Bidirectional Relationship Between Atopic Dermatitis and Psychiatric Comorbidities in Individuals of European Ancestry: A Mendelian Randomization Study

Abstract

Atopic dermatitis is a chronic inflammatory skin disorder that significantly impacts quality of life and is often associated with psychiatric comorbidities. How-ever, the causal relationship between atopic dermatitis and psychiatric disorders remains unclear. This study employed bidirectional 2-sample Mendelian randomization to investigate the potential causal relationships between atopic dermatitis and 8 psychiatric conditions: depression, anxiety, autism spectrum disorder, attention deficit hyperactivity disorder, suicidality, bipolar disorder, obsessive-compulsive disorder, and schizophrenia. Genetic instruments were derived from large-scale genome-wide association studies of European ancestry. Forward Mendelian randomization analysis indicated that atopic dermatitis causally increases the risk of anxiety (inverse variance weighting p = 0.016; odds ratio = 1.110, 95% confidence interval: 1.019-1.208). Reverse Mendelian randomization analysis revealed a significant causal effect of bipolar disorder on increasing the risk of atopic dermatitis (inverse variance weighting p = 0.005; odds ratio = 1.062, 95% confidence interval: 1.018-1.107). No significant causal relationships were found for other psychiatric conditions. Sensitivity analyses confirmed the robustness of these findings, with no evidence of horizontal pleiotropy. These results highlight the need for integrated dermatological and psychiatric care, emphasizing early mental health screening for atopic dermatitis patients and dermatological evaluation for individuals with bipolar disorder. Future research should explore underlying biological mechanisms and validate findings across diverse populations.

Fig. 1

Bidirectional Mendelian randomization (MR) design. MR analyses require 3 core assumptions. In the forward MR analysis (red arrows), atopic dermatitis (AD) is treated as the exposure and psychiatric comorbidities (PC) as the outcome. In the reverse MR analysis (blue arrows), PC is treated as the exposure and AD as the outcome.

Fig. 2

Mendelian randomization (MR) flowchart. The flowchart outlines the analytical pipeline used in the Mendelian randomization (MR) study. The process begins with selecting exposure data, represented by total SNPs, and follows with quality control measures to exclude confounding factors using the LDtrait tool and SNPs with weak instrumental strength (F < 10). Subsequently, SNPs linked to outcome data are extracted, with harmonization procedures employed to exclude palindromic SNPs. The MR analysis is performed using robust statistical methods, including MR-Egger, weighted median, inverse-variance weighting (IVW), simple mode, and weighted mode. Sensitivity analyses are incorporated to evaluate heterogeneity (Cochran’s Q statistics), pleiotropy (MR-Egger intercept method, MR-PRESSO), and robustness (leave-one-out test). SNPs identified as outliers during these tests are excluded to ensure valid causal inference. The thresholds for forward and reverse analyses are specified, and all analyses are executed using R software.