Randomized trials with patient-reported outcomes are commonly plagued by missing data. The analysis of such trials relies on untestable assumptions about the missing data mechanism. To address this issue, it has been recommended that the sensitivity of the trial results to assumptions should be a mandatory reporting requirement. In this paper, we discuss a recently developed methodology (Scharfstein et al., Biometrics, 2018) for conducting sensitivity analysis of randomized trials in which outcomes are scheduled to be measured at fixed points in time after randomization and some subjects prematurely withdraw from study participation. The methodology is explicated in the context of a placebo-controlled randomized trial designed to evaluate a treatment for bipolar disorder. We present a comprehensive data analysis and a simulation study to evaluate the performance of the method. A software package entitled SAMON (R and SAS versions) that implements our methods is available at www.missingdatamatters.org.
FieldingSMaclennanGCookJAet al.A review of RCTS in four medical journals to assess the use of imputation to overcome missing data in quality of life outcomes. Trials2008; 9: 51–51.
2.
LittleRCohenMDickersinKet al.The prevention and treatment of missing data in clinical trials, The National Academies Press, 2010.
3.
LittleRJD’AgostinoRCohenMLet al.The prevention and treatment of missing data in clinical trials. N Engl J Med2012; 367: 1355–1360.
4.
LiTHutflessSScharfsteinDOet al.Standards should be applied in the prevention and handling of missing data for patient-centered outcomes research: a systematic review and expert consensus. J Clin Epidemiol2014; 67: 15–32.
5.
ScharfsteinDMcDermottAOlsonWet al.Global sensitivity analysis for repeated measures studies with informative drop-out. Stat Biopharm Res2014; 6: 338–348.
6.
CopasJEguchiS. Local sensitivity approximations for selectivity bias. J R Stat Soc B2001; 63: 871–895.
7.
LiuTHeitjanDF. Sensitivity of the discrete-time Kaplan–Meier estimate to nonignorable censoring: application in a clinical trial. Stat Med2012; 31: 2998–3010.
8.
MaGToxelABHeitjanDF. An index of local sensitivity to nonignorable drop-out in longitudinal modelling. Stat Med2005; 24: 2129–2150.
9.
TroxelABMaGHeitjanDF. An index of local sensitivity to nonignorability. Stat Sinica2004; 14: 1221–1237.
10.
VerbekeGMolenberghsGThijsHet al.Sensitivity analysis for nonrandom dropout: a local influence approach. Biometrics2001; 57: 7–14.
11.
LittleRJARubinDB. Statistical analysis with missing data, Hoboken, New Jersey: John Wiley & Sons, 2014.
12.
DanielsMJHoganJW. Missing data in longitudinal studies: strategies for Bayesian modeling and sensitivity analysis, Boca Raton, Florida: CRC Press, 2008.
13.
RobinsJMRotnitzkyAScharfsteinDOSensitivity analysis for selection bias and unmeasured confounding in missing data and causal inference models. In: HalloranE (ed). Statistical models for epidemiology, New York: Springer-Verlag, 2000, pp. 1–94.
14.
RotnitzkyARobinsJMScharfsteinDO. Semiparametric regression for repeated outcomes with non-ignorable non-response. J Am Stat Assoc1998; 93: 1321–1339.
15.
RotnitzkyAScharfsteinDOSuTLet al.A sensitivity analysis methodology for randomized trials with potentially non-ignorable cause-specific censoring. Biometrics2001; 57: 103–113.
16.
ScharfsteinDOMcDermottADiazIet al.Global sensitivity analysis for repeated measures studies with informative drop-out: a semi-parametric approach. Biometrics2018; 74: 207–219.
17.
ScharfsteinDORotnitzkyARobinsJM. Adjusting for non-ignorable drop-out using semiparametric non-response models (with discussion). J Am Stat Assoc1999; 94: 1096–1146.
18.
CampbellGPennelloGYueL. Missing data in the regulation of medical devices. J Biopharm Stat2011; 21: 180–195.
19.
YanXLeeSLiN. Missing data handling methods in medical device clinical trials. J Biopharm Stat2009; 19: 1085–1098.
20.
CalabreseJRKeckPE JrMacfaddenWet al.A randomized, double-blind, placebo-controlled trial of quetiapine in the treatment of bipolar i or ii depression. Am J Psychiatry2005; 162: 1351–1360.
21.
EndicottJNeeJHarrisonWet al.Quality of life enjoyment and satisfaction questionnaire. Psychopharmacol Bull1993; 29: 321–326.
22.
Barndorff-NielsenOCoxDR. Edgeworth and saddle-point approximations with statistical applications. J R Stat Soc B1979; 41: 279–312.
23.
DigglePKenwardMG. Informative drop-out in longitudinal data analysis. Appl Stat1994; 43: 49–93.
24.
HallPRacineJLiQ. Cross-validation and the estimation of conditional probability densities. J Am Stat Assoc2004; 99: 1015–1026.
25.
EfronBGongG. A leisurely look at the bootstrap, the jackknife, and cross-validation. Am Stat1983; 37: 36–48.
26.
EfronBSteinC. The jackknife estimate of variance. Ann Stat1981; 9: 586–596.
27.
VansteelandtSGoetghebeurEKenwardMGet al.Ignorance and uncertainty regions as inferential tools in a sensitivity analysis. Stat Sinica2006; 16: 953–979.
28.
HuberPJRonchettiEM. Robust statistics, Hoboken, New Jersey: Wiley, 2009.
29.
RobinsJM. Non-response models for the analysis of non-monotone non-ignorable missing data. Stat Med1997; 16: 21–37.
30.
Tsiatis AA. Semiparametric theory and missing data. 2006, New York: Springer Verlag, 2006.
