Using Non-experimental Data to Estimate Treatment Effects.

Johns Hopkins Bloomberg School of Public Health, Baltimore.
Psychiatric Annals (Impact Factor: 0.71). 07/2009; 39(7):41451. DOI: 10.3928/00485713-20090625-07
Source: PubMed

ABSTRACT While much psychiatric research is based on randomized controlled trials (RCTs), where patients are randomly assigned to treatments, sometimes RCTs are not feasible. This paper describes propensity score approaches, which are increasingly used for estimating treatment effects in non-experimental settings. The primary goal of propensity score methods is to create sets of treated and comparison subjects who look as similar as possible, in essence replicating a randomized experiment, at least with respect to observed patient characteristics. A study to estimate the metabolic effects of antipsychotic medication in a sample of Florida Medicaid beneficiaries with schizophrenia illustrates methods.

  • [Show abstract] [Hide abstract]
    ABSTRACT: The present study used a monozygotic (MZ) cotwin-control (CTC) design to investigate associations between alcohol use and performance on the Iowa gambling task (IGT) in a sample of 96 adolescents (half female). The MZ CTC design is well suited to shed light on whether poor decision-making, as reflected on IGT performance, predisposes individuals to abuse substances or is a consequence of use. Participants completed structural MRI scans as well, from which we derived gray matter volumes for cortical and subcortical regions involved in IGT performance and reduced in adolescents with problematic alcohol use. Drinking was associated with poorer task performance and with reduced volume of the left lateral orbital-frontal cortex. CTC analyses indicated that the former was due to differences between members of twin pairs in alcohol use (suggesting a causal effect of alcohol), whereas the latter was due to factors shared by twins (consistent with a pre-existing vulnerability for use). Although these preliminary findings warrant replication, they suggest that normative levels of alcohol use may diminish the quality of adolescent decision-making and thus have potentially important public health implications.
    Behavior Genetics 03/2014; 44(4). DOI:10.1007/s10519-014-9651-0 · 2.84 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Objective: This study provides guidance on how propensity score methods can be combined with moderation analyses (i.e., effect modification) to examine subgroup differences in potential causal effects in nonexperimental studies. As a motivating example, we focus on how depression may affect subsequent substance use differently for men and women. Method: Using data from a longitudinal community cohort study (N = 952) of urban African Americans with assessments in childhood, adolescence, young adulthood, and midlife, we estimate the influence of depression by young adulthood on substance use outcomes in midlife, and whether that influence varies by gender. We illustrate and compare 5 different techniques for estimating subgroup effects using propensity score methods, including separate propensity score models and matching for men and women, a joint propensity score model for men and women with matching separately and together by gender, and a joint male/female propensity score model that includes theoretically important gender interactions with matching separately and together by gender. Results: Analyses showed that estimating separate models for men and women yielded the best balance and, therefore, is a preferred technique when subgroup analyses are of interest, at least in this data. Results also showed substance use consequences of depression but no significant gender differences. Conclusions: It is critical to prespecify subgroup effects before the estimation of propensity scores and to check balance within subgroups regardless of the type of propensity score model used. Results also suggest that depression may affect multiple substance use outcomes in midlife for both men and women relatively equally. (PsycINFO Database Record (c) 2014 APA, all rights reserved).
    Journal of Consulting and Clinical Psychology 04/2014; 82(5). DOI:10.1037/a0036515 · 4.85 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Community coalitions have been promoted as a strategy to help overcome challenges to the dissemination and implementation of evidence-based prevention programs. This paper explores the characteristics of coalitions that enable the provision of implementation support for prevention programs in general and for the implementation of evidence-based prevention programs with fidelity. Longitudinal cross-lagged panel models were used to study 74 Communities That Care (CTC) coalitions in Pennsylvania. These analyses provide evidence of a unidirectional influence of coalition functioning on the provision of implementation support. Coalition member knowledge of the CTC model best predicted the coalition's provision of support for evidence-based program implementation with fidelity. Implications for developing and testing innovative methods for delivering training and technical assistance to enhance coalition member knowledge are discussed.
    Prevention Science 12/2013; 16(1). DOI:10.1007/s11121-013-0447-x · 2.63 Impact Factor


1 Download
Available from