Propensity score stratification methods for continuous treatments

Continuous treatments propensity scoring remains understudied as the majority of methods are focused on the binary treatment setting. Current propensity score methods for continuous treatments typically rely on weighting in order to produce causal estimates. It has been shown that in some continuous treatment settings, weighting methods can result in worse covariate balance than had no adjustments been made to the data. Furthermore, weighting is not always stable, and resultant estimates may be unreliable due to extreme weights. These issues motivate the current development of novel propensity score stratification techniques to be used with continuous treatments. Specifically, the generalized propensity score cumulative distribution function (GPS-CDF) and the nonparametric GPS-CDF approaches are introduced. Empirical CDFs are used to stratify subjects based on pretreatment confounders in order to produce causal estimates. A detailed simulation study shows superiority of these new stratification methods based on the empirical CDF, when compared with standard weighting techniques. The proposed methods are applied to the "Mexican-American Tobacco use in Children" study to determine the causal relationship between continuous exposure to smoking imagery in movies, and smoking behavior among Mexican-American adolescents. These promising results provide investigators with new options for implementing continuous treatment propensity scoring.

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

[2]  P. Austin.  Assessing the performance of the generalized propensity score for estimating the effect of quantitative or continuous exposures on binary outcomes , 2018, Statistics in medicine.

[3]  Chad Hazlett,et al.  Covariate balancing propensity score for a continuous treatment: Application to the efficacy of political advertisements , 2018 .

[4]  M. Spitz,et al.  Correlates of susceptibility to smoking among Mexican origin youth residing in Houston, Texas: A cross-sectional analysis , 2008, BMC public health.

[5]  Elaine L. Zanutto,et al.  Using Propensity Score Subclassification for Multiple Treatment Doses to Evaluate a National Antidrug Media Campaign , 2005 .

[6]  B. McNeil,et al.  "Renalism": inappropriately low rates of coronary angiography in elderly individuals with renal insufficiency. , 2004, Journal of the American Society of Nephrology : JASN.

[7]  Elizabeth A Stuart,et al.  Improving propensity score weighting using machine learning , 2010, Statistics in medicine.

[8]  M. Spitz,et al.  Cognitive Susceptibility to Smoking: Two Paths to Experimenting among Mexican Origin Youth , 2009, Cancer Epidemiology, Biomarkers & Prevention.

[9]  Peter C Austin,et al.  A comparison of the ability of different propensity score models to balance measured variables between treated and untreated subjects: a Monte Carlo study , 2007, Statistics in medicine.

[10]  Francesca Dominici,et al.  Application of a propensity score approach for risk adjustment in profiling multiple physician groups on asthma care. , 2005, Health services research.

[11]  Til Stürmer,et al.  A review of the application of propensity score methods yielded increasing use, advantages in specific settings, but not substantially different estimates compared with conventional multivariable methods. , 2006, Journal of clinical epidemiology.

[12]  Kosuke Imai,et al.  Causal Inference With General Treatment Regimes , 2004 .

[13]  Joseph Kang,et al.  Demystifying Double Robustness: A Comparison of Alternative Strategies for Estimating a Population Mean from Incomplete Data , 2007, 0804.2958.

[14]  G. Imbens.  The Role of the Propensity Score in Estimating Dose-Response Functions , 1999 .

[15]  James D Sargent,et al.  Population-Based Assessment of Exposure to Risk Behaviors in Motion Pictures , 2008, Communication methods and measures.

[16]  Carlos A. Flores,et al.  A Stata Package for the Application of Semiparametric Estimators of Dose–Response Functions , 2014 .

[17]  Lane F Burgette,et al.  A tutorial on propensity score estimation for multiple treatments using generalized boosted models , 2013, Statistics in medicine.

[18]  P. Austin.  An Introduction to Propensity Score Methods for Reducing the Effects of Confounding in Observational Studies , 2011, Multivariate behavioral research.

[19]  Debashis Ghosh,et al.  A Boosting Algorithm for Estimating Generalized Propensity Scores with Continuous Treatments , 2015, Journal of causal inference.

[20]  Michael J. Lopez,et al.  Estimation of causal effects with multiple treatments: a review and new ideas , 2017, 1701.05132.

[21]  J. Robins,et al.  Marginal Structural Models and Causal Inference in Epidemiology , 2000, Epidemiology.

[22]  Brian K. Lee,et al.  Weight Trimming and Propensity Score Weighting , 2011, PloS one.

[23]  Dylan S. Small,et al.  The use of bootstrapping when using propensity-score matching without replacement: a simulation study , 2014, Statistics in medicine.

[24]  W. G. Cochran.  The effectiveness of adjustment by subclassification in removing bias in observational studies. , 1968, Biometrics.

[25]  D. Rubin,et al.  Reducing Bias in Observational Studies Using Subclassification on the Propensity Score , 1984 .

[26]  Andrew D. Martin,et al.  Untangling the Causal Effects of Sex on Judging , 2010 .

[27]  E. Stuart,et al.  Moving towards best practice when using inverse probability of treatment weighting (IPTW) using the propensity score to estimate causal treatment effects in observational studies , 2015, Statistics in medicine.

[28]  D. Brotman,et al.  Association of impaired diurnal blood pressure variation with a subsequent decline in glomerular filtration rate. , 2006, Archives of internal medicine.

[29]  Peter C Austin,et al.  Assessing covariate balance when using the generalized propensity score with quantitative or continuous exposures , 2018, Statistical methods in medical research.

[30]  Chunhao Tu,et al.  Comparison of clustering algorithms on generalized propensity score in observational studies: a simulation study , 2013 .

[31]  M. Spitz,et al.  Exposure to Smoking Imagery in the Movies and Experimenting with Cigarettes among Mexican Heritage Youth , 2009, Cancer Epidemiology, Biomarkers & Prevention.

[32]  Stephen R Cole,et al.  Constructing inverse probability weights for marginal structural models. , 2008, American journal of epidemiology.

[33]  Hulin Wu,et al.  A novel approach for propensity score matching and stratification for multiple treatments: Application to an electronic health record–derived study , 2020, Statistics in medicine.

[34]  D. Rubin,et al.  Constructing a Control Group Using Multivariate Matched Sampling Methods That Incorporate the Propensity Score , 1985 .

[35]  K. Imai,et al.  Covariate balancing propensity score , 2014 .

[36]  Martin Schumacher,et al.  Estimators and confidence intervals for the marginal odds ratio using logistic regression and propensity score stratification , 2010, Statistics in medicine.

[37]  Richard Grieve,et al.  Evaluation of the Effect of a Continuous Treatment: A Machine Learning Approach with an Application to Treatment for Traumatic Brain Injury , 2015, Health economics.

[38]  D. Marquardt.  An Algorithm for Least-Squares Estimation of Nonlinear Parameters , 1963 .

[39]  P. Rosenbaum,et al.  Invited commentary: propensity scores. , 1999, American journal of epidemiology.

[40]  Douglas G Altman,et al.  Dichotomizing continuous predictors in multiple regression: a bad idea , 2006, Statistics in medicine.

[41]  J. Schafer,et al.  Average causal effects from nonrandomized studies: a practical guide and simulated example. , 2008, Psychological methods.

[42]  J. Avorn,et al.  Variable selection for propensity score models. , 2006, American journal of epidemiology.

[43]  Megan S. Schuler,et al.  Propensity score weighting for a continuous exposure with multilevel data , 2016, Health Services and Outcomes Research Methodology.