Use of Propensity Score Matching with Claims Data to Assess the Resource Utilization and Costs Attributable to HZ
*Paul Buzinec, i3 Innovus
Carolyn Harley, i3 Innovus
Robbin F. Itzler, Merck Research Laboratories
Patricia Saddier, Merck Research Laboratories
Keywords: cost analysis, resource use, healthcare claims, herpes zoster
This study assesses the resource use (RU) and costs of herpes zoster (HZ) in the first 3 months and the first year after diagnosis using claims data from a US health plan offering commercial and Medicare Advantage health benefits. The analyses of average RU and cost were based on propensity score matching in combination with regression based approaches to estimate the incremental differences in RU and costs for the cases and a comparable set of controls. The limitations of this analysis include the inability to directly attribute care to HZ, and limited clinical evidence of disease which is especially problematic when identifying complications related to HZ. The strengths of this analysis include the use of propensity score matching to select controls, identification of healthcare potentially related to HZ, and the inclusion of enrollees with different types of health plan coverage.