Risk of depression among patients with acne in the U.K.: a population-based cohort study

Dear Editor, Acne has been associated with adverse psychiatric symptoms.1 In dermatology outpatient clinics, approximately 25.2% of patients with acne experience some psychiatric morbidity.2 However, few studies have evaluated the clinically significant diagnostic category of major depressive disorder (MDD) among people with acne. Here, we investigated whether patients with acne are at an increased risk of developing MDD compared to the general population, using one of the largest electronic medical records databases in the world. A retrospective cohort study was conducted using data from The Health Improvement Network (THIN) (1986–2012), a large primary care database in the U.K. that also includes data from specialists. All individuals between 7 and 50 years of age with ≥ 1 Read codes (diagnostic codes linked to International Classification of Diseases codes) for acne were identified. A general population cohort without acne was also identified. The study was approved by the Conjoint Health Research Ethics Board at the University of Calgary (ID 24423) and from the Scientific Review Committee in the U.K. (ID 16THIN036), authorizing access to extract relevant data from THIN. A complete list of codes used to identify exposures, outcomes and covariates is available from the authors. All patients were followed from their start date for ≥ 2 years in THIN until the earliest of either their first MDD Read code (main outcome),3,4 transfer out of practice, death or end of data collection. Observations were censored at the end of follow‐up in patients where MDD was not observed during the study period. To identify only incident cases, patients with an acne or MDD Read code prior to the start of follow‐up were excluded. Baseline covariates at the start of follow‐up included age (young ≤ 19; adult > 19 years),5 sex, obesity (BMI ≥ 30 kg m−2), smoking status (never, former, current), alcohol use (yes or no), medical comorbidities using the Charlson Comorbidity Index (0 or ≥ 1 comorbidity) and socioeconomic status using the Townsend Deprivation Index (quintiles 1–5, with 1 least and 5 most deprivation). Cox proportional hazards models stratified by time were constructed with estimates reported as hazard ratios (HR) with 95% confidence intervals (95% CIs). A likelihood‐ratio test was used to assess statistical interactions. Confounding was assessed by whether a substantial change (≥ 10%) to the estimated HR resulted with removal of each covariate. Sensitivity analyses were conducted whereby patients prescribed isotretinoin were removed from analyses. A total of 134 437 patients with incident acne and 1 731 608 patients without acne were identified in THIN. Patients with acne were more frequently younger (67.6% vs. 22.8%), female (58.6% vs. 48.6%), of higher socioeconomic status (24.4% vs. 22.1%), never smokers (58.4% vs. 48.6%) and had a comorbidity (17.2% vs. 13.8%), but less frequently were alcohol users (17.0% vs. 39.0%) or obese (2.3% vs. 6.6%) (all P < 0.001). Over the 15‐year follow‐up, the probability of developing MDD was 18.5% among patients with acne and 12.0% in the general population without acne. Patients with acne had a significantly increased risk of developing MDD after adjustment for all covariates but only in the first 5 years after being diagnosed with acne. This risk was highest within 1 year of acne diagnosis (adjusted HR 1.63, 95% CI 1.33–2.00) and decreased thereafter (Fig. 1). No statistical interactions were observed at any time point. In the first 5 years following an acne diagnosis, sex was the only covariate producing a confounding effect. At 10 and 15 years, both sex and obesity demonstrated confounding effects. The increased risk of MDD among patients with acne remained unchanged after excluding patients who were prescribed isotretinoin (n = 900), who may have the most severe acne. Hazard ratios (HR) for the development of depression among patients with acne over time. The HRs shown indicate the risk of depression in patients with acne relative to the general population at various time points post‐acne diagnosis. These HRs were determined from fully adjusted models (aHR) including age, sex, socioeconomic deprivation, Charlson Comorbidity Index, smoking, alcohol use and obesity. These results were unchanged after excluding patients who received isotretinoin (< 1‐year HR 1.63, 95% CI 1.33–2.00; 5‐year HR 1.02, 95% CI 0.88–1.17; 10‐year HR 1.06, 95% CI 0.86–1.31; 15‐year HR 1.15, 95% CI 0.76–1.74). CI, Confidence interval. *Statistical significance, P < 0.05. Our results indicate that the onset of acne, when patients present for treatment because of active disease, is associated with a greater risk of developing depression. Although the severity of acne was not assessed directly in the current study, this finding suggests a potential dose–response relationship such that more active disease may lead to a greater risk of depression. In this study, female patients were more likely to have acne (or to present to a physician) and were also more likely to develop MDD, which is consistent with existing literature.6,7 At 10 and 15 years, both sex and obesity demonstrated confounding effects, so future research on the disease course of acne should account for these variables. A potential limitation of this study is that cases of acne or MDD may have been misclassified if patients with either condition did not present to a physician for treatment. Additionally, treatment for acne other than isotretinoin was not considered; however, as treatment has been shown to improve depressive symptoms,8 we believe that the estimates reported here are conservative. Lastly, significant results should be interpreted based on clinical relevance given the large sample size. In conclusion, given the tremendous burden of MDD and its temporal association with active acne, it is critical that physicians monitor mood symptoms in patients with acne and initiate prompt MDD management or seek consultation from a psychiatrist when needed. Funding sources: I.A.V. received funding for this project from an Alberta Innovates Health Solutions MD‐PhD Studentship and from the Mach‐Gaensslen Foundation of Canada. Conflicts of interest: none declared.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.