Estimating treatment effects from randomized clinical trials with noncompliance and loss to follow-up: the role of instrumental variable methods

Abstract

Perfectly implemented randomized clinical trials, particularly of complex interventions, are extremely rare. Almost always they are characterized by imperfect adherence to the randomly allocated treatment and variable amounts of missing outcome data. Here we start by describing a wide variety of examples and then introduce instrumental variable methods for the analysis of such trials. We concentrate mainly on situations in which compliance is all or nothing (either the patient receives the allocated treatment or they do not - in the latter case they may receive no treatment or a treatment other than the one allocated). The main purpose of the review is to illustrate the use of latent class (finite mixture) models, using maximum likelihood, for complier-average causal effect estimation under varying assumptions concerning the mechanism of the missing outcome data.

Get full access to this article

View all access options for this article.

References

Efron B. Foreword . Statistics in Medicine 1998; 17: 249-250 .

Holland PW. Causal inference, path analysis, and recursive structural equations models (with discussion). In: Clogg CC ed. Sociological methodology 1988. Washington DC: American Sociological Association , 1988: 449-493.

Zelen M. A new design for randomized clinical trials . New England Journal of Medicine 1979; 300: 1242-1245 .

Zelen M. Randomized consent designs for clinical trials: an update . Statistics in Medicine 1990; 9: 645-656 .

Fiegl P , Blumenstein B , Thompson I et al. Design of the Prostate Cancer Prevention Trial (PCPT) . Controlled Clinical Trials 1994; 16: 150-163 .

Baker SG. Analyzing a randomised cancer prevention trial with a missing binary outcome, an auxiliary variable, and all-or-none compliance . Journal of the American Statistical Association 2000; 95: 43-50 .

Little RJA , Rubin DB. Statistical analysis with missing data (2nd edition). Hoboken, NJ: John Wiley & Sons , 2002.

Cuzick J , Edwards R , Segnan N. Adjusting for non-compliance and contamination in randomized clinical trials . Statistics in Medicine 1997; 16: 1017-1029 .

Pearl J. Causality. Cambridge: Cambridge University Press , 2000.

10.

Matthews JNS. An introduction to randomized controlled clinical trials. London: Arnold , 2000.

11.

European Coronary Surgery Study Group . Coronary-artery bypass surgery in stable angina pectoris: survival at two years . The Lancet, 1979; i: 889-893 .

12.

Creed F , Mbaya P , Lancashire S , Tomenson B , Williams B , Holme S. Cost effectiveness of day and inpatient psychiatric treatment: results of a randomized controlled trial . British Medical Journal 1997; 314: 1381-1385 .

13.

Asberg M , Perris C , Schalling D , Sedvall G. The CPRS - development and applications of a psychiatric rating scale . Acta Psychiatrica Scandinavica 1978; Supplement 271.

14.

Permutt T , Hebel JR. Simultaneous-equation estimation in a clinical trial of the effect of smoking and birth weight . Biometrics 1989; 45: 619-622 .

15.

Sexton M , Hebel JR. A clinical trial of change in maternal smoking and its effect on birth weight . Journal of the American Medical Association 1984; 252: 911-915 .

16.

Sommer A , Tarwotjo I , Djunaedi E et al. Impact of vitamin A supplementation on childhood mortality: a randomized controlled community trial . Lancet 1986; i: 1169-1173 .

17.

Sommer A , Zeger SL. On estimating efficacy from clinical trials . Statistics in Medicine 1991; 10: 45-52 .

18.

Kemeny MM , Adak S , Gray B et al. Combined-modality treatment for respectable metastatic colorectal carcinoma to the liver: surgical resection of hepatic metastases from colorectal cancer . The New England Journal of Medicine 2002; 341: 2039-2048 .

19.

Loeys T , Goetghebeur E. A causal proportional hazards estimator for the effect of treatment actually received in a randomized trial with all-or-nothing compliance . Biometrics 2003; 59: 100-105 .

20.

Vinokur AD , Price RH , Schul Y. Impact of the JOBS intervention on unemployed workers varying in risk for depression . American Journal of Community Psychology 1995; 23: 39-74 .

21.

Dowrick C , Dunn G , Ayuso-Mateos J-L et al. and The Outcomes of Depression International (ODIN) Group . Problem solving treatment and group psychoeducation for depression: multicentre randomised controlled trial . British Medical Journal 2000; 321: 1450-1454 .

22.

Dunn G , Maracy M , Dowrick C et al. and The Outcomes of Depression International (ODIN) Group . Estimating psychological treatment effects from an RCT with both non-compliance and loss to follow-up . British Journal of Psychiatry 2003; 183: 323-331 .

23.

Beck A , Ward C , Mendelson M , Mock J , Erbaugh J. An inventory for measuring depression . Archives of General Psychiatry 1961; 4: 561-571 .

24.

Imbens G , Angrist JD. Identification and estimation of local average treatment effects . Econometrika 1994; 62: 467-476 .

25.

Angrist JD , Imbens GW , Rubin DB. Indentification of causal effects using instrumental variables (with discussion) . Journal of the American Statistical Association 1996; 91: 444-472 .

26.

Wooldridge JM. Econometric analysis of cross section and panel data. Cambridge, MA: MIT Press , 2002.

27.

Wooldridge JM. Introductory econometrics: a modern approach (2nd edition). Cincinatti, OH. South-Western College Publishing , 2003.

28.

Bloom HS. Accounting for no-shows in experimental evaluation designs . Evaluation Review 1984; 8: 225-246 .

29.

Robins J. A new approach to causal inference in mortality studies with sustained exposure periods - application to control of the healthy worker survivor effect . Mathematical Modelling 1986; 7: 1393-1512 .

30.

Robins J. The analysis of randomized and nonrandomized AIDS treatment trials using a new approach to causal inference in longitudinal studies. In: Sechrest L , Freeman H , Mulley A eds. Health services research methodology: a focus on AIDS. Washington DC: US Public Health Service , 1989: 113-159.

31.

Tarwotjo I , Sommer A , West KP et al. Influence of participation on mortality in a randomized trial of vitamin A prophylaxis . American Journal of Clinical Nutrition 1984; 45: 1466-1471 .

32.

Newcombe RG. Explanatory and pragmatic estimates of the treatment effect when deviations from allocated treatment occur . Statistics in Medicine 1988; 7: 1179-1186 .

33.

Efron B , Feldman D. Compliance as an explanatory variable in clinical trials . Journal of the American Statistical Association 1991; 86: 9-17 .

34.

Zeger SL , Liang K-Y. Comment: dose-response estimands . Journal of the American Statistical Association 1991; 86: 18-19 .

35.

Trichter D. Explanatory analysis of randomized studies . Biometrics 1996; 52: 1450-1456 .

36.

Trichter D. Reasoning about data with directed graphs . Statistics in Medicine 1999; 18: 2067-2076 .

37.

Nagelkerke N , Fidler V , Bersen R , Borgdorff M. Estimating treatment effects in randomized clinical trials in the presence of non-compliance . Statistics in Medicine 2000; 19: 1849-1864 .

38.

StataCorp . Stata Statistical Software: Release 8.0. College Station TX: Stata Corporation , 2003.

39.

Frangakis CE , Rubin DB. Addressing complications of intention-to-treat analysis in the combined presence of all-or-none treatment-noncompliance and subsequent missing outcomes . Biometrika 1999; 86: 365-379 .

40.

Dunn G. Estimating the causal effects of treatment . Epidemiologia e Psichiatria Sociale 2002; 11: 206-215 .

41.

Imbens GW , Rubin DB. Bayesian inference for causal effects in randomized experiments with noncompliance . Annals of Statistics 1997; 25: 305-327 .

42.

Imbens GW , Rubin DB. Estimating outcome distributions for compliers in instrumental variables models . Review of Economic Studies 1997; 64: 555-574 .

43.

Little R , Yau LHY. Statistical techniques for analyzing data from prevention trials: treatment of no-shows using Rubin’s causal model . Psychological Methods 1998; 3: 147-159 .

44.

Jo B , Muthén B. Modeling of intervention effects with noncompliance: a latent variable approach for randomized trials. In: Marcooulides GA , Schumacher RH eds. New developments and techniques in structural equation modeling. Lawrence Erlbaum Associates, Mahwah, NJ ; 2001: 57-87.

45.

Jo B , Muthén B. Longitudinal studies with intervention and noncompliance: estimation of causal effects in growth mixture modeling. In: Duan N , Reise S eds. Multilevel modeling: methodological advances, issues and applications. Mahwah, NJ: Lawrence Erlbaum Associates , 2003: 112-139.

46.

Yau LHY , Little RJ. Inference for the complier-average causal effect from longitudinal data subject to noncompliance and missing data, with application to a job training assessment for the unemployed . Journal of the American Statistical Association 2001; 96: 1232-1243 .

47.

Peng Y , Little RJA , Raghunathan TE. An extended general location model for causal inferences from data subject to noncompliance and missing values . Biometrics 2002; 60: 598-607 .

48.

Mealli F , Imbens GW , Ferro S , Biggeri A. Analyzing a randomized trial on breast self-examination with noncompliance and missing outcomes . Biostatistics 2004; 5: 207-222 .

49.

Jo B. Sensitivity of causal effects under ignorable and latent ignorable missing-data mechanisms, 2002. Available at www.StatModel.com.

50.

Jo B. Model misspecification sensitivity analysis in estimating causal effects of intervention with noncompliance . Statistics in Medicine 2002; 21: 3161-3181 .

51.

Jo B. Estimation of intervention effects with noncompliance: alternative model specifications . Journal of Educational and Behavioral Statistics 2002; 27: 385-409 .

52.

Mealli F , Rubin DB. Assumptions when analyzing randomized experiments with noncompliance and missing outcomes . Health Services & Outcomes Research Methodology 2003; 3: 225-232 .

53.

Hirano K , Imbens GW , Rubin DB , Zhou X-H. Assessing the effect of an influenza vaccine in an encouragement design . Biostatistics 2000; 1: 69-88 .

54.

Skrondal A , Rabe-Hesketh S. Generalized linear latent variable modeling: multilevel, longitudinal and structural equations models. London: Chapman & Hall=CRC , 2004.

55.

Muthén LK , Muthén BO. Mplus User’s Guide. Los Angeles, CA: Muthén & Muthén 1998-2004. See www.StatModel.com for details of Mplus Version 3.0.

56.

Angrist JD , Imbens G. Two-stage least squares estimation of average causal effects in models with variable treatment intensity . Journal of the American Statistical Association 1995; 90: 431-442 .

57.

Fischer-Lapp K , Goetghebeur E. Practical properties of some structural mean analyses of the effect of compliance in randomized trials . Controlled Clinical Trials 1999; 20: 531-546 .

58.

Cuzick J. Contamination and non-compliance in screening trials. In: Duffy SW , Hill C , Este `ve J eds. Quantitative methods for the evaluation of cancer screening. London: Arnold , 2001: 26-33.

59.

Branson M , Whitehead J. Estimating a treatment effect in survival studies in which patients switch treatment . Statistics in Medicine 2002; 21: 2449-2463 .

60.

Rabe-Hesketh S , Skrondal A , Pickles A. Reliable estimation of generalized linear mixed models using adaptive quadrature . Stata Journal 2002; 2: 1-21 .

61.

Robins JM. Correcting for non-compliance in randomized trials using structural nested means models . Communications in Statistics - Theory and Methods 1994; 23: 2379-2412 .

62.

Robins JM. Correction for non-compliance in equivalence trials . Statistics in Medicine 1998; 17: 269-302 .

63.

Robins JM. Structural nested failure time models. In: Armitage P , Colton T eds. Encylopedia of Biostatistics. Chichester: John Wiley & Sons , 1998: 4372-4378.

64.

Vansteelandt S , Goetghebeur E. Causal inference with generalized structural mean models . Journal of the Royal Statistical Society, Series B 2003; 65: 817-835 .