Subjective well-being and allostatic load in multimorbidity transitions: A multi-state survival analysis of three international longitudinal cohorts.

High allostatic load increased overall physical-psychological-cognitive multimorbidity risk, specifically driving transitions involving cognitive decline.

• High AL was associated with overall PPC-MM risk (HR = 1.66, 95% CI: 1.16–2.38)
• High AL specifically drove the None→physical-cognitive (P1C) transition (HR = 1.48)
• AL was derived from cumulative biomarkers across three national cohorts
• Sample comprised 5629 adults aged ≥45 free of baseline PPC-MM

Low subjective well-being acted as a broad risk factor, strongly predicting physical-psychological multimorbidity onset.

• Low SWB predicted the None→physical-psychological (P1P2) transition with HR = 2.14
• SWB was standardized across all three cohorts (HRS, ELSA, CHARLS)
• Low SWB was described as a 'generalized driver of multimorbidity' affecting multiple transition pathways
• SWB effects were distinct from those of allostatic load, particularly in non-cognitive pathways

The combination of high allostatic load and low subjective well-being tripled the risk of developing complex triple multimorbidity.

• Combined high AL and low SWB was associated with None→P1P2C (triple multimorbidity) transition HR = 3.23
• This combined risk was substantially greater than either factor alone
• The authors described this combination as creating 'a high-risk profile for complex multimorbidity'
• Multi-state survival models were used to estimate hazard ratios across four transition types: P1P2, P1C, P2C, and P1P2C

Meta-analysis confirmed high consistency of findings across the three international cohorts.

• Three cohorts included: Health and Retirement Study (US), English Longitudinal Study of Aging (UK), and China Health and Retirement Longitudinal Study (CHARLS)
• Meta-analysis of cohort-specific results showed high consistency across cohorts
• Data were harmonized across the three national cohorts for comparability
• Total sample was 5629 adults aged ≥45 free of baseline PPC-MM

Physical-psychological-cognitive multimorbidity was studied using multi-state survival models examining four distinct transition pathways.

• Transitions modeled included: None→P1P2 (physical-psychological), None→P1C (physical-cognitive), None→P2C (psychological-cognitive), and None→P1P2C (triple multimorbidity)
• Multi-state survival analysis allowed estimation of hazard ratios for each specific transition type
• Participants were required to be free of baseline PPC-MM at study entry
• AL and SWB were examined both independently and jointly in relation to multimorbidity transitions