A digital intervention to protect cognitive health in older adults with cognitive impairment: a three-arm, open-label, randomised controlled trial.

BACKGROUND: Cognitive impairment affects health and social care. Intensive multicomponent interventions work but are impractical for primary care. The study aimed to trial brief approaches for cognitive health. METHODS: This was an open-label trial among adults aged 60-85 years with internet access and low cognitive scores (SD ≥1 below normal on the Baddeley verbal reasoning test, the cognitive criteria for age-associated-cognitive decline [AACD]). We used computer-generated random numbers to allocate participants to one of three groups: (1) Active Brains website (supporting physical activity, a Mediterranean diet, and cognitive exercises); (2) Active Brains plus brief support (three email or telephone sessions); or (3) control (evidence-based advice for cognitive health). The primary outcome was Baddeley verbal reasoning score at 12 months analysed by intention-to-treat with multiple imputation for missing data; follow-up for a co-primary outcome of dementia at 5 years is ongoing. The key secondary outcome was to describe the clinically important cutpoints of the Baddeley test. The trial is registered with International Standard Randomised Controlled Trial (ISRCTN17349359), is closed to new participants, and follow-up is ongoing. FINDINGS: Between Oct 3, 2020, and Oct 30, 2023, 11 563 participants from 670 practices were randomly assigned to either the active brains group (n=3169), active brains plus support (n=3087), or control (n=3927; intention-to-treat population). Participants were predominantly White (11 087 [95·8%]), 6697 (57·9%) were female, and 8142 (70·4%) were retired. Recruitment and 12-month follow-up (median 366 days in all groups) are complete. Baseline mean Baddeley verbal reasoning score was 15·1 (SD 6·4) in the control group, 15·3 (6·2) in the Active Brains group, and 15·1 (6·4) in the Active Brains plus support group. 12-month data were available for the primary outcome for 3113 (79·3%) of 3927 participants in the control group, 2564 (67·3%) of 3810 in the Active Brains group, and 2566 (67·1%) of 3826 in the Active Brains plus support group. At 12 months, mean Baddeley verbal reasoning score was 18·3 (SE 0·14) in the control group, 18·7 (SE 0·16) in the Active Brains group (adjusted mean difference vs control +0·50, 95% CI 0·13-0·87), and 19·2 (SE 0·16) in the Active Brains plus support group (+0·97, 0·59-1·34). This resulted in a substantial minority in all groups no longer meeting the cognitive criteria for AACD (883 [28·4%] of 3113 in the control group; 842 [32·8%] of 2566 in the Active Brains group; 811 [31·6%] of 2546 in the Active Brains plus support group), and fewer in the Active Brains plus support group (but not the Active Brains group) met cognitive criteria for mild or severe cognitive impairment. Serious adverse events and deaths, all unrelated to treatment, were similar in all groups. Serious adverse events occurred in 33 (0·87%) of 3810 in the Active Brains group, 32 (0·84%) of 3826 in the Active Brains plus support group, and 28 (0·71%) of 3927 in the control group. Deaths occurred in 20 (0·5%) in the Active Brains group, 17 (0·4%) in the Active Brains plus support group, and 19 (0·5%) in the control group. INTERPRETATION: Very brief intervention (eg, cognitive testing, documenting brain-stimulating activities, and evidence-based advice on brain health) in primary care is likely to improve cognitive health among individuals with low cognitive functioning. This study further supports the additional use of scalable online interventions for cognitive impairment in that require minimal support. Research is needed to assess the longer term effects of these interventions. FUNDING: National Institute for Health and Care Research Programme Grants for Applied Research.