Causal Inference Methods in Action

Featured Research

  • State-Level Opioid Policy Analyses: Moving Beyond the Classic Difference-in-Differences Model

    These videos provide an overview of key methodological challenges faced when evaluating the effectiveness of state-level opioid policies using annual state-level data, as well as potential solutions and practice guidelines.

  • Unintended Consequences of Policies that Punish Pregnant Women for Substance Use

    A retrospective, difference-in-difference analysis of live births in the State Inpatient Databases from 8 U.S. states in varying years between 2003 and 2014 found that states that impose punitive action against pregnant women who use illicit substances are associated with higher rates of infants being born with opioid withdrawal.

  • Gaussian Process Framework Models Treatment Probability

    Propensity scores are commonly employed in observational study settings where the goal is to estimate average treatment effects. The paper introduces a flexible propensity score modelling approach, where the probability of treatment is modelled through a Gaussian process framework.

  • Allowing Pharmacists to Directly Dispense Opioid Antidote Can Sharply Cut Opioid Overdose Deaths

    States that adopted naloxone access laws giving direct prescription authority to pharmacists had significant decreases in opioid-related deaths, according to a difference-in-differences approach that estimates the magnitude of the association for each year relative to time of the laws' adoption. Other types of naloxone access laws did not seem to affect opioid overdose rates.

  • How the Opioid Crisis Is Driving the Rise in Hepatitis C

    Using difference-in-differences methods on data for the period 2004–15 allowed researchers to assess whether states with higher rates of OxyContin misuse prior to reformulation—states where the reformulation had more impact—experienced faster growth in Hepatitis C infections after the reformulation.

  • Opening School-Based Health Clinics Can Lower Teacher Health Costs

    Opening onsite health clinics to provide comprehensive primary care to teachers and their families can lower a school district's health care costs and decrease teacher absenteeism, according to a difference-in-differences analysis.

  • One Size Fits All? Disentangling the Effects of Tobacco Policies

    Among three state-level tobacco policies (cigarette taxation, tobacco control spending, and smoke-free air laws), a difference-in-differences analysis with generalized propensity scores found that only taxation significantly reduced smoking among the general adult population.

  • The Not-So-Marginal Value of Weather Warning Systems

    Estimates of the benefits of weather warning systems are sparse, perhaps because there is often no clear counterfactual of how individuals would have fared without a particular warning system. Researchers used conditional variation in the initial broadcast dates of the National Oceanic and Atmospheric Administration's Weather Radio All Hazards (NWR) transmitters to produce both cross-sectional and fixed effects estimates of the causal impact of expanding the NWR transmitter network.

  • No Link Found Between Pets and Kids' Health

    Contrary to popular belief, having a dog or cat in the home does not improve the mental or physical health of children. Researchers used a weighted propensity score regression approach and double robust regression analyses to examine the association between living with a dog or cat and health outcomes, while accounting for confounding factors.

  • Does Legalization Affect Teen Marijuana Use?

    Difference-in-difference estimates comparing Washington with states that did not legalize recreational drug use indicated that 8th and 10th graders reported a significant increase in marijuana use and decrease in perceptions of harmfulness.

  • Testing a Video Game Intervention to Recalibrate Physician Heuristics in Trauma Triage

    Researchers created a video game to recalibrate how trauma triage physicians determine whether a patient's injuries appear typical. They then conducted a randomized controlled trial to compare the effect of this game with that of another educational program on physicians' triage decisions.

  • A Cluster-Randomized Trial of Restorative Practices

    The Study of Restorative Practices was a 5-year, cluster-randomized controlled trial (RCT) of the Restorative Practices Intervention in 14 middle schools in Maine to assess whether the intervention affects both positive developmental outcomes and problem behaviors; it was the first RCT of its kind.

  • Evaluating the Impact of Parent-Reported Medical Home Status on Children's Health Care

    To study how changes in medical home status over a 2-year period affected children's health care outcomes, researchers took a causal difference-in-differences approach using inverse probability weighting and doubly robust estimators. They found that having a medical home may help improve health care quality for children, and losing a medical home may lead to higher utilization of emergency care.

  • A Synthetic Control Approach to Evaluating Place-Based Crime Interventions

    To evaluate a drug market intervention, researchers used a synthetic control model to reduce the bias introduced by models that use non-equivalent comparison groups. The research also demonstrated the method and its versatility for evaluating programs retrospectively.

  • TWANG Short Course/Educational Videos: Three Videos — Introduction, Propensity Score Weighted Analyses with 2 Groups, and Propensity Score Weighted Analyses with More Than 2 Groups

    This series of three training videos provides researchers and analysts with an overview of causal inference and the role of propensity scores in analysis. The videos provide step-by-step procedures for implementing propensity score analyses involving two or more treatment groups using the TWANG (Toolkit for Weighting and Analysis of Nonequivalent Groups) data analysis package.

  • The Effect of School District Nutrition Policies on Dietary Intake and Overweight: A Synthetic Control Approach

    Results of a study implementing cohort and cross-section estimators using "synthetic" control groups—combinations of unaffected districts that are reweighted to closely resemble the treatment unit in the pre-intervention period—indicate that the policy was mostly ineffective at reducing the prevalence of overweight or obesity

  • Community Justice Center in San Francisco Is Associated with Lower Rearrest Rates

    Using a differences-in-differences design, researchers examined one-year rearrest rates among those arrested for eligible offenses within the four police districts that include a part of the CJC catchment area. Their findings support the hypothesis that the CJC reduces criminal recidivism and are robust to a number of sensitivity analyses.

  • Causal Inference Using Mixture Models: A Word of Caution

    Mixture models are useful for monitoring the behavior of data and for offering comparisons to supplemental data, especially in the presence of unobserved heterogeneity, but one should be highly cautious when drawing causal inferences as to which population each component of the fitted mixture model represents.

  • Toolkit for Weighting and Analysis of Nonequivalent Groups (TWANG)

    The Toolkit for Weighting and Analysis of Nonequivalent Groups, or TWANG, contains a set of functions to support causal modeling of observational data through the estimation and evaluation of propensity score weights. TWANG has been released as a package in R, and as downloadable macros for SAS users. The team has also released a series of tutorials intended to guide analysts using the toolkit for the first time.

  • Tool Helps Defense Planners Match Priorities for Security Cooperation

    For each of the world's 195 countries, a diagnostic tool developed by RAND for the U.S. military produces an overall security cooperation propensity score. Planners can then compare these scores with available funding and security cooperation priorities.