This study aims to evaluate the comparative risk of dementia/Alzheimer's disease onset between patients treated with medications that target specific metabolic pathways and patients treated with alternative medications for the same indication.

This is a non-randomized, non-interventional study that is part of the DREAM Study of Brigham and Women's Hospital. DREAM is led by Dr. Madhav Thambisetty, MD, PhD, Chief of the Clinical and Translational Neuroscience Section, Laboratory of Behavioral Neuroscience, National Institute on Aging (NIA) intramural research program. This study aims to evaluate the comparative risk of dementia/Alzheimer's disease onset between patients treated with medications that target specific metabolic pathways and patients treated with alternative medications for the same indication using healthcare claims data.

Please see https://docs.google.com/spreadsheets/d/1N3IVoirGdWS7OiKiOWWgPHhgsSxX7SWDIhpeeZ-Ft7c/edit?usp=sharing or Appendix A for full code and algorithm definitions.

Medicare timeframe: 2007 to 2017 (end of data availability).

Inclusion Criteria:

• 1. No prior use of urate-lowering agents, including probenecid and allopurinol anytime prior to cohort entry date
• 2. Enrollment in Medicare Part A, B, and D with no HMO coverage for 365 days prior to and including cohort entry date

Exclusion Criteria:

• 1. Prior history of dementia measured anytime prior to cohort entry date
• 2. No prior history of gout recorded in the 365 days prior to cohort entry date
• 3. Prior history of nursing home admission in the 365 days prior to the cohort entry date

This study will involve a new user, parallel group, cohort study design comparing probenecid to allopurinol. The patients will be required to have continuous enrollment during the baseline period of 365 days before initiation of probenecid or allopurinol (cohort entry date). Follow-up for the outcome (incident dementia), begins the day after drug initiation for analyses 1, 3, and 4 and 180 days after drug initiation for analysis 2.

Study cohort after 1:1 propensity score matching

'As-treated' follow-up approach - Median follow-up days of 36 days (for Probenecid) and 214 days (for Allopurinol)

Our study did not capture Serious Adverse Events.