Trajectories of chronic multimorbidity patterns in older patients: MTOP study

Abstract

Background: Multimorbidity is associated with negative results and poses difficulties in clinical management. New methodological approaches are emerging based on the hypothesis that chronic conditions are non-randomly associated forming multimorbidity patterns. However, there are few longitudinal studies of these patterns, which could allow for better preventive strategies and healthcare planning. The objective of the MTOP (Multimorbidity Trajectories in Older Patients) study is to identify patterns of chronic multimorbidity in a cohort of older patients and their progression and trajectories in the previous 10 years.

Methods: A retrospective, observational study with a cohort of 3988 patients aged > 65 was conducted, including suspected and confirmed COVID-19 patients in the reference area of Parc Taulí University Hospital. Real-world data on socio-demographic and diagnostic variables were retrieved. Multimorbidity patterns of chronic conditions were identified with fuzzy c-means cluster analysis. Trajectories of each patient were established along three time points (baseline, 5 years before, 10 years before). Descriptive statistics were performed together with a stratification by sex and age group.

Results: 3988 patients aged over 65 were included (58.9% females). Patients with ≥ 2 chronic conditions changed from 73.6 to 98.3% in the 10-year range of the study. Six clusters of chronic multimorbidity were identified 10 years before baseline, whereas five clusters were identified at both 5 years before and at baseline. Three clusters were consistently identified in all time points (Metabolic and vascular disease, Musculoskeletal and chronic pain syndrome, Unspecific); three clusters were only present at the earliest time point (Male-predominant diseases, Minor conditions and sensory impairment, Lipid metabolism disorders) and two clusters emerged 5 years before baseline and remained (Heart diseases and Neurocognitive). Sex and age stratification showed different distribution in cluster prevalence and trajectories.

Conclusions: In a cohort of older patients, we were able to identify multimorbidity patterns of chronic conditions and describe their individual trajectories in the previous 10 years. Our results suggest that taking these trajectories into consideration might improve decisions in clinical management and healthcare planning.

Trial registration number: NCT05717309.

Keywords: Ageing; Chronic conditions; Cluster analysis; Longitudinal study; Multimorbidity; Older patients; Trajectories.

Fig. 1

Prevalence of chronic multimorbidity clusters defined at three different time points over ten years and their trajectories. Bar heights represent the number of patients belonging to the cluster and stripe heights represent patients moving from one cluster to another, in an independent manner for each of the two depicted cluster transitions.

Fig. 2

Prevalence of chronic multimorbidity clusters defined at three different time points over ten years and their trajectories, stratified by sex and age group. Bar heights represent the number of patients belonging to the cluster and stripe heights represent patients moving from one cluster to another, in an independent manner for each of the two depicted cluster transitions.