Estimating the causal effect of smoking cessation in the presence of confounding factors using a rank preserving structural failure time model.

Estimating the causal effect of quitting smoking on time to death or first myocardial infarction requires that one control for the differences in risk factors between individuals who elect to quite at each time t versus those who elect to continue smoking at time t. In this paper we examine the limitations of standard time varying Cox proportional hazards models to yield tests and estimates of this effect. Implementing the method of G-estimation proposed by Robins, we perform an observational analysis of data from the Multiple Risk Factor Intervention Trial (MRFIT) and estimate the causal effect of cigarette cessation while controlling for such time varying confounders as angina. We reject the null hypothesis of no effect of quitting on time to failure, and estimate that by quitting smoking, an individual increases by 50 per cent his time to death or first myocardial infarction (MI).

[1]  J. Robins.  A graphical approach to the identification and estimation of causal parameters in mortality studies with sustained exposure periods. , 1987, Journal of chronic diseases.

[2]  D. McLeish.  Dependent Central Limit Theorems and Invariance Principles , 1974 .

[3]  J. Robins.  Correcting for non-compliance in randomized trials using structural nested mean models , 1994 .

[4]  C. Seltzer.  Framingham study data and "established wisdom" about cigarette smoking and coronary heart disease. , 1989, Journal of clinical epidemiology.

[5]  I. Holme,et al.  The Oslo Study: diet and antismoking advice. Additional results from a 5-year primary preventive trial in middle-aged men. , 1985, Preventive medicine.

[6]  W. Newey,et al.  Semiparametric Efficiency Bounds , 1990 .

[7]  R. Kronmal,et al.  Effect of cigarette smoking on survival of patients with angiographically documented coronary artery disease. Report from the CASS registry. , 1986, JAMA.

[8]  J. Robins,et al.  A method for the analysis of randomized trials with compliance information: an application to the Multiple Risk Factor Intervention Trial. , 1993, Controlled clinical trials.

[9]  J. Robins.  A new approach to causal inference in mortality studies with a sustained exposure period—application to control of the healthy worker survivor effect , 1986 .

[10]  L H Kuller,et al.  The relationship of smoking cessation to coronary heart disease and lung cancer in the Multiple Risk Factor Intervention Trial (MRFIT). , 1990, American journal of public health.

[11]  J. Robins.  The control of confounding by intermediate variables. , 1989, Statistics in medicine.

[12]  Donald B. Rubin,et al.  Bayesian Inference for Causal Effects: The Role of Randomization , 1978 .

[13]  R. Benfari.  The multiple risk factor intervention trial (MRFIT). III. The model for intervention. , 1981, Preventive medicine.

[14]  J. Ockene.  Baseline rest electrocardiographic abnormalities, antihypertensive treatment, and mortality in the Multiple Risk Factor Intervention Trial. Multiple Risk Factor Intervention Trial Research Group. , 1985, The American journal of cardiology.

[15]  P. Holland.  Statistics and Causal Inference , 1985 .

[16]  J. Robins.  Addendum to “a new approach to causal inference in mortality studies with a sustained exposure period—application to control of the healthy worker survivor effect” , 1987 .

[17]  D. Petitti,et al.  Mortality in cigarette smokers and quitters. Effect of base-line differences. , 1981, The New England journal of medicine.

[18]  R. Kronmal,et al.  Beneficial six-year outcome of smoking cessation in older men and women with coronary artery disease. Results from the CASS registry. , 1988, The New England journal of medicine.

[19]  W. Kannel,et al.  STOPPING SMOKING AND C.H.D. , 1977, The Lancet.

[20]  J. Robins,et al.  Estimating exposure effects by modelling the expectation of exposure conditional on confounders. , 1992, Biometrics.

[21]  T. Vogt,et al.  The multiple risk factor intervention trial (MRFIT). V. Intervention on smoking. , 1981, Preventive medicine.

[22]  J. Cutler,et al.  Coronary heart disease and all-causes mortality in the Multiple Risk Factor Intervention Trial: subgroup findings and comparisons with other trials. , 1985, Preventive medicine.

[23]  W. Butts,et al.  Automated method for determining serum thiocyanate, to distinguish smokers from nonsmokers. , 1974, Clinical chemistry.

[24]  M. Shipley,et al.  A randomised controlled trial of anti-smoking advice: 10-year results. , 1982, Journal of epidemiology and community health.

[25]  J. Robins,et al.  G-Estimation of the Effect of Prophylaxis Therapy for Pneumocystis carinii Pneumonia on the Survival of AIDS Patients , 1992, Epidemiology.

[26]  S. Shapiro,et al.  The risk of myocardial infarction after quitting smoking in men under 55 years of age. , 1985, The New England journal of medicine.

[27]  J. Ockene.  Relationship between baseline risk factors and coronary heart disease and total mortality in the Multiple Risk Factor Intervention Trial. Multiple Risk Factor Intervention Trial Research Group. , 1986, Preventive medicine.

[28]  L. Dales,et al.  Characteristics predictive of coronary heart disease in ex-smokers before they stopped smoking: comparison with persistent smokers and nonsmokers. , 1979, Journal of chronic diseases.

[29]  L. Goldman,et al.  Long-term impact of smoking cessation on the incidence of coronary heart disease. , 1990, American journal of public health.

[30]  D. Cox,et al.  Analysis of Survival Data. , 1985 .

[31]  J. Robins,et al.  Identifiability and Exchangeability for Direct and Indirect Effects , 1992, Epidemiology.

[32]  J. Robins.  Estimation of the time-dependent accelerated failure time model in the presence of confounding factors , 1992 .

[33]  D. Cox.  Regression Models and Life-Tables , 1972 .

[34]  K Y Liang,et al.  Longitudinal data analysis for discrete and continuous outcomes. , 1986, Biometrics.

[35]  D. Rubin,et al.  The central role of the propensity score in observational studies for causal effects , 1983 .

[36]  W. Kannel,et al.  Death and coronary attacks in men after giving up cigarette smoking. A report from the Framingham study. , 1974, Lancet.

[37]  K. Warner.  Health and economic implications of a tobacco-free society. , 1987, JAMA.

[38]  P. Rosenbaum.  Conditional Permutation Tests and the Propensity Score in Observational Studies , 1984 .

[39]  J. Ockene.  Coronary heart disease death, nonfatal acute myocardial infarction and other clinical outcomes in the Multiple Risk Factor Intervention Trial. Multiple Risk Factor Intervention Trial Research Group. , 1986, The American journal of cardiology.