Affiliations: [a] Department of Pharmacology and Toxicology, University of Toronto, Toronto, ON, Canada
| [b] Sandra Black Centre for Brain Resilience and Recovery, Hurvitz Brain Sciences Program, Sunnybrook Research Institute, Toronto, ON, Canada
Correspondence:
[*]
Correspondence to: Walter Swardfager, Department of Pharmacology and Toxicology, University of Toronto,
1 King’s College Circle, Toronto, ON, M5 S 1A8, Canada. E-mail: w.swardfager@utoronto.ca.
Abstract: Pharmacoepidemiologic studies using routinely collected data allow researchers to propose drugs for repurposing trials for dementia prevention or treatment. A recent cohort study reported a 54% lower dementia risk among users of sildenafil compared to users of certain cardiovascular medications. We caution that “confounding by indication” can arise when outcomes are compared between a drug of interest and an inappropriate comparator. Here, we emphasize important considerations in selecting an active comparator. We assess the implications of substantial risk of confounding by indication in pharmacoepidemiologic studies linking phosphodiesterase-5 inhibitors to lower dementia risk.
Keywords: Alzheimer’s disease, confounding by indication, dementia, pharmacoepidemiology, phosphodiesterase-5 inhibitors
