyle is increasingly a subject of research. Diverse lifestyle factors can influence the immune system and alter inflammatory processes. Therefore, it is hypothesized that when improving overall health, HE might benefit in conjunction. Lifestyle and behavioral changes might be of great importance in future complementary medicine, with a possible role for prevention and personalized treatment programs for HE. Tobacco smoking has been the most extensively studied in HE, but results are pointing in different directions. 11,12 In addition, evidence of the role of other lifestyle factors in HE is scanty.

| Smoking
Smoking behavior was categorized as never-, former-, and current smokers. Current smokers were categorized in smoking <8 and ≥8 cigarettes per day. One pack year was defined as smoking 20 cigarettes per day for 1 year; cigars were regarded as three cigarettes.

| Stress
The occurrence of stressful life events and chronic stress experienced in the last 12 months were measured using the List of Threatening Experiences (LTE) 16 and the Long-term Difficulties Inventory (LDI), 17 respectively. Both have been validated for large population-based cohorts. 18 The LTE compromises 12 life events, for which participants indicated whether or not the life events occurred, with a maximum total score of 12. Total scores were categorized in 0, 1, 2, and ≥3 points. The LDI consists of 12 life aspects for which participants indicate how they experienced these life aspects with respect to difficulty and stress on a three-point scale: 0 = not stressful, 1 = slightly stressful, and 2 = very stressful. Total scores range from 0 to 24, and were categorized as 0, 1-2, 3-4, and ≥5 points, with higher scores indicating more stress.

| Physical activity
Physical activity was measured using the Short Questionnaire to Assess Health-enhancing physical activity (SQUASH), which includes questions regarding commute activity, physical activity at work, household activities, and leisure time activities (including sports) of an average week in the past months. 19 Intensity was categorized in to light, moderate, or vigorous based on age-specific Metabolic Equivalent Tasks (METs) derived from Ainsworth's compendium of physical activity combined with the self-reported intensity of each activity. 20 Outcomes were presented as moderate and vigorous physical activity (MVPA), vigorous physical activity (VPA), and their tertiles in minutes per week (min/wk).

| Diet
Diet was categorized as a vegetarian and/or vegan diet, and the overall diet quality was assessed by using the LifeLines Diet Score (LLDS), a tool based on the 2015 Dutch Dietary Guidelines. 21 It consists of 12 food groups, including 9 food groups with proven positive health effects (vegetables, fruit, whole grain products, legumes and nuts, fish, oil and soft margarines, unsweetened diary, coffee, and tea) and 3 food groups with negative effects (red and processed meat, butter and hard margarines, and sugar-sweetened beverages). Per food group, the intake in grams per 1000 kcal is categorized into quintiles, awarded 0 to 4 points (negative groups scored inversely) and summed. The total LLDS ranges from 0 to 48, with higher scores representing a higher diet quality. For the present study the quintiles of the total LLDS were used.

| Statistical analyses
Analyses were performed using the Statistical Products and Service Solutions package version 25 (SPSS Inc., Chicago, IL, U.S.A.). All proportions were computed excluding missing answers. Differences between responders and nonresponders were assessed using an independent Student t-test, Mann-Whitney U test, or a chi-square test. Univariate and multivariate logistic regression models were performed with HE in the past year vs never HE as the dependent variable, and sex, age, self-reported physician diagnosed AD, exposure to wet activities, and all lifestyle factors as the independent variables. To verify a possible dose-response relationship between lifestyle factors and HE, all categorized continuous variables were also entered as continuous variables in separate models. Associations were expressed as odds ratios (ORs) with 95% confidence intervals (CIs). P-values of <.05 were considered to be statistically significant.

| Study population
The lifetime prevalence of HE was 15.0% (95% confidence interval [CI] 14.7-15.3) and the 1-year prevalence of HE was 7.3% (95% CI 7.1-7.5). There was an increased proportion of female responders compared to nonresponders (60.3% vs 57.2%, respectively), and older individuals were less likely to answer the questionnaire compared with younger individuals (mean age ± standard deviation [SD] 55.8 ± 12.2 vs 50.6 ± 12.3 years, respectively; age at the time of answering the add-on questionnaire in 2020). Overall, responders reported a more favorable lifestyle at the baseline assessment with less current smoking, less stress, and more moderate and vigorous activity compared to nonresponders (see online supplementary material S2).

| Lifestyle factors and hand eczema
Subjects with HE in the past year were more often female (70.5%) compared with subjects without HE ever (58.3%). The 1-year prevalence of HE decreased with age. More subjects with HE in the past year than subjects without HE ever reported AD (33.7% vs 5.7%, respectively) and exposure to wet activities (33.1% vs 23.2%, respectively) ( Table 1).
In the univariate analysis a positive association between HE in the past year and being female, having AD, and exposure to wet activities was found. Age showed a negative association with HE in the past year. Regarding lifestyle factors, there was a positive association between HE in the past year and smoking, stress, and a vegetarian/ vegan diet at baseline. A negative association was found between HE and former smokers, a history of ≥15 pack years, alcohol consumption, BMI (25-30 kg/m 2 ), waist circumference, the LLDS score, and physical activity at baseline.
Associations between HE and age, sex, AD, and exposure to wet activities were similar in all adjusted models compared with the univariate analysis. When adjusting for age and sex (model 1), HE in the past year showed a positive association with smoking, a history of ≥15 pack years, stress, overweight and obesity, a waist circumference of >90 cm and sleeping 5-7 hours per 24 hours at baseline.
When also adjusting for AD and exposure to wet activities in addition to age and sex (model 2), HE in the past year was more common in individuals reporting smoking ≥8 cigarettes/day, or a smoking history of ≥15 pack years at baseline. Furthermore, positive associations between HE and a BMI >30, a waist circumference of >90 cm, and individuals reporting more stress according to both the LTE and LDI at baseline were found. No association between HE and amount of sleep at baseline was found. In addition, no statistically significant associations for former smoking, alcohol consumption, physical activity, and diet at baseline were found in either adjusted models.  It has been debated whether tobacco smoking is associated with HE, as previous studies have been pointing in different directions. A systematic review and a meta-analysis reported the evidence of HE and smoking. 11,12 No association between smoking and the prevalence of HE was found in the meta-analysis. However, due to a lack of numerical data to perform the meta-analysis, this conclusion was based on only three studies, all conducted in the same country. 11 On the other hand, another systematic review reported cautiously that smoking may cause an increased prevalence and severity of HE, especially in high-risk occupations. 12 Several other chronic inflammatory skin diseases are linked to smoking, such as AD, 25 palmoplantar pustulosis, 26 and psoriasis. 27 Smoking has an immunomodulatory effect with elevated levels of immunoglobulin E (IgE), increased macrophage and dendritic cell activity, a release of pro-inflammatory cytokines, and a favored activity of the Th2 pathway. 28 In addition, nonimmunologic effects such as cutaneous vasoconstriction with delayed wound healing and chronic damages to the microcirculation might play a role in HE. 29 Moreover, especially in HE, the direct toxic effects of holding tobacco products, or the direct effect of tobacco products causing allergic contact dermatitis, might also influence the course of HE. 30 In the current study, a positive association between HE and smoking eight or more cigarettes per day was found, pointing in a direction of a positive dose-dependent relation between HE and smoking. It is possible that due to different categorization of smoking habits in previous studies, often categorized as smoking yes/no, the effect of higher daily smoking amounts is not revealed. Another possible explanation might be that smoking does not have much influence on the occurrence of HE itself, but it might act as a catalyst in individuals already prone to develop HE, leading to a higher severity and/or a worse prognosis of HE. This is supported by a few studies. A prospective multicenter cohort study in 1 608 patients with occupational HE found an increased severity and a worse prognosis of HE in smokers. 31 In another, questionnaire-based cross-sectional study a strong association between tobacco smoking and HE severity was reported. 32 In addition, in a recent register-based cohort study of 1 491 individuals with HE, current smoking was inversely associated with healing of HE. 33 The Another possible reason could be that one of the main causes of HE is atopic HE, which can be triggered by stress as well. 37 HE was associated with obesity and a higher waist circumference in the current study. Two previous studies have also reported the association between a BMI ≥30 kg/m 2 and the occurrence of HE. 22,24 This is the first study including results on waist circumference in individuals with HE. Waist circumference might act as an indicator of central obesity, which is associated with increased visceral fat. Increased visceral fat acts as an endocrine organ that activates macrophages and releases pro-inflammatory cytokines that might lead to immune dysregulation. 38 In the current study no association between HE and alcohol consumption was found, which is in line with previous studies. 22,[39][40][41] Only one other study investigated HE and a vegetarian/vegan diet in 6 095 adolescents, and also no association was found. 42 The association between diet quality and HE has not been studied before.

| CONCLUSION
In conclusion, this study showed a positive association between HE in the past year, smoking, chronic stress, obesity, and waist circumference. Because personalized medicine is a subject of increased interest, and future health care is moving forward to a more individual approach, attention to lifestyle interventions such as reducing stress, losing weight, and quitting smoking to promote better overall health may be important to include when counseling patients with HE. However, to get a better understanding of the effect of lifestyle factors on HE, it is important that further studies also focus on the association between lifestyle factors and the severity and prognosis of HE rather than on occurrence alone.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. ORCID