中止理由別（AE とその他の理由で区別）の pattern imputation

この補完方法は，脱落パターンごとに欠測値を補完するのではなく，たとえば脱落理由として有害事象の影 響で脱落した人とそれ以外の理由で脱落した人でパターンを作り，欠測値の補完を行う(Ratitch et al., 2013)．

投与中止後の推移を想定しているため，対象とするestimandはeffectivenessとなる．有害事象で途中脱落した 試験治療群及び対照群の被験者の欠測データには，両群のベースラインの応答変数データで補完する．一方，

その他の理由で脱落した試験治療群/対照群の被験者の欠測値には，それぞれの群の試験を完了したパターンの 被験者のデータでそれぞれ補完をする(CCMV)．この補完方法の特徴は，特に疼痛の領域において使用される ベースラインのデータで単補完するBOCFに近いコンセプトをもっている．もし，被験者が薬剤の服用を続け ることができないならば，脱落後のその被験者に治療のベネフィットは与えられなかったと考えることは自然 である．このため，脱落後の被験者の欠測データはベースラインの応答変数データで補完し保守的な取扱いを する．BOCFでは単補完であり，補完値の不確実性を考慮していないことから，この補完方法はこの不確実性 を考慮した多重補完とBOCFを組み合わせた方法となっている．

 上記以外にも，中止理由ごとに別々の補完モデルを用いることや，残りの症例がMARを仮定した補完を 行うことなど，中止理由別のpatter imputaionの事例が出始めていることに注目されたい（計量生物セミナー，

2015）．

参考文献

[1] Ayele, B. T., Lipkovich, I., Molenberghs, G.&Mallinckrodt, C. H. (2014). A multiple-imputation-based approach to sensitivity analysis and effectiveness assesments in longitudinal clinical trials.Journal of Biopharmaceutical Statistics. 24, 211-228.

[2] Diggle, P.,&Kenward, M. G. (1994). Informative drop-out in longitudinal data analysis. Applied statistics, 49-93.

[3] Dmitrienko, A., Molenberghs, G., Chuang-Stein, C.,&Offen, W. W. (2005). Analysis of clinical trials using SAS: A practical guide. SAS Institute.

[4] Graham, J. W., Olchowski, A. E.&Gilreath, T. D. (2007). How many imputations are really needed? Some practical clarifications of multiple imputation theory. Prevention Science, 8, 206-213.

[5] Kenward, M. G., Molenberghs, G.,&Thijs, H. (2003). Pattern‐ mixture models with proper time dependence.

Biometrika, 90(1), 53-71.

[6] Li, K. H., Raghunathan, T. E.,&Rubin, D. B. (1991). Large-sample significance levels from multiply imputed data using moment-based statistics and an F reference distribution. Journal of the American Statistical Association, 86(416), 1065-1073.

[7] Little, R. J. (1993). Pattern-mixture models for multivariate incomplete data. Journal of the American Statistical Association, 88(421), 125-134.

[8] Little, R. J. (1994). A class of pattern-mixture models for normal incomplete data. Biometrika, 81(3), 471-483.

[9] Little, R. J. (1995). Modeling the drop-out mechanism in repeated-measures studies. Journal of the American Statistical Association, 90(431), 1112-1121.

[10] Little, R. J.&Yau, L. (1996). Intent-to-treat analysis for longitudinal studies with drop-outs. Biometrics, 1324-1333.

[11] Little, R. J. A. and Rubin, D. B. (2002).Statistical analysis with missing data. 2nd edition. John Wiley&Sons.

New York.

[12] Mallinckrodt, C. H. (2013).Preventing and treating missing data in longitudinal clinical trials: a practical guide. Cambridge University Press.

[13] Mallinckrodt, C., Roger, J., Chuang-Stein, C., Molenberghs, G., O Kelly, M., Ratitch, B., ...&Bunouf, P.

(2013). Recent Developments in the Prevention and Treatment of Missing Data. Therapeutic Innovation&

Regulatory Science, 2168479013501310.

[14] Molenberghs, G., Michiels, B., Kenward, M. G.,&Diggle, P. J. (1998). Monotone missing data and pattern‐ mixture models. Statistica Neerlandica, 52(2), 153-161.

[15] Molenberghs, G.,&Kenward, M. (2007). Missing data in clinical studies (Vol. 61). John Wiley&Sons.

[16] National Research Council (2010). The Prevention and Treatment of Missing Data in Clinical Trials.

Washington, DC: The National Academies Press.

[17] O’Kelly, M.,&Ratitch, B. (2014). Clinical Trials with Missing Data: A Guide for Practitioners. John Wiley&

Sons.

[18] Ratitch, B.,&O Kelly, M. (2011). Implementation of pattern-mixture models using standard SAS/STAT procedures. PharmaSUG, Paper-SP04.

[19] Ratitch, B., O’Kelly, M.,&Tosiello, R. (2013). Missing data in clinical trials: from clinical assumptions to statistical analysis using pattern mixture models.Pharmaceutical statistics, 12(6), 337-347.

[20] Royston, P.,&White, I. R. (2011). Multiple imputation by chained equations (MICE): implementation in Stata.J Stat Softw, 45(4), 1-20.

[21] Rubin, D.B. (1978), Multiple Imputations in Sample Surveys - A Phenomenological Bayesian Approach to Nonresponse, Proceedings of the Survey Research Methods Section of the American Statistical Association, 20-34.

[22] Rubin, D. B. (1987). Multiple imputation for nonresponse in surveys. John Wiley&Sons.

[23] Siddiqui, O. (2011). MMRM versus MI in dealing with missing data―a comparison based on 25 NDA data sets. Journal of Biopharmaceutical Statistics. 21, 423-436.

[24] Thijs, H., Molenberghs, G., Michiels, B., Verbeke, G.,&Curran, D. (2002). Strategies to fit pattern‐ mixture models. Biostatistics, 3(2), 245-265.

[25] Verbeke, G.,&Molenberghs, G. (2009). Linear mixed models for longitudinal data. Springer Science&

Business Media.

[26] Wu, M. C.,&Bailey, K. R. (1989). Estimation and comparison of changes in the presence of informative right censoring: conditional linear model. Biometrics, 939-955.

[27] White, I. R., Royston, P.,&Wood, A. M. (2011). Multiple imputation using chained equations: issues and guidance for practice. Statistics in medicine, 30(4), 377-399.

[28] Yuan, Y. (2011). Multiple imputation using SAS software. Journal of Statistical Software, 45(6), 1-25.

[29] Yuan, Y. (2014). Sensitivity Analysis in Multiple Imputation for Missing Data. In Proceedings of the SAS Global Forum 2014 Conference. (http://support.sas.com/resources/papers/proceedings14/SAS270-2014.pdf).

[30] 大江基貴，鵜飼裕之，大浦智紀，土居正明，藤原正和，横山雄一. (2015). (2)多重補完(Multiple

Imputation：MI)に基づく連続量経時データの解析. SASユーザー総会論文集.