The Alopecia Areata in a Global Noninterventional Observational Cohort (ADAAGIO) study was a retrospective medical record review conducted in France, Germany, Spain, and the UK [18]. Medical record review was led by dermatologists experienced in managing patients with AA. The study was subjected to local ethics committee reviews for exemption (in France and the UK) or approval (in Spain and Germany) per local data privacy requirements on the basis that all data collected were fully anonymized prior to analysis.

Patients eligible for inclusion in ADAAGIO were required to meet the following criteria: diagnosis of AA with ≥ 50% scalp hair loss, aged ≥ 12 years at index, and ≥ 6 months of available post-index follow-up. Patients also needed to be receiving ongoing treatment for AA at index, or, if treatment naive, initiating new treatment for AA within 60 days post index. Furthermore, all patients were required to have ≥ 1 post-index clinic visit during which percent scalp hair loss was recorded. Patients were excluded if they had other types of alopecia, other diseases that can cause hair loss, or other scalp diseases that could interfere with assessments of hair loss or regrowth.

The index date was defined as the first observed date of de novo or progression to ≥ 50% scalp hair loss occurring between January 1, 2015, and December 31, 2019. This cross-sectional analysis of the ADAAGIO study included patients who had a DLQI score recorded at their index date.

The DLQI is a questionnaire consisting of 10 questions regarding symptoms and feelings, daily activities, leisure, work and school, personal relationships, and treatment as dimensions of life [15]. Each item is scored on a scale of 0–3 points, and scores are added for a total DLQI score of 0–30 points. Higher scores indicate greater impact of disease on quality of life. DLQI scores of 0–1 indicate no effect at all on quality of life, scores of 2–5 indicate a small effect, scores of 6–10 indicate a moderate effect, scores of 11–20 indicate a very large effect, and scores of 21–30 indicate an extremely large effect [19].

Analyses were stratified by age group at index: adults (aged ≥ 18 years) and adolescents (aged 12–17 years). Patient demographics and clinical characteristics were described using descriptive statistics. Standardized mean differences were calculated to assess the balance of baseline characteristics between the two age groups, with a standardized mean difference of < 0.1 indicating good balance.

To assess the effect of age group on DLQI score as a continuous variable, multivariable linear regression was conducted, with adjustment for a pre-specified list of covariates selected a priori based on primary literature and expert knowledge. Covariates included country, sex, race, AA type, Severity of Alopecia Tool (SALT) score at index, scalp hair loss ≥ 50% at diagnosis, eyebrow involvement, eyelash involvement, index year, presence of concomitant dermatologic conditions, presence of comorbid anxiety, and presence of comorbid depression. Regression coefficients (β, standard errors [SEs], and P values) were reported.

To assess the effect of age group on DLQI category, DLQI score was discretized into three mutually exclusive categories (none to moderate effect [DLQI score 0–10], very large effect [DLQI score 11–20], and extremely large effect [DLQI score 21–30]), with the none to moderate effect category serving as the reference group. Relative risks (RRs) and 95% CI between age group and DLQI category were estimated using modified Poisson regression analyses with a log link and robust SEs. The first model compared the RR of being in the very large effect category vs the none to moderate effect category, and the second model compared the RR of being in the extremely large effect category vs the none to moderate effect category. Both models were adjusted for the covariates described above. Two-sided P values < 0.05 were considered statistically significant.