Structured vs Self-Guided Multidomain Lifestyle Interventions for Global Cognitive Function

Importance Identifying new interventions to slow and prevent cognitive decline associated with dementia is critical. Nonpharmacological interventions targeting modifiable risk factors are promising, relatively low-cost, accessible, and safe approaches. Objective To compare the effects of two 2-year lifestyle interventions on cognitive trajectory in older adults at risk of cognitive decline and dementia. Design, Setting, and Participants Single-blind, multicenter randomized clinical trial enrolling 2111 participants from May 2019 to March 2023 (final follow-up, May 14, 2025) at 5 clinical sites in the US. Participant inclusion criteria enriched risk of cognitive decline and included age 60 to 79 years, sedentary lifestyle, and suboptimal diet plus at least 2 additional criteria related to family history of memory impairment, cardiometabolic risk, race and ethnicity, older age, and sex. Interventions Participants were randomly assigned with equal probability to structured (n = 1056) or self-guided (n = 1055) interventions. Both interventions encouraged increased physical and cognitive activity, healthy diet, social engagement, and cardiovascular health monitoring, but differed in structure, intensity, and accountability. Main Outcomes and Measures The primary comparison was difference between intervention groups in annual rate of change in global cognitive function, assessed by a composite measure of executive function, episodic memory, and processing speed, over 2 years. Results Among the 2111 individuals enrolled (mean age, 68.2 [SD, 5.2] years; 1455 [68.9%] female), 89% completed the year 2 assessment. The mean global cognitive composite z score increased from baseline over time in both groups, with a mean rate of increase per year of 0.243 SD (95% CI, 0.227-0.258) for the structured intervention and 0.213 SD (95% CI, 0.198-0.229) for the self-guided intervention. The mean rate of increase per year was statistically significantly greater for the structured group than the self-guided group by 0.029 SD (95% CI, 0.008-0.050; P = .008). Based on prespecified secondary subgroup comparisons, the structured intervention benefit was consistent for APOE ε4 carriers and noncarriers ( P = .95 for interaction) but appeared greater for adults with lower vs higher baseline cognition ( P = .02 for interaction). Fewer ascertained adverse events were reported in the structured group (serious: 151; nonserious: 1091) vs the self-guided group (serious: 190; nonserious: 1225), with a positive COVID-19 test result being the most common adverse event overall and more frequent in the structured group. Conclusions and Relevance Among older adults at risk of cognitive decline and dementia, a structured, higher-intensity intervention had a statistically significant greater benefit on global cognition compared with an unstructured, self-guided intervention. Further investigation of functional outcomes, biomarkers, and ongoing extended follow-up will help address clinical relevance and sustainability of the observed cognitive benefits. Trial Registration ClinicalTrials.gov Identifier: NCT03688126