Propensity score weighting for a continuous exposure with multilevel data

被引：39

作者：

Schuler M.S. ^{[1
]}

Chu W. ^{[2
]}

Coffman D. ^{[3
]}

机构：

[1] Department of Health Care Policy, Harvard Medical School, 180 Longwood Avenue, Boston, 02215, MA

[2] Google, Inc., Mountain View, 94043, CA

[3] Department of Epidemiology and Biostatistics, Temple University, Philadelphia, 19122, PA

来源：

Health Services and Outcomes Research Methodology | 2016年 / 16卷 / 4期

基金：

美国国家科学基金会; 美国国家卫生研究院;

关键词：

Continuous exposure; Multilevel data; Observational study; Propensity score;

D O I：

10.1007/s10742-016-0157-5

中图分类号：

学科分类号：

摘要：

Propensity score methods (e.g., matching, weighting, subclassification) provide a statistical approach for balancing dissimilar exposure groups on baseline covariates. These methods were developed in the context of data with no hierarchical structure or clustering. Yet in many applications the data have a clustered structure that is of substantive importance, such as when individuals are nested within healthcare providers or within schools. Recent work has extended propensity score methods to a multilevel setting, primarily focusing on binary exposures. In this paper, we focus on propensity score weighting for a continuous, rather than binary, exposure in a multilevel setting. Using simulations, we compare several specifications of the propensity score: a random effects model, a fixed effects model, and a single-level model. Additionally, our simulations compare the performance of marginal versus cluster-mean stabilized propensity score weights. In our results, regression specifications that accounted for the multilevel structure reduced bias, particularly when cluster-level confounders were omitted. Furthermore, cluster mean weights outperformed marginal weights. © 2016, Springer Science+Business Media New York.

引用

页码：271 / 292

页数：21

共 40 条

[31]

Rubin D.B., Statistics and causal inference: comment: Which ifs have causal answers, J. Am. Stat. Assoc., 81, 396, pp. 961-962, (1986)

[32]

Scharfstein D.O., Rotnitzky A., Robins J.M., Adjusting for non-ignorable drop-out using semiparametric non-response models, J. Am. Stat. Assoc., 94, 448, pp. 1096-1120, (1999)

[33]

Setoguchi S., Schneeweiss S., Brookhart M.A., Glynn R.J., Cook E.F., Evaluating uses of data mining techniques in propensity score estimation: A simulation study, Pharmacoepidemiol Drug Saf, 17, 6, pp. 546-555, (2008)

[34]

Stuart E.A., Estimating causal effects using school-level data sets, Educ. Res., 36, 4, pp. 187-198, (2007)

[35]

Stuart E.A., Matching methods for causal inference: a review and a look forward, Stat. Sci., 25, 1, pp. 1-21, (2010)

[36]

Su Y.S., Cortina J., What do we gain? Combining propensity score methods and multilevel modeling. Paper presented at the Annual Meeting of the American Political Science Association, (2009)

[37]

Thoemmes F.J., West S.G., The use of propensity scores for nonrandomized designs with clustered data, Multivar. Behav. Res., 46, 3, pp. 514-543, (2011)

[38]

Xiang Y., Tarasawa B., Propensity score stratification using multilevel models to examine charter school achievement effects, J. Sch. Choice, 9, 2, pp. 179-196, (2015)

[39]

Zhang Z., Zhou J., Cao W., Zhang J., Causal inference with a quantitative exposure, Stat. Methods Med. Res, 25, 1, pp. 315-335, (2016)

[40]

Zhu Y., Coffman D.L., Ghosh D., A boosting algorithm for estimating generalized propensity scores with continuous treatments, J. Causal inference, 3, 1, pp. 25-40, (2014)

← 1 2 3 4 →