Risk factors of hand eczema: A population‐based study among 900 subjects

Abstract

Background

Many risk factors such as atopic dermatitis (AD) have shown to associate with hand eczema (HE). However, studies concerning other atopic diseases, parental or longitudinal risk factors of HE are scarce.

Objectives

To examine the association between HE and atopic diseases, parental factors, environmental factors (keeping animals, exposure to moulds) and lifestyle factors (obesity, tobacco smoking, alcohol consumption and physical activity) at population level.

Methods

Subjects belonging to the Northern Finland Birth Cohort 1966 Study (NFBC1966) (n = 6830) answered a comprehensive health questionnaire. The data was completed with parental information.

Results

HE was reported in 900 (13.3%) individuals. All atopic diseases, parental allergy, female gender and obesity increased the risk of HE whereas physical activity decreased the risk of HE. A statistically significant association was not found between HE and tobacco smoking or alcohol consumption.

Conclusions

All atopic diseases, not only AD, seem to have influence on the presence of HE. In addition, parental and environmental factors associated with HE.

In birth cohort setting (N = 6830), this study examined the prevalence and multiple risk factors of hand eczema (HE). HE was reported in 900 (13.3%) subjects. All atopic diseases, parental allergy, female gender and obesity increased the risk of HE whereas physical activity decreased the risk of HE.

Abbreviations

AD

atopic dermatitis

BMI

body mass index

CI

confidential interval

HE

hand eczema

NFBC1966

the Northern Finland Birth Cohort 1966 Study

OR

odds ratio

SD

standard deviation

1. INTRODUCTION

Hand eczema (HE) is one of the most frequent dermatological disorders with a significant impact on quality of life and remarkable costs to society. ¹ , ² In addition, HE is known as the most common occupational skin disease. ³ , ⁴ The prevalence of HE varies slightly between studies. According to a recent systematic review and meta‐analysis, the pooled estimates for point, 1‐year and lifetime prevalence of HE in general population were 4.0%, 9.1% and 14.5%, respectively. ⁵ The incidence of HE is 7.3 cases/1000 person‐years ⁵ and HE is more common among women than men. ⁵ , ⁶

Pathomechanistically, HE is an inflammatory disease with a chronic and relapsing course and multifactorial aetiology. History of atopic dermatitis (AD) is known to be an endogenous risk factor of HE whereas wet work, exposure to irritants and contact sensitization are exogenous risk factors. ¹ , ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰

Researchers have shown increasing interest towards associations between lifestyle factors and HE. ⁶ , ⁷ , ¹¹ The association between tobacco smoking and HE is controversial. ¹² , ¹³ Smoking has been shown to be associated with HE. ⁷ , ¹¹ , ¹⁴ , ¹⁵ , ¹⁶ A Danish review evaluated that smoking is associated with increasing prevalence of HE, particularly in occupational settings. ¹² However, a systematic review and meta‐analysis performed in Sweden did not confirm tobacco smoking as a risk factor for HE. ¹⁷ An association between HE and stress ¹¹ , ¹⁶ , ¹⁸ and obesity ¹¹ , ¹⁶ , ¹⁸ has been found, whereas high‐physical activity has shown to be a protective factor of HE. ¹¹ , ¹⁹ In addition, HE has not been shown to be associated with alcohol consumption. ¹⁴ , ¹⁶

Previous studies have mostly evaluated risk factors of HE in cross‐sectional settings ⁴ , ⁷ , ¹¹ , ¹⁴ , ¹⁵ , ¹⁹ whereas studies addressing longitudinal or parental risk factors of HE are scarce. In addition, even though AD is a known risk factor of HE, the association between HE and other atopic diseases has not been studied widely. The aim of this population‐based study was to evaluate the association between atopic diseases (asthma, allergic rhinoconjunctivitis and AD) and HE besides several lifetime and environmental risk factors of HE in middle‐aged adults by using the Northern Finland Birth Cohort 1966 Study (NFBC1966). In addition, in this longitudinal cohort study we investigated the parental risk factors for the offspring's HE.

2. MATERIALS AND METHODS

The NFBC1966 data, collected since 1965, includes mothers whose expected delivery date fell during the year 1966 and their offspring. At that time, all the mothers were living in northern Finland in the provinces of Oulu and Lapland. The data included 12 055 mothers with 12 058 live‐born children (12 068 deliveries). The children of NFBC1966 have been followed regularly via health questionnaires and clinical examinations. Four main follow‐up surveys were conducted when the cohort members were 1, 14, 31 and 46 years old. The parents of the cohort subjects have also been followed via national registers and medical reports. ²⁰ , ²¹

2.1. Questionnaire

As part of the 46‐year follow‐up, study members completed a comprehensive questionnaire, which included 132 questions regarding physical health, lifestyle and environmental factors and socio‐economic status (SES). HE was recognized by the question ‘Do you have or have you ever had HE diagnosed or treated by a physician?’ Asthma, allergic rhinoconjunctivitis and AD were identified by two questions, which (a) consisted of the respondent's own evaluation of disease as well as (b) disease diagnosed and treated by a physician. For example, (a) ‘Do you have asthma past 12 months?’/‘have you had asthma previously?’ and (b) ‘Have you had asthma diagnosed or treated by a physician?’ The diagnosis was determined if one of the questions had been answered in the affirmative. Information about owning pets and farm animals, apartment mould exposure (smell or visible) and symptoms or sickness at home or in place of work because of exposure to mould was also gathered via the same questionnaire. SES was defined based on education level, estimated as a most usable measure of socioeconomic status, ²² and classified in three subgroups (basic, secondary and tertiary education).

Maternal data was collected by a questionnaire from 24th to 28th gestational week and if needed, the questionnaire was completed later during pregnancy or after delivery (10.1% of mothers). Parental data regarding history of AD or allergic rhinitis (later named parental allergy) and asthma was obtained by 31‐year and 46‐year follow‐up studies. Definitions of body mass index, alcohol consumption, smoking, and physical activity are described in Supplementary Methods S1.

2.2. Statistical analysis

The overall prevalence of HE was calculated. Distributions of continuous variables were expressed as mean and SD and categorical variables as numbers and percentage of proportions. A Chi square test and Fisher's exact test were used to test difference between HE and possible risk factors. Continuous variables (maternal age at birth, parity, BMI and menarche age) were tested with the Mann–Whitney U test. A logistic regression analysis was used to examine associations between the presence of HE and risk factors with crude and adjusted odds ratios (ORs) and 95% confidence intervals (CIs). The following potential confounding factors were adjusted for: AD, asthma, allergic rhinoconjuntivitis, SES, living on a farm, mould exposure (smell or visible) and symptoms or sickness at home or place of work because of exposure to mould, physical activity age at 46, sex, BMI, maternal BMI during pregnancy and parental allergy. Statistical analyses were performed using R version 4.1.0. A P value <.05 was considered statistically significant.

2.3. Ethics statement

The Ethical Committee of the Northern Ostrobothnia Hospital District approved the study, which was performed according to the principles of the 1983 Helsinki Declaration. Written informed consent for scientific purposes was received from all participants.

3. RESULTS

A questionnaire was sent to 10 321 study subjects who were living in Finland and whose address was known. Of these, 6830 (66.2%) subjects aged 45 to 46 responded to the questionnaire. Of the respondents, 3715 (54.4%) were women. Of the study subjects, 40 (1.3%) men and 41 (1.1%) women did not answer the question about HE.

HE was reported in 900 (13.3%) individuals. HE was more common among women (592, 16.1%) than in men (308, 10.0%) (OR 1.73, 95%CI: 1.49–2.0, P < .001). HE was slightly more common among those with primary and secondary education (13.8%) than those with tertiary education (12.6%), but this relationship was not statistically significant (p = 0.187). Descriptive data regarding potential risk factors of HE is presented in Table 1.

TABLE 1.

Characteristics of potential risk factors for the presence of HE (n = 900)

	HE n(%)	OR (95%CI)	P value	P overall
Sex				<.001
Men	308 (34.2)	1
Women	592 (65.8)	1.73 (1.49‐2.00)
Education				.187
Primary/Secondary	575 (63.9)	1
Tertiary	325 (36.1)	0.90 (0.78‐1.05)
Physical activity				.010
Very low	220 (24.4)	1
Low	394 (43.8)	0.93 (0.78‐1.12)	.449
Moderate	267 (29.7)	0.78 (0.64‐0.94)	.010
High	18 (2.0)	0.56 (0.33‐0.91)	.018
BMI (kg/m2)				.006
<18.5	2 (0.2)	1
18.5‐<25	315 (35.0)	2.57 (0.79‐17.0)	.134
25‐<30	353 (39.2)	2.88 (0.88‐19.1)	.087
>30	227 (25.2)	3.44 (1.05‐22.8)	.041
Alcohol consumption				.213
Never	106 (11.8)	1
Light	716 (79.6)	0.86 (0.69‐1.08)	.182
Moderate	38 (4.2)	0.72 (0.48‐1.05)	.092
Heavy	39 (4.3)	0.71 (0.48‐1.05)	.085
Tobacco smoking				.193
Never	432 (48.0)	1
Former	239 (26.6)	1.13 (0.96‐1.34)	.147
Current smoker	223 (24.8)	1.14 (0.96‐1.36)	.131
Mould exposure a	126 (14.0)	1.32 (1.07‐1.61)		.011
Symptoms/sickness of mould exposure
At home	60 (6.7)	1.94 (1.43‐2.59)		<.001
At workplace	244 (27.1)	1.84 (1.56‐2.16)		<.001
Parental asthma	200 (22.2)	1.16 (0.98‐1.37)		.098
Paternal	91 (10.1)	1.16 (0.91‐1.46)		.251
Maternal	128 (14.2)	1.17 (0.95‐1.43)		.147
Parental allergy	328 (36.4)	1.98 (1.70‐2.30)		<.001
Paternal	162 (18.0)	2.06 (1.69‐2.49)		<.001
Maternal	221 (24.6)	1.81 (1.52‐2.14)		<.001
Maternal b age (years), mean (SD)	28.4 (6.5)	1.01 (1.00‐1.02)		.048
Maternal b BMI, mean (SD)	23.3 (3.4)	1.02 (1.00‐1.05)		.047
Maternal b menarche age (years), mean (SD)	13.9 (1.6)	0.94 (0.90‐0.99)		.024
Parity, mean (SD)	2.97 (2.25)	1.02 (0.99‐1.06)		.142
Gestational age (weeks), mean (SD)	40.1 (1.85)	1.01 (0.98‐1.05)		.470
Birth weight (grams), mean (SD)	3491 (503)	0.995(0.982‐1.009) d		.482
Dog owners c	297 (33.0)	0.83 (0.71‐0.97)		.024
Cow owners c	29 (3.2)	1.82 (0.63‐6.77)		.403

Note: Significant P values are shown in bold. There are some missing data because not all cohort members answered all the questions. Parental history of AD or allergic rhinitis (named as parental allergy).

Abbreviations: BMI, body mass index; CI, confidence interval; OR, odds ratio; SD standard deviation.

^a In apartment.

^b Maternal data was collected during pregnancy or after delivery.

^c Data of keeping animal was collected with the 46‐year follow up questionnaire.

^d Estimates are per 100 g.

3.1. Atopic diseases as a risk factor of HE

Individuals with atopic diseases had an increased risk of having HE. There was a strong association between HE and asthma, allergic rhinoconjunctivitis and AD (P < .001 separately) (Table 2). After adjusting for female gender, SES, BMI, maternal BMI, parental allergy, physical activity, living on a farm, mould exposure and reported symptoms at home or at place of work because of mould in multivariable logistic regression analysis, all factors remained associated with HE. Additionally, after further adjusting for AD, asthma and allergic rhinoconjunctivitis the majority of other atopic diseases (AD, asthma and allergic rhinoconjunctivitis) still remained associated with HE (P < .001, P = .003 and P = .02, respectively). Sex did not modify the association between atopic diseases and HE.

TABLE 2.

Characteristic of atopic diseases to the presence of HE

	HE n (%)	OR (95%CI)	P value	aOR (95%CI)	P value
Asthma	253 (28.1)	2.39 (2.03–2.81)	<.001	1.38 (1.12–1.71)	.003
Allergic rhinoconjunctivitis a , b	621 (69.0)	2.96 (2.55–3.44)	<.001	1.28 (1.04–1.56)	.02
Atopic dermatitis	652 (72.4)	11.8 (10.1–13.9)	<.001	9.66 (8.03–11.66)	<.001

Note: Odds ratio (OR) and adjusted odds ratio (aOR) of HE, adjusted for AD, asthma, allergic rhinoconjunctivitis, SES, living on a farm, mould exposure (smell or visible) and symptoms or sickness at home or place of work because of exposure to mould, physical activity age at 46, sex, BMI, maternal BMI during pregnancy and parental history of AD or allergic rhinitis (named as parental allergy). Confidence interval (CI). Significant P values are shown in bold.

^a Pruritus or water leaking from eyes when near animals or during pollen season.

^b For example, regarding animals or pollen, for example, hay fever.

3.2. Lifestyle factors as a risk factor of HE

Obesity was associated with HE (OR 3.44, 95%CI: 1.05‐22.8, P = .041). Individuals who reported moderate or high‐physical activity had decreased risk of having HE (OR 0.78, 95%CI: 0.64‐0.94, P = .010 and OR 0.56 95%CI: 0.33‐0.91, P = .018 respectively) when compared with those who were less active. HE was reported slightly more often among former and current smokers than those who had never smoked, but this relationship was not statistically significant (P = .193). Individuals with moderate or heavy alcohol consumption reported HE less frequently compared with those with light alcohol consumption or never drinkers, but this association was not statistically significant (P = .213). (Table 1).

3.3. Parental factors as a risk factor of HE

We also examined parental risk factors of the offspring's HE and found that parental allergy increased the risk of HE (OR 1.98, 95%CI: 1.70‐2.30, P < .001). Correspondingly, maternal factors such as age, BMI and menarche age also increased the risk of the offspring's HE, that is, as the variables increase the risk of HE increases (P = .048, P = .047 and P = .024, respectively) (Table 1). There was no relationship between maternal tobacco smoking, parental asthma, birth weight, parity, gestational age and offspring's HE. (Table 1).

3.4. Other factors as a risk factor of HE

When comparing those with pets and farm animals to those without, we found that dog owners had decreased risk of HE compared to those without a dog (OR 0.83, 95%CI: 0.71‐0.97, P = .024). There was a slightly greater amount of HE among cow owners compared to those without, but that did not reach statistical significance (P = .403). There was no association between having other farm animals or cats and HE (data not shown). (Table 1).

Those who reported visible or smell of mould in their apartment had increased risk of HE when compared with those without a history of mould exposure (OR 1.32, 95%CI: 1.07–1.61, P = .011). Individuals who reported” Getting symptoms or being sick at home or at workplace because of exposure to mould” had HE more commonly than those who did not report this exposure (P < .001) (Table 1).

4. DISCUSSION

In this population‐based study, all atopic diseases, not only AD, were found as individual risk factors for HE. In addition, female gender, obesity and mould exposure increased the risk of HE. Parental allergy was also a risk factor of offspring's HE. Moderate or high‐physical activity as well as owning a dog appeared as protective factors of HE. No association was found between other lifestyle factors and HE.

In the present study, the prevalence of HE was 13.3%, which is comparable to a recent meta‐analysis (conducted in general population of 568 100 individuals) in which the lifetime prevalence of HE was 14.5%. ⁵ It is well known that HE is more common among women than men ⁵ , ⁶ , ⁹ and our results were in line with this outcome. The sex difference has been explained by differing exposure among women than men, not by sex differences in skin susceptibility. ⁵ , ⁶ Thus, it is speculated that sex differences in beauty care and use of cosmetics could explain the higher prevalence of HE among women. ³ Furthermore, the prevalence of AD (a known risk factor of HE) among females exceeds the prevalence among males. ³ , ⁶ It is known that female‐dominated occupations such as hairdressing, cleaning and health‐care work are more often exposed to wet work, ²³ and in addition to this wet work exposure continues at leisure time in women more often than men, ²⁴ which may also explain the higher prevalence of HE among females.

In line with previous studies, AD was a significant risk factor of HE in the present study. ³ , ⁵ , ¹⁰ AD was confirmed in 72.4% of the HE patients in our study, which was considerably more than in a recent meta‐analysis where the pooled proportion of adults with history of AD was 34.4% in HE subjects. ⁵ Interestingly, we also found an association between other atopic diseases (asthma and allergic rhinoconjunctivitis) and HE even though the impact of other atopic diseases were taken into account. However, these were lower than associations between AD and HE or between AD and obesity. Previously, some studies have demonstrated the association between other atopic diseases and HE. ³ , ⁴ , ²⁴ In a Norwegian study (n = 4206) asthma, hay fever and allergic rhinitis were also reported as predictive factors of HE. ⁴ In a Swedish study, an association between history of asthma/hay fever has been described, even when the effect of childhood eczema was eliminated. ³ This highlights the effect of other atopic diseases as an associative factor in HE. In a Finnish study of occupational HE (n = 605), both skin atopy and respiratory atopy were shown to associate with the continuation of HE. ²⁵ Whereas, two separate Swedish studies (n = 2927 adolescents) and (n = 1516 young adults) did not found association between asthma or rhinoconjunctivitis and HE. ²⁶ , ²⁷ However, most of the studies concerning risk factors of HE did not acknowledge other atopic diseases besides AD. ²⁵ Thus, more studies are needed to strengthen our findings of association between other atopic diseases and HE.

Obesity has appeared as a risk factor of HE in previous examinations ¹¹ , ¹⁶ but the results are partly inconsistent. ⁷ In our study, the association between HE and obesity was demonstrated. Obesity alters the normal physiology of the skin in many ways. ²⁸ For instance, obesity modifies the epidermal barrier of the skin causing increased transepidermal water loss and dry skin. Obesity also modifies cutaneous and systemic inflammation leading to chronic low‐inflammatory condition and possibly inducing hypersensitivity reactions. ²⁹ Interestingly, we found that moderate or high‐physical activity itself had a significant negative association with HE. Physical activity as a protective factor on HE has been documented in previous studies as well. ¹¹ , ¹⁹ It is known that regular physical activity has anti‐inflammatory effects that suppress systemic low‐grade inflammation. ³⁰ , ³¹ That could explain our results as HE is known to be an inflammatory disorder. ³²

Smoking and alcohol consumption did not appear as risk factors of HE in the present study. The association between smoking and HE has shown inconsistent results in previous studies. A Swedish cross‐sectional study found a slight increase in HE among smokers and a positive association between the two was also found in a recent Dutch study. ¹¹ , ¹⁶ In addition, according to a Danish review, smoking may increase the risk of HE particularly in high‐risk occupations, ¹² but a meta‐analysis did not find smoking to be a risk factor of HE. ¹⁷ More studies are needed to clarify the association between smoking and HE. Our findings regarding the association between alcohol consumption and HE are in line with previous studies. ¹¹ , ¹⁴ , ¹⁶

Due to the birth cohort study design, we were able to study parental factors and HE as well. We found that parental allergies had a strong association with HE whereas parental asthma did not appear as a risk factor of HE. Our findings are consistent with a Norwegian study in which a family background of allergies, hay fever, rhinitis (not asthma) and tobacco smoking associated with HE ⁴ . However, we did not find parental smoking to be a predictive factor of HE. Similarly, in a Chinese study of 934 nurses, personal or familial history of atopy increased the risk of HE in nurses. ³³ However, the authors do not report how the family history for atopy was defined in their study.

Surprisingly, there was more HE among individuals who reported visible or smell of mould in their apartment and among those who reported getting symptoms or being sick because of mould exposure at home or at workplace compared to those who did not report this exposure. According to our knowledge, the effect of mould exposure on HE has not been studied previously. Sick building syndrome (SBS) classically includes symptoms in the eyes, nose, respiratory organ and skin, for example. ³⁴ , ³⁵ In previous studies, self‐reported allergy like hay fever ³⁶ and atopy ³⁵ increased the risk for reporting SBS. A recent Finnish study found an association between toxicity of classroom dust and airborne microbes and teachers' risk of work‐related respiratory and ocular symptoms. ³⁷ More studies are needed to establish the association between mould exposure and HE.

The major strength of this study is the large and unique birth cohort data of general population. This population‐based, longitudinal study included also parental factors. The study design allowed us to evaluate the risk factors of HE comprehensively. A limitation of the study is self‐reported HE, which is, however, largely used in HE studies. ⁴ , ¹¹ , ¹⁶ , ³⁸ Furthermore, as a limitation we must admit that we could not classify different types of HE, such as AD, allergic contact eczema or irritant contact eczema, which could have an impact on the associations between risk factors and HE. Unfortunately, the appearance, severity and chronicity of HE was also uncharted, as were the effects of occupational exposures. Additionally, parental allergy (AD and allergic rhinitis) were asked together in the questionnaire and could not be analysed separately even though AD is a risk factor for developing IgE hypersensitivity that can be cause of both allergic rhinitis and allergic asthma. In addition, due to prospective research setting, we were unable to clarify the timing between owning an animal and the appearance of HE. More studies are needed to confirm these associations.

In conclusion, HE was a common finding in this middle‐aged population. Several atopic diseases, female gender and obesity as well as parental allergy appeared to be risk factors of HE whereas HE was less common among individuals who reported moderate or high‐physical activity. In clinical practice it is important to take into account all atopic diseases, not only AD, as well as lifestyle factors as risk factors of HE.

AUTHOR CONTRIBUTIONS

Suvi‐Päivikki Sinikumpu, Laura Huilaja, and Jari Jokelainen were involved in the study conception and design. Suvi‐Päivikki Sinikumpu and Laura Huilaja provided study materials and collated data. All authors were involved in analysis and interpretation of the data. Jari Jokelainen provided statistical expertise. Marjut Koskelo drafted the manuscript; Suvi‐Päivikki Sinikumpu and Laura Huilaja revised it critically for important intellectual content; all authors approved the final version. Suvi‐Päivikki Sinikumpu and Laura Huilaja are the guarantors. The corresponding author attests that all listed authors meet authorship criteria and that no others meeting the criteria have been omitted.

CONFLICT OF INTEREST

The authors have no conflicts of interest to declare.

Supporting information

Appendix S1 Supporting Information.

Koskelo M, Sinikumpu S‐P, Jokelainen J, Huilaja L. Risk factors of hand eczema: A population‐based study among 900 subjects. Contact Dermatitis. 2022;87(6):485‐491. doi: 10.1111/cod.14205

DATA AVAILABILITY STATEMENT

Data Available: NFBC data is available from the University of Oulu, Infrastructure for Population Studies. Permission to use the data can be applied for research purposes via electronic material request portal. In the use of data, we follow the EU general data protection regulation (679/2016) and Finnish Data Protection Act. The use of personal data is based on cohort participants written informed consent at his/her latest follow‐up study, which may cause limitations to its use. Please, contact NFBC project center (NFBCprojectcenter(at)oulu.fi) and visit the cohort website for more information.