(263) Methods used to manage missing data in recent analgesic clinical trials: ACTTION systematic review and recommendations

Abstract

Missing data in clinical trials can bias estimates of treatment effects. Statisticians and government agencies recommend making every effort to minimize missing data. When data are missing, however, they recommend pre-specifying a primary method to manage the missing data and using additional methods to examine the sensitivity of the results to different assumptions about the missingness mechanism. The objective of this study was to assess the frequency with which randomized clinical trials published in three major pain journals (i.e., European Journal of Pain, Journal of Pain, Pain) reported strategies to prevent missing data, the amount of data missing, and statistical methods to manage missing data. A total of 160 randomized clinical trials investigating treatments for pain, published between 2006 and 2012, were included. Approximately 2/3 of the trials reported at least one method to reduce the amount of missing data, the most common being allowing rescue medications. Fewer than 50% of articles provided sufficient information to determine how many participants provided complete data for the efficacy analyses. While only 8 articles (5%) reported no missing data, fewer than 50% reported any statistical method to manage missing data. Last observation carried forward (LOCF) was used most commonly (42%). Only 13 of the articles (8%) used more than one method to manage missing data. This systematic review demonstrates deficiencies in the reporting of missing data and methods to manage it. These deficiencies can be rectified by changes in, or better enforcement of, journal policies pertaining to the reporting of missing data and methods to manage them in clinical trials. Financial support for this project was provided by the ACTTION public-private partnership which has received research contracts, grants, or other revenue from the FDA, Astellas, Bristol-Myers Squibb, Collegium, Depomed, Eli Lilly, Horizon, Jazz, Johnson & Johnson, Mallinckrodt, Pfizer, Purdue, Zalicus, and other sources.