Demographic and clinical characteristics of the study cohort at index date

Comparison of medication purchases in the 10 years preceding the index date in the AD and control groups of the LHS cohort

Cumulative incidence of dementia across matched cohorts in TriNetX U.S. data.

Panels A-C show reduced dementia incidence among individuals previously treated with atovaquone/proguanil (A+P) compared to matched controls across three age groups. Panel D shows increased dementia incidence among Toxoplasma gondii seropositive vs. seronegative individuals. Panels E and F stratify the A+P dementia risk by prior varicella-zoster vaccination status. Shaded bands represent 95% confidence intervals. All cohorts were matched 1:1 on age, gender, race/ethnicity, BMI, diabetes, and smoking status. “Control” refers to individuals without A+P exposure (A-C) or with any matched medication use (E, F).

Outcome Comparisons Studies in TriNetX U.S. cohort

Dementia outcome comparisons in matched cohorts from the TriNetX U.S. network. This table summarizes hazard ratios (HR), confidence intervals, and p-values for dementia incidence across multiple matched comparisons. Subplots A-F correspond to specific analyses displayed in Figure 1. Comparisons include patients exposed to atovaquone/proguanil across three age strata (50-59, 60-69, and 70-79), individuals with positive vs. negative Toxoplasma gondii serology (ages 60-75), and stratification by zoster vaccination status at age 70. All cohorts were matched 1:1 by propensity scores on demographics and clinical covariates. Dementia was defined by ICD-10 codes G30 or F03.