Social isolation, lifestyle and cognitive functioning in midlife and early‐late life: Implications for risk reduction of cognitive decline and dementia

Abstract

AbstractBackgroundSocial isolation (SI) is considered a modifiable risk factor for cognitive decline and dementia, is associated with unhealthy lifestyle. We investigated independent and combined associations of SI and lifestyle with cognitive functioning (CF) in a midlife to early‐late life population without dementia. This informs developing more targeted preventive strategies against cognitive decline and dementia.MethodThe “LIfestyle for BRAin health” (LIBRA) score was computed for 6,203 baseline participants of the LIFE‐Adult‐Study, a population‐based German cohort. LIBRA is validated for dementia prediction in midlife to early‐late life populations, consisting of 12 modifiable factors (heart disease, kidney disease, diabetes, obesity, hypertension, hypercholesterolemia, alcohol consumption, smoking, physical inactivity, diet, depression, cognitive activity). Higher scores (‐5.9,+12.7) indicate “worse” lifestyle. SI was assessed with the short Lubben Social Network Scale (score < 12). For CF, we calculated a z‐standardized composite score of the Verbal Fluency Test, the Trail Making Test A and B. Associations of SI and LIBRA with CF were investigated using generalized linear modelling. Bootstrapped structural equation modelling (SEM) tested whether LIBRA mediated the association between SI and CF. Sampling weights were applied, variables z‐standardized for SEM.ResultsParticipants were M = 57.7 (SD = 12.0, range: 40‐79) years old; 50.7% were female; 15.8% were socially isolated. SI (β: ‐.17, 95%CI[‐.24,‐.10]; Wald = 21.45, p < .001) and a higher LIBRA score (β: ‐.03, 95%CI[‐.04,‐.02]; Wald = 31.49, p < .001) were independently associated with lower CF; there was no interaction (β: ‐.003, 95%CI[‐.03,.02]; Wald = 0.06, p = .814; Fig.1). Bootstrapped SEM (reps = 10,000; R² = .34) indicated that LIBRA partially mediated the association between SI and CF (IE = ‐.006, 95%CI[‐.008,‐.003], p < .001; proportion mediated = 8.5%; Fig.2).ConclusionCF is lower in socially isolated than socially integrated individuals without dementia. The CF difference can be partially attributed to modifiable health and lifestyle factors; however, only to a small extent. Therefore, public health initiatives aiming at preventing SI might be more effective in promoting good CF than lifestyle interventions in socially isolated individuals in midlife to early‐late life. This would likely benefit risk reduction of cognitive decline and dementia in later life.