Abstract
OBJECTIVE: To characterize the heterogeneity of Long COVID (LC) by identifying distinct patient profiles based on symptoms and quality of life (QoL), and to examine the sociodemographic and clinical predictors associated with these profiles. STUDY DESIGN: A cross-sectional observational study was conducted. METHODS: We recruited 363 patients with LC in Spain via an online survey. Symptom patterns were identified through latent class analysis of 15 binary symptoms. QoL was assessed with the patient-derived LC-6D-QoL across six dimensions, and cluster analysis defined QoL subgroups. Logistic regression was applied to examine clinical and sociodemographic predictors of QoL profiles. RESULTS: Two symptom profiles emerged: a low-burden profile, dominated by fatigue and cognitive problems, and a high-burden profile with multisystem involvement. QoL clustered into three profiles-high, middle, and low QoL-with more than half of participants in the low QoL group. Symptom burden and employment status were the strongest predictors of poor QoL, whereas age, sex, education, and income showed limited associations. Social support was more frequently reported among participants with low QoL. CONCLUSIONS: LC is characterized by distinct clinical and QoL profiles, with strong interactions between multisystem symptom burden and social determinants. Identifying patients at greatest risk of poor QoL can inform stratified interventions and integrated policies that combine medical care, psychosocial support, and workplace reintegration.