Figure 4.

Considering the time since diagnosis when evaluating the relationship between cancer and Alzheimer’s disease (AD). Hazard rate ratios for the effect of cancer diagnosis on AD estimated using Cox regression. Models displayed are run separately by sex and stratified by birth year. No other covariates are included in the models. Results displayed have separate variables for prebaseline (prevalent) and postbaseline (incident) cancer diagnosis (Model 3 in Figure 4). dx 10+years = cancer diagnosis 10 or more years prior to baseline; dx 5–9 years = cancer diagnosis 5–9 years prior to baseline; dx 0–4 years = cancer diagnosis 0–4 years prior to baseline; TVEM = time-varying effect for cancer diagnosed after baseline. Source: Utah Population Database, Utah Cancer Registry, Centers for Medicare and Medicaid Services.