Trajectories of functional decline in older adults with neuropsychiatric and cardiovascular multimorbidity: A Swedish cohort study

Abstract

Background: Functional decline is a strong health determinant in older adults, and chronic diseases play a major role in this age-related phenomenon. In this study, we explored possible clinical pathways underlying functional heterogeneity in older adults by quantifying the impact of cardiovascular (CV) and neuropsychiatric (NP) chronic diseases and their co-occurrence on trajectories of functional decline.

Methods and findings: We studied 2,385 people ≥60 years (range 60-101 years) participating in the Swedish National study of Aging and Care in Kungsholmen (SNAC-K). Participants underwent clinical examination at baseline (2001-2004) and every 3 or 6 years for up to 9 years. We grouped participants on the basis of 7 mutually exclusive clinical patterns of 0, 1, or more CV and NP diseases and their co-occurrence, from a group without any CV and NP disease to a group characterised by the presence of CV or NP multimorbidity, accompanied by at least 1 other CV or NP disorder. The group with no CV and/or NP diseases served as the reference group. Functional decline was estimated over 9 years of follow-up by measuring mobility (walking speed, m/s) and independence (ability to carry out six activities of daily living [ADL]). Mixed-effect linear regression models were used (1) to explore the individual-level prognostic predictivity of the different CV and NP clinical patterns at baseline and (2) to quantify the association between the clinical patterns and functional decline at the group level by entering the clinical patterns as time-varying measures. During the 9-year follow-up, participants with multiple CV and NP diseases had the steepest decline in walking speed (up to 0.7 m/s; p < 0.001) and ADL independence (up to three impairments in ADL, p < 0.001) (reference group: participants without any CV and NP disease). When the clinical patterns were analyzed as time varying, isolated CV multimorbidity impacted only walking speed (β -0.1; p < 0.001). Conversely, all the clinical patterns that included at least 1 NP disease were significantly associated with decline in both walking speed (β -0.21--0.08; p < 0.001) and ADL independence (β -0.27--0.06; p < 0.05). Groups with the most complex clinical patterns had 5%-20% lower functioning at follow-up than the reference group. Key limitations of the study include that we did not take into account the specific weight of single diseases and their severity and that the exclusion of participants with less than 2 assessments may have led to an underestimation of the tested associations.

Conclusions: In older adults, different patterns of CV and NP morbidity lead to different trajectories of functional decline over time, a finding that explains part of the heterogeneity observed in older adults' functionality. NP diseases, alone or in association, are prevalent and major determinants of functional decline, whereas isolated CV multimorbidity is associated only with declines in mobility.

Fig 1. Flowchart of study participation over 9 years.

Dropouts are due to either refusal of the participant/relative, loss of contact with the participant, or moving of the participant from the city where the study took place. FU1, follow-up 1; FU2, follow-up 2; FU3, follow-up 3; SNAC-K, Swedish National study of Aging and Care in Kungsholmen.

Fig 2. Model showing the grouping of chronic diseases.

Participants were grouped by patterns of chronic diseases; that is, by number of CV and NP diseases. CV, cardiovascular; Mult., multimorbidity; NP, neuropsychiatric; Ref., reference group.

Fig 3. Trajectories of walking speed and ADL impairment over 9 years by clinical patterns.

Trajectories derived from multilevel mixed-effect linear regression models adjusted for baseline age, sex, education, malnutrition, institutionalization, and number of medications. Reference group: participants free from cardiovascular and/or neuropsychiatric diseases. ADL, activities of daily living; CV, cardiovascular; Multim., multimorbidity; NP, neuropsychiatric; Ref., reference group.

Fig 4. Association of chronic disease patterns with walking speed and number of intact ADL based on 6-year repeated measures of disease patterns, covariates, and outcomes.

Multilevel mixed-effect linear regression models adjusted for age, sex, education, malnutrition, institutionalization, and number of medications. The exposure (the patterns of chronic diseases), the time-varying covariates (institutionalization, malnutrition, and number of medications), and the outcomes were introduced in the models as repeated measures. Reference group: participants free from CV/NP diseases. ADL, activities of daily living; CI, confidence interval; CV, cardiovascular; Multim., multimorbidity; NP, neuropsychiatric; Ref., reference group.