Observational data are often used to address prevention questions such as, “If alcohol initiation could be delayed, would that in turn cause a delay in marijuana initiation?” This question is concerned with the total causal effect of the timing of alcohol initiation on the timing of marijuana initiation. Unfortunately, when observational data are used to address a question such as the above, alternative explanations for the observed relationship between the predictor, here timing of alcohol initiation, and the response abound. These alternative explanations are due to the presence of confounders. Adjusting for confounders when using observational data is a particularly challenging problem when the predictor and confounders are time-varying. When time-varying confounders are present, the standard method of adjusting for confounders may fail to reduce bias and indeed can increase bias. In this paper, an intuitive and accessible graphical approach is used to illustrate how the standard method of controlling for confounders may result in biased total causal effect estimates. The graphical approach also provides an intuitive justification for an alternate method proposed by James Robins [Robins, J. M. (1998). 1997 Proceedings of the American Statistical Association, section on Bayesian statistical science (pp. 1–10). Retrieved from http://www.biostat.harvard.edu/robins/research.html; Robins, J. M., Hernán, M., & Brumback, B. (2000). Epidemiology, 11(5), 550–560]. The above two methods are illustrated by addressing the motivating question. Implications for prevention researchers who wish to estimate total causal effects using longitudinal observational data are discussed.
Similar content being viewed by others
Notes
In this paper, we assume that all variables in the subset \(\mathbf{O}\) are measured precisely. Consequently, we consider neither measurement error nor measurement models.
This model makes the proportional hazards assumption that the effect of alcohol initiation is the same at all time points. This modeling assumption is maintained throughout the paper; it has no bearing on the causal issues that are discussed throughout.
In addition, \(\beta_1\) would be interpreted as the direct causal effect of alcohol initiation on marijuana initiation. For simplicity and reasons of space, this is not elaborated upon in this paper.
Here, PPress\(_2\) is known as a collider in Panel B of Fig. 2 because it is affected by the \(\mathbf{U}_t\)s and Alc\(_1\). As Pearl notes, conditioning on colliders has the effect of inducing noncausal relations among parents of the collider (the \(\mathbf{U}_t\)s and Alc\(_1\)) that are otherwise not related causally. Because we are unable to condition further on the unobserved or unknown variables (the \(\mathbf{U}_t\)s) in our setting, the induced noncausal relations become part of the observed relation between the collider's observed parent (Alc\(_1\)) and the response (Mj\(_2\)).
The weighting method is also known as inverse-probability-of-treatment-weighting, or IPTW, where “treatment” refers to the exposure or putative cause.
The analyses presented here include only those participants who had no missing data on heart rate, performance IQ, verbal IQ, average sensation seeking, first peer pressure resistance measurement, and time of initiation of alcohol, cigarettes, conduct disorder, and other drug use. Therefore, this sample may not be representative of any subset of adolescents.
Last observation carried forward (LOCF) imputation was used only for peer pressure resistance because it was the only time-varying variable for which there was missing data among the participants in these analyses.
Sensation Seeking is conceived as a relatively stable personality trait (Zuckerman, 1994), thus, the average score across assessments is used in these analyses. Empirically, sensation seeking is quite stable; in the larger sample from which these data are drawn, the 1-year stabilities for sensation seeking approach the maximum correlation possible given the reliabilities of the scales (average 1-year stability=.70).
Robins (1998) shows that confidence intervals according to the robust standard errors calculated in this way have coverage probability of at least 95%; that is, confidence intervals using this method are (possibly) conservative.
These assumptions, concerning unmeasured variables, are by definition not testable.
REFERENCE
Agerbo, E. (2005). Effect of psychiatric illness and labour market status on suicide: A healthy worker effect? Journal of Epidemiology and Community Health, 59(7), 598–602.
Allison, P. D. (1995). Survival analysis using the SAS system: A practical guide. Cary, NC: SAS Institute.
Angrist, J. D., Imbens, G. W., & Rubin, D. B. (1996). Identification of causal effects using instrumental variables. Journal of the American Statistical Association, 91, 434, 444–455.
Barber, J. S., Murphy, S. A., & Verbitsky, N. (2004). Adjusting for time-varying confounding in survival analysis. Sociological Methodology, 34, 163–192.
Berkson, J. (1946). Limitations of the application of fourfold table analysis to hospital data. Biometric Bulletin, 2, 47–53.
Bohrnstedt, G. W., & Knoke, D. (1982). Statistics for social data analysis. Itasca, IL: P. E. Peacock Publishers.
Bollen, K. A. (1989). Structural equations with latent variables. New York: Wiley.
Bray, B. C., Zimmerman, R. S., Lynam, D., & Murphy, S. (2003). Assessing the total effect of time-varying predictors in prevention research (Technical Report 03-59). University Park, PA: The Methodology Center, Pennsylvania State University.
Clayton, R. R., Cattarello, A., Day, L. E., & Walden, K. P. (1991). Persuasive communications and drug prevention: An evaluation of the D.A.R.E. program. In L. Donohew, H. E. Sypher, & W. J. Bukoski (Eds.), Persuasive communication and drug abuse prevention (pp. 279–294). Hillsdale, NJ: Erlbaum.
Clayton, R. R., Cattarello, A. M., & Johnstone, B. M. (1996). The effectiveness of drug abuse resistance education (project DARE): 5-year follow-up results. Preventive Medicine, 25, 307–318.
Cochran, W., & Rubin, D. (1973). Controlling bias in observational studies. Sankya—The Indian Journal of Statistics, Series A, 35(Dec), 417–446.
Heckman, J. J. (1979). Sample selection bias as a specification error. Econometrica, 47(1), 153–161.
Heckman, J. J., & Hotz, V. J. (1989). Choosing among alternative nonexperimental methods for estimating the impact of social programs—The case of manpower training. Journal of the American Statistical Association, 84, 408, 862–874.
Hernán, M., Brumback, B., & Robins, J. M. (2000). Marginal structural models to estimate the causal effect of zidovudine on the survival of HIV-positive men. Epidemiology, 11(5), 561–570.
Hernán, M., Brumback, B., & Robins, J. M. (2001). Marginal structural models to estimate the joint causal effect of nonrandomized treatments. Journal of the American Statistical Association, 96, 454, 440–448.
Imbens, G. (2000). The role of the propensity score in estimating dose–response functions. Biometrika, 87(3), 706–710.
Joffe, M. M., Ten Have, T. R., Feldman, H. I., & Kimmel, S. E. (2004). Model selection, confounder control, and marginal structural models: Review and new applications. American Statistician, 58(4), 272–279.
Little, R. J., & Yau, L. H. Y. (1998). Statistical techniques for analyzing data from prevention trials: Treatment of no-shows using Rubin's causal model. Psychological Methods, 3(2), 147–159.
Oakes, J. M. (2004). The (mis)estimation of neighborhood effects: Causal inference for a practicable social epidemiology. Social Science and Medicine, 58(10), 1929–1952.
Pearl, J. (1998). Graphs, causality, and structural equation models. Sociological Methods and Research, 27, 226–284.
Raine, A. (1993). The psychopathology of crime: Criminal behavior as a clinical disorder. San Diego, CA: Academic Press.
Robins, J. M. (1998). Marginal structural models. 1997 proceedings of the American Statistical Association, section on Bayesian statistical science (pp. 1–10). Retrieved from http://www.biostat.harvard.edu/~robins/research.html
Robins, J. M. (1999). Association, causation, and marginal structural models. Synthese, 121(1/2), 151–179.
Robins, J. M., Hernán, M., & Brumback, B. (2000). Marginal structural models and causal inference in epidemiology. Epidemiology, 11(5), 550–560.
Rosenbaum, P. R., & Rubin, D. B. (1984). Reducing bias in observational studies using subclassification on the propensity score. Journal of the American Statistical Association, 79, 387, 516–524.
Rosenbaum, P. R., & Rubin, D. B. (1985). Constructing a control-group using multivariate matched sampling methods that incorportate the propensity score. American Statistician, 39(1), 33–38.
Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi-experimental designs for generalized causal inference. Boston: Houghton-Mifflin.
Singer, J. D., & Willet, J. B. (1993). It's about time: Using discrete-time survival analysis to study duration and the timing of events. Journal of Education Statistics, 18, 155–195.
Wechsler, D. (1981). The psychometric tradition: Developing the Wechsler Adult Intelligence Scale. Contemporary Educational Psychology, 6(2), 82–85.
White, H. (1982). Maximum likelihood estimation of misspecified models. Econometrica, 50(1), 1–25.
Winship, C., & Mare, R. D. (1992). Models for sample selection bias. Annual Review of Sociology, 18, 327–350.
Winship, C., & Morgan, S. L. (1999). The estimation of causal effects from observational data. Annual Review of Sociology, 25, 659–707.
Zuckerman, M. (1994). Behavioral expressions and biosocial bases of sensation seeking. Cambridge: Cambridge University Press.
ACKNOWLEDGMENTS
Preparation of this paper was supported by grant P50-DA-10075 from the National Institute on Drug Abuse to the Methodology Center at the Pennsylvania State University, by grant T32-DA-017629 to the Methodology Center and the Prevention Research Center for the Promotion of Human Development at the Pennsylvania State University, and by the National Institute on Drug Abuse award K02-DA-15674-01.
Author information
Authors and Affiliations
Corresponding author
APPENDIX: WEIGHT COMPUTATION
APPENDIX: WEIGHT COMPUTATION
Complete details and explanation of generic SAS programming code that can be used for the weight creation, naïve method, standard method, and weighted method are available at http://methodology.psu.edu/publications/tvpappen.html. Also provided are two simulated datasets that allow for practice and analysis in conjunction with a review of the SAS code. At each measured time point, \(t\), where a participant is at risk for response initiation (e.g., marijuana initiation) a weight component is created. Each weight component is the ratio of two predicted probabilities. The numerator is the predicted probability of a participant's observed predictor initiation or noninitiation in period \(t\), given past predictor initiation status (e.g., alcohol initiation or noninitiation) and baseline variables (e.g., sex and race), for those still at risk of response initiation. The denominator is the predicted probability of a participant's observed predictor initiation or noninitiation in period \(t\), given past predictor initiation status, baseline variables, and confounders, for those still at risk of response initiation. Thus, the numerator and denominator models only differ in that the confounders are present in the denominator predicted probability. Note that if alcohol initiation (i.e., predictor initiation) occurs prior to time \(t\), but before marijuana initiation (i.e., response initiation), the ratio at time \(t\) is 1 because both the numerator and denominator predicted probabilities are 1 (i.e., the predicted probability of initiating alcohol use after the initiation of alcohol use is 1 for any values of the confounders and baseline variables). Hence, the weight component does not need to be computed after predictor or response initiation (whichever occurs first). Because the numerator and denominator probabilities are computed only for those still at risk of predictor and response initiation, conditioning on \(\overline{{\rm Alc}}_{i-1}\) and \(\overline{{\rm Mj}}_{i-1}\) (complete past predictor and response patterns, respectively), as shown in Eq. (5), is not necessary. Note that conditioning on \(\overline{{\rm Mj}}_{i-1}\) in Eq. (5) is implicit because this is a survival analysis setting in which a participant is no longer in the dataset if the participant has initiated marijuana use; hence \(\overline{{\rm Mj}}_{i-1}=\mathbf{0}\) for all participants in the dataset at time \(i\). Thus, the equations below do not include past Alc or Mj. The model for the numerator regression model is
whereas the model for the denominator regression model is
Thus, the weight component for participant \(i\) before alcoholinitiation is
and the weight component for participant \(i\) at alcohol initiation is
The weight at time \(t\), \(W_t\), is the product of these weight components up to time \(t\). If at time \(t-1\) participant \(i\) has yet to initiate alcohol use then the weight at time \(t-1\) is
If at time \(t\) participant \(i\) initiates alcohol use then the form of the weight at time \(t\) is
The weight \(W_s\) for all times, \(s\) larger than \(t\), remains equal to \(W_t\) as each weight component is now equal to 1. Each participant has a weight for each time point until either the participant initiates the response or the study ends.
Rights and permissions
About this article
Cite this article
Bray, B.C., Almirall, D., Zimmerman, R.S. et al. Assessing the Total Effect of Time-Varying Predictors in Prevention Research. Prev Sci 7, 1–17 (2006). https://doi.org/10.1007/s11121-005-0023-0
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11121-005-0023-0