How do fitness measures affect dementia risk? We investigated two mechanistic pathways: circulating proteins and structural brain changes.
Multivariable linear regression of each fitness measure against 2,918 Olink plasma proteins (Model 3 covariates). Select exposure to query the corresponding table; filter by FDR threshold and Beta direction.
Step 1b stratified analyses. Select Stratum: Age uses the age table; Male or Female uses the sex table. Filter by FDR, Beta direction, and Protein (All = traverse table).
Elastic Net Cox regression identified the most predictive proteins for dementia. Query by Exposure and Protein; results show all matching rows.
Cox models linking selected proteins to dementia risk. Query by Outcome, Exposure Context (multi-select), FDR threshold, HR direction, and Protein name.
Multivariable linear regression of fitness measures against 1,430 brain MRI structural features (IDPs) from UK Biobank. Select Exposure, filter by p_fdr threshold, and search for specific MRI features.
Causal mediation analysis testing whether brain MRI features mediate the fitness–dementia relationship. Select Exposure, filter by TNIE_p_formatted threshold, and search for specific Mediators.