Health-related quality of life profiles and trajectories of patients with post COVID-19 condition: a latent Markov model

BACKGROUND: The aim of this study was to identify subgroups with similar health states based on health-related quality of life (HRQoL) profile (i.e. differences in affected dimensions and severity of problems per dimension) among PCC patients, to examine transitions between these subgroups over time, and to examine the association with sociodemographic and medical characteristics. METHODS: In this longitudinal cohort study, data from 5,737 PCC patients, collected through two online surveys, were analysed using a latent Markov model. Dichotomized HRQoL scores per EQ-5D-5L dimension were used as indicators to determine latent states and covariates were added into the model. RESULTS: A model with six latent states was selected. The largest state, consisting of 31% of respondents, was characterized by a high probability of problems on all dimensions, whereas the smallest state (7%) was characterized by a low probability of problems on all dimensions, except usual activities. The remaining four states were all characterized by a high probability of problems on usual activities and pain/discomfort, but differed based on the probability of problems on the dimensions mobility, self-care and anxiety/depression. The probability of transitioning to a different state was low, with states with fewer affected dimensions being most stable. Women, younger respondents, those with a lower educational level, and those with comorbidity were more likely to be in states with more affected dimensions. CONCLUSIONS: The identification of different HRQoL profiles shows that there is substantial heterogeneity in the HRQoL dimensions that are affected and in observed changes between two measurement points.