Clinical and Utilization Outcomes for a Heart Failure Care Support Program: A Matched-Cohort Study

Objective: To investigate the effect of a heart failure disease management program for patients >=18 years of age enrolled in a commercial health plan. Background: Disease management provides a framework for managing the chronic illness of large populations. Evaluating the comparative benefits of disease management program participation remains a central challenge for researchers, clinicians, and administrators. A growing consensus in the field of disease management is that more rigorous methodologies are required to assess program outcomes. However, many heart failure disease management programs have been evaluated by the use of non-experimental designs (pre-/post-methodologies), or matching and stratification methods that have been used with limited success. Methods: To investigate the program effects of a heart failure care support program, we conducted a matched-cohort study on 521 participants using propensity scores. This methodology constructed matched samples of treated and control pairs for a wide range of observed characteristics and may reduce the bias in estimates of treatment effects to provide a relatively more accurate assessment of program outcomes. Administrative claims provided the source data for evaluating rates of hospitalizations, emergency department visits, and physician office visits. The study also included selected clinical indicators from administrative claims data to estimate the effects of program enrollment. Results: Participants exhibited significantly fewer cardiac-related inpatient admissions and bed days compared with those for matched cohorts. A greater proportion of participants received cardiography testing and pneumococcal and influenza immunizations compared with matched cohorts. Participants experienced less use of medical services overall, suggesting that there were beneficial effects with monitoring and education for this group. Conclusions: This study documents the beneficial outcomes of participation in a commercially delivered heart failure care support program. In those cases where controlled randomized clinical trials cannot be performed because of ethical, cost, or feasibility issues, the use of propensity scores provides an alternative for estimating treatment effects based on observational data.