Identifying the optimal combinations of modifiable dementia risk factors to target in multidomain intervention - Three-year longitudinal findings from the Canadian longitudinal study on aging.

Background: Recent multidomain prevention trials for dementia have shifted toward more targeted approaches, focusing on specific combinations of risk factors and interventions at certain times. However, the optimal combinations of modifiable risk factors that can be targeted to maximize intervention effect remain unclear. Identifying risk factor combinations with the highest prevalence and largest effect sizes can enhance efficiency of trial design.

Objectives: To identify risk factor combinations that are both highly prevalent and have the most detrimental effect on cognition, and to assess their interaction effect and synergism.

Design: Longitudinal analysis of Canadian Longitudinal Study on Aging (CLSA).

Setting: Community.

Participants: 30,097 adults aged 45 to 85 at baseline MEASUREMENTS: The five most prevalent dyad, triad, and tetrad combinations of 12 modifiable risk factors were identified. Cognition was assessed with a composite Z-score from a neuropsychological test battery. Linear mixed effect models were used to examine the association between the identified combinations and 3-year cognitive changes. Interaction was assessed on additive scale, and synergism was explored.

Results: The combinations that were both highly prevalent and had the most detrimental effect on global cognition were: hearing loss and physical inactivity for the dyad (mean difference in change score = -0.07 SD; 95 % CI: -0.09 to -0.06; p < 0.001; effect size = -0.28), hearing loss, physical inactivity, and hypertension for the triad (mean difference in change score = -0.07; 95 % CI: -0.09 to -0.06; p < 0.001; effect size = -0.28), and hearing loss, physical inactivity, hypertension, and sleep disturbance for the tetrad (mean difference in change score = -0.05; 95 % CI: -0.07 to -0.03; p < 0.001; effect size = -0.20). Similar patterns were observed for memory and executive function. A significant synergistic interaction was observed between hearing loss and physical inactivity for global cognition (p = 0.005).

Conclusions: The combined effect of multiple risk factors varied by its combinations. The combination of hearing loss and physical inactivity offers a greater potential benefit than other dyad combinations. Hypertension and sleep disturbance can be further included for triad and tetrad combinations. Auditory health and exercise should be prioritized for multidomain interventions.