Interval From Cancer Diagnosis to Depression Onset and Risk of Dementia in Cancer Survivors.

BACKGROUND: To examine how the timing of depression after cancer diagnosis relates to the risk of all-cause dementia (ACD) and Alzheimer disease (AD), and to compare time-fixed and time-varying modeling approaches. METHODS: We conducted a nationwide cohort study using Korean National Health Insurance Service data on 385,092 individuals aged ≥55 years diagnosed with cancer between 2009 and 2013, with no dementia or depression, followed through 2023. Depression was identified using ICD-10 codes F32-F33 and categorized by interval from cancer diagnosis. Depression was modeled as a baseline time-fixed or a time-varying exposure. The primary outcome was ACD and the secondary outcome was AD. Cox proportional hazards models estimated adjusted hazard ratios (aHRs) and 95% confidence intervals (CIs), adjusting for demographic, lifestyle, and cardiometabolic factors. RESULTS: Over a median follow-up of 11.7 years, 71,600 participants (18.6%) developed ACD. Time-fixed models suggested increased dementia risk for depression diagnosed within 5 years after cancer, but an apparent inverse association for depression diagnosed >5 years after cancer (aHR, 0.90; 95% CI, 0.87-0.93). In contrast, time-varying models showed consistently elevated risks across all depression-onset intervals, including >5 years after cancer (aHR, 2.09; 95% CI, 2.02-2.17). Results were similar for AD. CONCLUSIONS: Depression at any time after cancer diagnosis is associated with increased dementia risk when modeled. Time-fixed analyses may introduce immortal time bias and mischaracterize risk, underscoring the importance of time-varying approaches in survivorship research. IMPACT: Modeling post-cancer depression as a time-varying exposure clarifies dementia risk estimates and informs survivorship mental health strategies.