Mortality among Older Adults Across Multimorbidity Categories and Lifestyle Patterns

Multimorbidity is common in older adults, and certain combinations of chronic conditions may confer higher mortality risk. Unhealthy lifestyle behaviours are also linked to shorter life expectancy. This study examined whether lifestyle patterns (LPs) modify the association between multimorbidity configurations (MCs) and mortality. We analysed data from 20,853 adults aged 60 and older in the Northern Netherlands Lifelines cohort, followed for a mean of 12 years. Five LPs were previously identified via latent class analysis, based on diet, physical activity, substance use, sleep, social connection, and stress. Multimorbidity was defined both as a disease count (≥2 non-communicable diseases [NCDs]) and as five latent MCs reflecting distinct NCD combinations. All-cause mortality was estimated using Kaplan-Meier plots and Cox models, reporting hazard ratios (HRs) with 95% confidence intervals (CIs), stratified by LPs. Compared to having no NCDs, mortality risk differed across MCs, with associations varying by LPs. Among participants with a ‘Healthy in a balanced way’ lifestyle, the ‘Complex-Treatment’ (HR 3.18, 95%CI: 2.24-4.51) and ‘CVD-&-Vascular’ (HR 2.35, 95%CI: 1.97-2.79) configurations showed the highest risks. In the ‘Unhealthy but no substance use’ group, mortality risk across MCs was more heterogeneous, with larger effect sizes. In contrast, multimorbidity defined by disease count showed limited variation in effect sizes across other LPs. LPs shape the mortality risk associated with multimorbidity. Risk variability across MCs was more pronounced in healthier lifestyles. These findings support the value of considering specific multimorbidity profiles—beyond disease count—for prognostic assessment and targeted interventions in older adults.