Improving the causal treatment effect estimation with propensity scores by the bootstrap

When observational studies are used to establish the causal effects of treatments, the estimated effect is affected by treatment selection bias. The inverse propensity score weight (IPSW) is often used to deal with such bias. However, IPSW requires strong assumptions whose misspecifications and strategies to correct the misspecifications were rarely studied. We present a bootstrap bias correction of IPSW (BC-IPSW) to improve the performance of propensity score in dealing with treatment selection bias in the presence of failure to the ignorability and overlap assumptions. The approach was motivated by a real observational study to explore the potential of anticoagulant treatment for reducing mortality in patients with end-stage renal disease. The benefit of the treatment to enhance survival was demonstrated; the suggested BC-IPSW method indicated a statistically significant reduction in mortality for patients receiving the treatment. Using extensive simulations, we show that BC-IPSW substantially reduced the bias due to the misspecification of the ignorability and overlap assumptions. Further, we showed that IPSW is still useful to account for the lack of treatment randomization, but its advantages are stringently linked to the satisfaction of ignorability, indicating that the existence of relevant though unmeasured or unused covariates can worsen the selection bias.

Health Service Utilization and State Costs Among Adults Aging With Early-Acquired Physical Disabilities in Medicaid Managed Care

Objective: We evaluated the impact of Medicaid managed care (MMC) on health service use and state costs among adults with early-acquired physical disabilities. Method: Using claims data, we tracked utilization of the emergency department (ED), inpatient admissions, outpatient physician visits, and state expenditures on enrollees who transitioned to MMC ( n = 881). The inverse propensity score weight and a difference-in-differences regression model were used to estimate the impact of MMC using their counterparts who remained in fee-for-service ( n = 1,552) as the comparison group. Results: MMC reduced ED use by 3.2% points/month ( p < .001). Relative to younger enrollees (age ⩽45 years), MMC reduced inpatient admissions of older enrollees (age ⩾46 years) by 3.3% points/month ( p < .001), and state expenditures by US$839/month ( p < .01). Discussion: MMC could reduce the hospital service use of and state spending on enrollees with early-acquired physical disabilities. This impact may vary depending on the enrollees’ age.