Psychiatric Comorbidities in Adult Patients with Atopic Dermatitis: A Nationwide Cohort Study Compared to Melanocytic Naevi

Abstract

Background/Objectives: Atopic dermatitis (AD) is a chronic inflammatory skin disorder increasingly recognized for its association with psychiatric comorbidities. However, the extent of this association compared to dermatologic controls in Asian populations remains underexplored. We sought to evaluate the prevalence and risk of psychiatric comorbidities in adult patients with AD compared to those with melanocytic naevi using a nationwide population-based cohort. Methods: We conducted a retrospective cohort study utilizing the Korean National Health Insurance Service (NHIS) database, including individuals diagnosed with AD (ICD-10 code L20.0) or melanocytic naevi (ICD-10 code D22, excluding melanoma) between 1 January 2010 and 31 December 2023. Patients were required to have at least five years of diagnostic history and be 25 years or older at the end of the study. Psychiatric comorbidities were identified based on ICD-10 codes. Adjusted odds ratios (ORs) and 95% confidence intervals (CIs) were calculated to compare psychiatric morbidity between groups. Results: Among 1,902,114 individuals (1,813,320 with AD and 88,794 with naevi), psychiatric comorbidities were more prevalent in the AD group (28.2%) compared to the naevi group (27.1%) (adjusted OR 1.04, 95% CI 1.02–1.05). While differences for major depression, bipolar disorder, and personality disorders were not statistically significant, other psychiatric categories suggested significantly higher prevalence in the AD group. Sex-stratified analysis revealed a higher overall psychiatric morbidity in women compared to men; however, the relative risk increase associated with AD was slightly greater in men than in women. Comparison with previous international studies suggests that Korea’s healthcare accessibility and nationwide mental health programs may contribute to the smaller observed difference. Conclusions: This large-scale cohort study highlights a modest but significant association between AD and psychiatric comorbidities in adults. Our findings underscore the importance of integrating mental health assessment into routine dermatologic care for AD patients to improve comprehensive disease management.

1. Introduction

Atopic dermatitis (AD) is one of the major chronic inflammatory skin diseases with an increasing global prevalence, causing significant health burdens across all age groups [1,2]. AD is characterized by persistent pruritus and inflammatory skin lesions, which lead to sleep disturbances, social withdrawal, and decreased productivity, ultimately deteriorating the patient’s quality of life (QoL) [3,4]. Recent studies suggest that AD may not be merely a skin condition but may also be closely linked to various psychological comorbidities. Patients with AD are at significantly higher risk of developing depression (Major Depressive Disorder) and anxiety disorders compared to the general population [5,6]. Epidemiological studies in children have also reported associations with neurodevelopmental disorders such as attention-deficit hyperactivity disorder (ADHD) and autism spectrum disorder (ASD) [5,7,8,9,10]. Among adolescent AD patients, increased risks of suicidal ideation, suicide plans, and suicide attempts have consistently been observed [4,11,12], suggesting that the impact of AD on mental health is not transient but may be clinically significant.

Nevertheless, some studies have reported that the utilization rates of psychiatric care or psychiatric hospitalization among AD patients do not differ significantly from those of the general population [13]. These findings raise concerns that mental health issues in AD patients may be systematically underdiagnosed, or that structural barriers to mental healthcare accessibility may exist [14]. Therefore, there is a need for more definitive evidence to clarify the relationship between AD and mental health problems, and to support the necessity of public health interventions.

Previous studies have often been limited by cross-sectional designs or small clinical samples, which restricted causal interpretation and generalizability. Therefore, the following study utilized a large-scale healthcare database accumulated over the past decade to compare the AD patient group with a control group of melanocytic naevi (ICD-10 code D22) patients. The naevi group shares the skin condition but has a relatively low mental health risk, and patients with a history of malignant melanoma (C43 and D03) were excluded from the analysis.

In this study, the adjusted odds ratio (adj OR) was calculated, accounting for key confounders such as gender, age, and disability registration status to evaluate the independent association between AD and the occurrence of psychiatric disorders.

The results of this study are expected to provide scientific evidence for enhanced mental health screening and early intervention strategies for AD patients and contribute to the development of public health policies aimed at addressing mental health disparities.

2. Methods

2.1. Data Source and Study Populations

Data were obtained from the nationwide population-based National Health Insurance Service (NHIS) database of South Korea [15]. The NHIS operates a universal health insurance system, with the central government as the sole insurer for the entire population, including citizens and long-term residents. However, specific groups, such as short-term foreign residents, may have different coverage eligibility. As of 2020, the NHIS covered over 52 million individuals, with 97.1% enrolled in health insurance and 2.9% receiving medical aid. All insurance claims are recorded within this system, and healthcare providers exclusively supply medical data. The NHIS database includes socio-demographic details such as sex, age, insurance premiums, medical aid status, disability status, and residential region [15].

For this study, a retrospective cohort of individuals who received a diagnosis of either atopic dermatitis (AD) or melanocytic naevi between 1 January 2010 and 31 December 2023 was constructed based on NHIS claims data.

AD cases were identified using the International Classification of Diseases (ICD) code L20.0, and the control group comprised individuals diagnosed with melanocytic naevi (ICD code D22), excluding those with a history of melanoma (ICD codes C43 or D03). A key rationale for selecting individuals with melanocytic naevi as a control group was to mitigate potential confounding factors associated with healthcare-seeking behavior for dermatological conditions. As melanocytic naevi, like atopic dermatitis (AD), are dermatological conditions, they serve as a suitable control group to partially account for potential confounding variables related to dermatological characteristics, such as frequency of hospital visits and the likelihood of medical record documentation. Unlike AD, however, naevi are not inflammatory conditions and have not been directly associated with psychiatric comorbidities in prior literature. This makes them an optimal control group for this study. Individuals who had records of both AD and melanocytic naevi diagnoses during the study period were excluded to avoid overlap.

Given that this study focused on the psychiatric comorbidities associated with AD, we restricted our analysis to individuals aged 25 years or older as of 31 December 2023. This age threshold was selected to ensure that all included participants had at least five years of healthcare utilization data available following the transition into adulthood, thus enhancing the reliability of psychiatric comorbidity assessment.

We initially identified 2,713,320 individuals who had received a diagnosis of AD and 92,964 individuals diagnosed with melanocytic naevi during the study period. In the first exclusion step, we removed individuals with a documented history of psychiatric disorders before their initial AD or naevi diagnosis. In the second step, we restricted the cohort to those with at least five years of continuous healthcare utilization related to the respective diagnoses. After applying all inclusion and exclusion criteria, 1,902,114 individuals remained for final analysis: 1,813,320 in the AD group and 88,794 in the melanocytic naevi control group.

2.2. Statistical Analysis

Crude odds ratios (ORs) and 95% confidence intervals (CIs) were first calculated to estimate the association between atopic dermatitis (AD) and psychiatric comorbidities. Logistic regression models were then applied to obtain adjusted odds ratios (aORs), accounting for key potential confounders including age, sex, residential region, income level (insurance premium quartile), disability registration status, and medical aid status. Separate models were constructed for each psychiatric comorbidity category (major depression, anxiety disorders, schizophrenia, etc.). Statistical significance was defined as a two-tailed p-value < 0.05. All analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC, USA).

3. Results

3.1. Characteristics of the Study Population

The baseline characteristics of patients with atopic dermatitis (AD) and those with melanocytic naevi (control group) are summarized in

Table 1. Characteristics of patients in AD and melanocytic naevi groups.

	AD (n = 1,813,320)	Naevi (n = 88,794)
Age, years, mean ± SD	43.0 ± 12.3	44.9 ± 14.0
Male, n (%)	733,761 (40.5%)	30,290 (34.1%)
Female, n (%)	1,079,559 (59.5%)	58,504 (65.9%)

. The mean age was 43.0 years in the AD group and 44.9 years in the naevi group, indicating comparable age distributions. Regarding sex distribution, 40.5% of patients with AD were male, compared to 34.1% in the naevi group, suggesting a higher proportion of female patients in the control group.

3.2. Association Between AD and Psychiatric Disorders

The comparison of psychiatric disorder prevalence between the two groups is summarized in

Table 2. Prevalence of psychiatric comorbidities and associated odds ratios in AD and naevi groups.

Comorbidities	Group	n	%	Crude OR (95% CI)	Adjusted OR (95% CI)
All psychiatric disorders	AD	510,896	28.2%	1.053 (1.037–1.069)	1.04 (1.02–1.06)
	Naevi	24,098	27.1%	(Reference)	(Reference)
Major depression (F32, F33, F34.1)	AD	219,062	12.1%	1.019 (0.998–1.041)	1.01 (0.99–1.03)
	Naevi	10,548	11.9%	(Reference)	(Reference)
Bipolar disorder or manic episodes (F30-F31)	AD	15,556	0.9%	1.074 (0.995–1.158)	1.07 (0.99–1.15)
	Naevi	710	0.8%	(Reference)	(Reference)
All psychiatric disorders (severe cases) (F20-F29)	AD	24,160	1.3%	1.086 (1.021–1.154)	1.08 (1.01–1.15)
	Naevi	1091	1.2%	(Reference)	(Reference)
Schizophrenia and schizotypal disorder (F20-F21)	AD	17,290	1.0%	1.082 (1.007–1.163)	1.10 (1.02–1.18)
	Naevi	783	0.9%	(Reference)	(Reference)
Neurotic, stress-related, and somatoform disorders (F40-F45, F48)	AD	398,742	22.0%	1.054 (1.037–1.072)	1.04 (1.02–1.05)
	Naevi	18,719	21.0%	(Reference)	(Reference)
Anxiety disorders (F40-F42)	AD	225,324	12.4%	1.091 (1.068–1.114)	1.06 (1.03–1.08)
	Naevi	10,222	11.5%	(Reference)	(Reference)
Disorders of adult personality and behavior (F60-F69)	AD	3581	0.2%	1.124 (0.958–1.32)	1.11 (0.95–1.31)
	Naevi	156	0.2%	(Reference)	(Reference)

. Among patients with AD, 28.2% had at least one diagnosed psychiatric disorder, compared to 27.1% in the naevi group. This difference was statistically significant, with an adjusted odds ratio (OR) of 1.04 (95% confidence interval [CI]: 1.02–1.05). When stratified by specific psychiatric disorders, no statistically significant differences were observed between the two groups for major depressive disorder, bipolar disorder or manic episodes, and disorders of adult personality and behavior. However, for all other psychiatric disorder categories, patients with AD exhibited a significantly higher prevalence compared to the control group.

According to

Table 3. Prevalence (%) of psychiatric disorders by sex in AD and naevi groups.

Comorbidities	Sex	AD (%)	Naevi (%)
All psychiatric disorders	Male	22.0	20.4
	Female	32.4	30.6
Major depression (F32, F33, F34.1)	Male	8.5	8.1
	Female	14.5	13.9
Bipolar disorder or manic episodes (F30-F31)	Male	0.8	0.6
	Female	0.9	0.9
All psychiatric disorders (severe cases) (F20-F29)	Male	1.3	1.1
	Female	1.4	1.3
Schizophrenia and schizotypal disorder (F20-F21)	Male	0.9	0.8
	Female	1.0	0.9
Neurotic, stress-related, and somatoform disorders (F40-F45, F48)	Male	17.2	15.7
	Female	25.3	23.8
Anxiety disorders (F40-F42)	Male	9.6	8.6
	Female	14.4	13.0
Disorders of adult personality and behavior (F60-F69)	Male	0.3	0.2
	Female	0.2	0.1

, the overall prevalence of psychiatric disorders was higher in the AD group (22.0% for males, 32.4% for females) compared to the Naevi group (20.4% for males, 30.6% for females), with females exhibiting a substantially higher prevalence in both groups. While the AD group consistently showed marginally higher prevalences across most categories, the largest absolute differences were observed for major depression and anxiety disorders, particularly among females. Specifically, for neurotic, stress-related, and somatoform disorders, the prevalence was notably high in both groups but higher in the AD group (17.2% vs. 15.7% for males; 25.3% vs. 23.8% for females), suggesting a potential association between AD and this spectrum of conditions.

3.3. Sex-Specific Differences in Psychiatric Disorders

Among patients with AD, females exhibited a higher prevalence of psychiatric disorders compared to males across all diagnostic categories, except for disorders of adult personality and behavior. The proportion of patients with at least one psychiatric disorder was 32.4% in females and 22.0% in males. Additionally, the proportion of patients diagnosed with two or more psychiatric disorders was 26.0% among females and 11.7% among males, indicating a higher psychiatric burden in female patients.

Although the overall prevalence of psychiatric disorders was higher in females than in males within both the AD and naevi groups, the relative risk of psychiatric disorders associated with AD compared to naevi was slightly higher in males than in females (adjusted OR for males: 1.045 vs. females: 1.038; p = 0.01) (Figure 1). No other significant sex-based differences in the odds of individual psychiatric disorders were observed.

4. Discussion

This large, population-based cohort study utilizing data from the Korean National Health Insurance Service (NHIS) suggested that patients with atopic dermatitis (AD) exhibited a higher risk of psychiatric comorbidities compared to patients with melanocytic naevi.

The primary finding of our study is that 28.2% of AD patients had at least one psychiatric diagnosis, compared to 27.1% in the naevi group, with an adjusted odds ratio (OR) of 1.04 (95% confidence interval [CI] 1.02–1.05).

Compared to previous studies, the observed difference in psychiatric morbidity between groups was relatively modest. For instance, a large-scale Finnish study using a similar dermatologic control group reported a 4.2 percentage point difference (17.3% vs. 13.1%) with an OR of 1.25. The relatively smaller difference observed in our study could be partially attributed to Korea’s high healthcare accessibility, where universal insurance coverage allows individuals free access to mental health services without significant financial barriers. Moreover, nationwide mental health counseling programs targeted at young adults may contribute to earlier diagnosis and management of psychiatric conditions, potentially diminishing disease-specific disparities. Nevertheless, findings from other countries have shown a wide range of associations. A Taiwanese study reported that patients with AD had a sevenfold increased risk of major depressive disorder and a sixfold risk for depressive disorders overall [16]. Similarly, studies from the United States and South Korea reported a twofold increased risk of depression among AD patients compared to the general population [17,18]. Conversely, a Danish nationwide study found that while severe AD was associated with increased use of antidepressants and anxiolytics, no significant difference in clinical diagnosis of depression or anxiety was observed [19]. This discrepancy may stem from methodological differences, such as sample size and diagnostic definitions. Interestingly, another insurance-based study similar to ours [20] reported a psychiatric comorbidity rate of 27.4% among AD patients, with notable associations with stress-related disorders, behavioral disorders, mood disorders, and psychosis. The use of claims-based datasets, capturing diagnosed and treated cases rather than undiagnosed psychiatric conditions, was cited as a possible reason for higher observed prevalences, aligning with our findings.

5. Limitations

Despite the strengths of this study, including the largest sample size to date and the use of real-world healthcare utilization data over 14 years, several limitations must be acknowledged. First, diagnostic validity for AD could not be externally verified. To mitigate this, the cohort was restricted to individuals with at least five years of diagnostic history; however, it is recognized that future studies incorporating medication records or detailed clinical criteria may improve specificity. Second, a potential for differential reporting bias is noted between the AD and naevi groups due to the nature of the NHIS database. Specifically, melanocytic naevi may be less consistently coded or reported compared to atopic dermatitis, which could potentially underestimate the true prevalence of naevi. This differential reporting may have influenced the observed associations. Third, although statistical adjustment was performed for sex and age, data on other variables known to influence psychiatric morbidity, such as disease severity and more granular socioeconomic factors, were not available. Consequently, the possibility of residual confounding remains. Fourth, due to the nature of claims data, the precise temporal sequence between AD onset and psychiatric diagnoses could not be established, although inclusion was limited to individuals aged 25 or older with a sufficient preceding medical record history.

6. Conclusions

This study provides large-scale epidemiologic evidence of the association between adult atopic dermatitis and psychiatric comorbidities, based on a nationwide cohort spanning 14 years. Our findings suggest that dermatologic care for AD should incorporate mental health screening and counseling, highlighting the need for an integrated approach to the management of patients with chronic inflammatory skin diseases.