Longitudinal course and phenotypes of health-related quality of life in adults with atopic dermatitis.

BACKGROUND The real-world course of health-related quality of life (HRQOL) in atopic dermatitis (AD) is not well established. OBJECTIVE To examine predictors, longitudinal course, and phenotypes of HRQOL in adult AD patients. METHODS A prospective dermatology-practice based study of 955 AD patients (age 18-97 years). Patients were assessed at baseline and approximately 6, 12, 18 and 24 months. Health-related quality of life was assessed using PROMIS Global Health (PGH) 10-item short-form. AD severity and impact was assessed by patient-reported global AD severity, Patient-Oriented Eczema Measure (POEM), Eczema Area and Severity Index (EASI), objective-SCORing AD (SCORAD), Investigator's Global Assessment (IGA), Numerical Rating Scale (NRS) average and worst-itch, PROMIS sleep-related impairment, and Patient Health Questionnaire (PHQ)-9. Repeated-measures regression models were constructed to examine itch over time. RESULTS In multivariable linear regression models controlling for age, race/ethnicity, history of asthma, hay fever, and food allergy, baseline PGH-physical T-scores were inversely associated with patient-reported global AD severity, POEM, EASI, objective-SCORAD, IGA, NRS average and worst-itch, PROMIS SRI, and PHQ-9, with stepwise decreases of physical health with worsening severity. PGH-mental T-scores were associated with all aforementioned severity measures aside from POEM. In multivariable repeated measures linear regression models, decreased PGH-mental and PGH-physical T-scores and mapped EQ-5D over time were associated with self-reported global AD severity, NRS worst and average-itch, POEM, PROMIS SD and SRI, EASI, objective-SCORAD, IGA, and PHQ-9 scores. Latent class analysis identified six classes of HRQOL, which were associated with measures of AD severity, non-white race, Hispanic ethnicity, and public insurance, but not age or sex. CONCLUSION AD patients have a heterogeneous longitudinal course and distinct patterns of HRQOL. Many patients had fluctuating HRQOL over time. Most patients with moderate-severe disease at baseline had persistent HRQOL impairment over time.