A review of the handling of missing longitudinal outcome data in clinical trials

Matthew Powney, Paula Williamson, Jamie Kirkham, Ruwanthi Kolamunnage-Dona

The aim of this review was to establish the frequency with which trials take into accountmissingness, and to discover what methods trialists use for adjustment in randomised controlledtrials with longitudinal measurements. Failing to address the problems that can arise from missingoutcome data can result in misleading conclusions. Missing data should be addressed as a means of asensitivity analysis of the complete case analysis results. One hundred publications of randomisedcontrolled trials with longitudinal measurements were selected randomly from trial publications fromthe years 2005 to 2012. Information was extracted from these trials, including whether reasons fordropout were reported, what methods were used for handing the missing data, whether there was anyexplanation of the methods for missing data handling, and whether a statistician was involved in theanalysis. The main focus of the review was on missing data post dropout rather than missing interimdata. Of all the papers in the study, 9 (9%) had no missing data. More than half of the papersincluded in the study failed to make any attempt to explain the reasons for their choice of missingdata handling method. Of the papers with clear missing data handling methods, 44 papers (50%)used adequate methods of missing data handling, whereas 30 (34%) of the papers used missing datamethods which may not have been appropriate. In the remaining 17 papers (19%), it was difficult toassess the validity of the methods used. An imputation method was used in 18 papers (20%) toaddress the problems caused by missingness. Multiple imputation methods were introduced in 1987and are an efficient way of accounting for missing data in general, and yet only 4 papers used thesemethods. All these 4 papers were published in the past four years. Out of the 18 papers which usedimputation, only 7 displayed the results as a sensitivity analysis of the complete case analysis results.61% of the papers that used an imputation explained the reasons for their chosen method. Just undera third of the papers made no reference to reasons for missing outcome data. There was littleconsistency in reporting of missing data within longitudinal trials; some papers provided great detailabout the missing data, while others made no reference to it.