Back to Annual Meeting

Back to Annual Meeting

APHA Scientific Session and Event Listing

Recorded presentation

Stuart A. Gansky, MS, DrPH, Center to Address Disparities in Children's Oral Health, University of California, San Francisco, 3333 California St, Ste 495, San Francisco, CA 94143-1361, 415-502-8094, stuart.gansky@ucsf.edu

In prevention trials, it is important to monitor adverse events or side effects, even non-serious ones, to maintain safety of otherwise healthy participants.

Objective: To develop an easy to implement method to monitor adverse events.

Methods: A simple Bayesian approach to monitor the prevalence of specific adverse events was developed. Prior information about adverse event prevalances was be updated with the number of adverse events and number of non-adverse events during the trial. A 2-parameter beta distribution was used to update the prior distribution with the observed data to provide posterior distributions. One-sided 95% upper bound credible intervals were generated for the posterior distribution and compared to that for prior prevalence; if the posterior upper bound exceeds the prior upper bound, the adverse event rate should be examined more closely by trial monitors.

Results: This method was easy enough to implement in standard spreadsheet software. Sample sizes and frequencies from published trials, package inserts, etc was utilized to provide priors. The number of adverse events and number of non-adverse events during the trial were data used to update the priors. Posterior distributions were then calculated. Adverse event rates were flagged if the upper bound exceeded the prior prevalence upper bound.

Conclusions: A simple Bayesian rule can provide relatively easy prevention trial monitoring.

Support: US DHHS / NIH / NIDCR & NCMHD U54 DE 14251

Learning Objectives:

• Identify reasons for non-serious adverse event monitoring in prevention clinical trials
• Learn a simple Bayesian rule for monitoring non-serious adverse events in an ongoing prevention trial

Keywords: Clinical Trials, Biostatistics

Presenting author's disclosure statement:

Any relevant financial relationships? No

Recorded presentation