OBJECTIVES: FDA guidance documents describe the agency’s current thinking on a regulatory issue and are not legally binding on the public or the FDA itself. This study examines the impact of guidances set forth by the FDA in 2013 and 2018 for primary endpoint selection in Alzheimer’s disease (AD) trials. The 2013 guideline recommended cognitive/functional composite instruments as suitable tools for early AD, suggesting the Clinical Dementia Rating – Sum of Boxes (CDR-SB) as an example. The 2018 guideline revoked this stance for early-stage AD trials, amidst controversy whether a composite instrument could serve as a proper AD endpoint. METHODS: We utilized a regression discontinuity design framework to examine trialist response to these guidances. Our logistic models evaluated the presence of cognitive/functional composite endpoints in AD trials starting after each of the guidances. Models were conducted by AD stage, and controlled for disease-modifying therapy status and FDA registration status. This study utilized Phase 3 AD trial data through June 2020 from TrialTrove, which is a global registry of trials. RESULTS: We find that the rate of usage of cognitive/functional composite endpoints significantly increased after the 2013 guidance in pre-onset AD trials (OR = 2.01, p = 0.03; unit in years), suggesting adherence to the 2013 guidance, and significantly decreased after the 2018 guidance (OR = 0.11, p < 0.01; unit in years). Interestingly, this effect of the 2013 guideline is observed in the overt AD trials as well (OR = 1.51, p = 0.03; unit in years), which suggests a spillover effect. These findings also hold for CDR-SB. We also find evidence that privately-sponsored trials may be more adherent to such nonbinding recommendations than publicly-sponsored trials. CONCLUSIONS: Despite not setting new legal standards or imposing new requirements, FDA guidelines have an effect on trialist behavior, particularly in the private sector.