Details

Autor(en) / Beteiligte

• Zhang, Yuqing
• Alyass, Akram
• Vanniyasingam, Thuva
• Sadeghirad, Behnam
• Flórez, Iván D
• Pichika, Sathish Chandra
• Kennedy, Sean Alexander
• Abdulkarimova, Ulviya
• Zhang, Yuan
• Iljon, Tzvia
• Morgano, Gian Paolo
• Colunga Lozano, Luis E
• Aloweni, Fazila Abu Bakar
• Lopes, Luciane C
• Yepes-Nuñez, Juan José
• Fei, Yutong
• Wang, Li
• Kahale, Lara A
• Meyre, David
• Akl, Elie A
• Thabane, Lehana
• Guyatt, Gordon

Titel

Reporting quality and optimal methods of handling participants with missing outcome data for continuous outcomes in randomized controlled trials: a systematic survey of the methods literature

Ist Teil von

• Journal of clinical epidemiology, 2017

Erscheinungsjahr

2017

Quelle

Elsevier ScienceDirect Journals

Beschreibungen/Notizen

• Abstract Objective To conduct 1) a systematic survey of the reporting quality of simulation studies dealing with how to handle missing participant data (MPD) in randomized control trials and 2) summarize the findings of these studies. Study Design and Settings We included simulation studies comparing statistical methods dealing with continuous MPD in RCTs addressing bias, precision, coverage, accuracy, power, type I error and overall ranking. For the reporting of simulation studies, we adapted previously developed criteria for reporting quality and applied them to eligible studies. Result Of 16,446 identified citations, the 60 eligible generally had important limitations in reporting, particularly in reporting simulation procedures. Of the 60 studies 47/60 addressed ignorable and 32/60 non-ignorable data. For ignorable missing data, mixed model was most frequently the best on overall ranking (9 times best, 34.6% of times tested) and bias (10, 55.6%). Multiple imputation also performed well. For non-ignorable data, mixed model was most frequently the best on overall ranking (7, 46.7%), and bias (8, 57.1%). Mixed model performance varied on other criteria. Last observation carried forward was very seldom the best performing, and for non-ignorable MPD frequently the worst. Conclusion Simulation studies addressing methods to deal with MPD suffered from serious limitations. The mixed model approach was superior to other methods in terms of overall performance and bias. Last observation carried forward performed worst.