From Multimorbid to General Population, Documenting The Trajectories of Chronic Diseases Multimorbidity Patterns at Older Ages

Hoang Khanh Linh Dang
Nicola Caranci, General Directorate for Health and Social Care, Emilia-Romagna Region, Bologna, Italy
Giulia Roli, University of Bologna
Rosella Rettaroli, University of Bologna
Rossella Miglio, Department of Statistical Sciences "Paolo Fortunati", University of Bologna

In aging populations, to support the transition from healthcare centering around one single disease to a patient-centered approach, the possibility of identifying precise multimorbidity patterns and seizing their trajectories in time became increasingly urgent. Using longitudinal data of individuals aged 50 and above residing Emilia-Romagna region (northern Italy) in 2011 and followed up to 2019, our study aims to document the multimorbidity patterns transition leading to death at older ages across three time-points (in 2011, 2016 and 2019). Our analysis is structured in three steps, corresponding to three questions: (i) what are the multimorbidity patterns at each time-point? (ii) how can we document multimorbidity patterns transition in data at individual level? (iii) how can we model this transition in patterns across time to extract meaningful information? First, we combine mixed graphical model and network analysis to identify coherent clusters of chronic diseases at each time point. Second, we propose a scoring method using network metrics to assign one unique multimorbidity pattern to each individual at each time-point. Third, we apply the hidden Markov model to the transition process between multimorbidity patterns. We stratify our population by sex and age groups (50-59, 60-69, 70-79 and 80+) and conduct the three-steps analyses systematically on all 8 subgroups, starting with the population with multimorbidity, before extending to the general population. To our best knowledge, this paper figures among the first studies to investigate the trajectories of multimorbidity patterns under network perspective across time with an objective to extend to the general population.

See extended abstract

 Presented in Session P5. Health, Mortality, and Ageing 1