Likelihood Ratio-Based Tests for Longitudinal Safety Data

On October 24, 2014, the Engelberg Center for Health Care Reform at Brookings hosted a webinar, “Likelihood Ratio-Based Tests for Longitudinal Drug Safety Data.” This webinar highlighted a recent development of longitudinal likelihood ratio test (LongLRT) methods with application to a pooled clinical trial dataset for drugs treating osteoporosis with concomitant use of proton pump inhibitors (PPIs). The assessment also considered the special case of LongLRT, referred to as sequential LRT (SeqLRT), which is utilized when evaluating a signal of an adverse event for a particular drug compared with a placebo or a comparator. Results suggested that concomitant use of PPIs could lead to more adverse events associated with osteoporosis. The performance of the proposed LongLRT and SeqLRT methods was evaluated using simulated datasets and shown to retain power and control of type I error over time. The proposed methods can also be applied to large observational databases with exposure information under the U.S. Food and Drug Administration’s Sentinel Initiative for active surveillance.

Presenters included Dr. Ram Tiwari, Associate Director, Office of Biostatistics in CDER at the U.S. Food and Drug Administration and Dr. Lan Huang, Mathematical Statistician, Office of Biostatistics in CDER at the U.S. Food and Drug Administration.

For more information about Brookings’s active product surveillance projects, please visit Brookings’s Sentinel project page.