The Hospital Anxiety and Depression Scale in People Living With HIV: A Validity Study in a Hard-to-Reach Population

While the Hospital Anxiety and Depression Scale (HADS) is widely used in diverse populations, countries, and languages, its scores' factor structure remains under debate. Only three studies have examined this in People Living with HIV (PLWH), with none conducted in Europe, Latin America, or Spanish-speaking samples. This study aimed to test five possible structural models of the HADS scores (one-, two-, and three-factor, and bifactor models with two and three factors) in Spanish-speaking PLWH, and to examine evidence of reliability and sensitivity.

A convenience sample of 580 PLWH was recruited via online dissemination and health center referrals from Spain, Mexico, Colombia, and other Latin American countries. Participants completed the HADS and provided information on sociodemographic and medical variables (e.g., gender, age, means of HIV transmission). Confirmatory factor analyses were conducted, McDonald's omega coefficients computed for internal consistency, and independent-samples t-tests performed for sensitivity analyses.

A bifactor solution, comprising two specific factors (Anxiety and Depression) along with a strong general Distress factor, best fitted the data (CFI = 0.99, TLI = 0.98, RMSEA = 0.03, SRMR = 0.02). The Anxiety, Depression, and Distress scales showed good internal consistency (ω = 0.89, 0.87, and 0.93, respectively) and were generally able to distinguish between groups (men-women, employed-unemployed, connected-not connected to an HIV-related group).

The HADS scores appear multidimensional, reliable, and sensitive in Spanish-speaking PLWH. However, their ability to reliably measure anxiety and depression separately is in question and warrants further investigation.