Abstract
Background/Aims Early evidence suggests that a patient-centered medical home (PCMH) has the potential to improve quality of care while reducing cost. However, it is unclear how PCMH achieves such desirable outcomes, particularly in regards to its impact on cost of care. In this study, we examined claims data from Geisinger Health System’s version of PCMH, known as ProvenHealth Navigator® (PHN) to answer the following questions: 1) From the perspective of a primary care clinic, what is the overall total cost savings associated with exposure to PHN; and 2) where do the cost savings come from? In answering the latter question, we deconstruct the total cost savings into their major component parts and examined the association separately between the clinic’s exposure to PHN and each cost component.
Methods Data were obtained from Geisinger Health Plan’s claims database among its Medicare Advantage members who received care from the 65 primary care sites that had been transformed into PHN sites between 2006 through 2012. The main outcome variable was the mean per-member-per-month (PMPM) total cost of care for a given PHN site. The total PMPM cost was further divided into five components (outpatient, inpatient, professional, prescription generic and brand drugs). The key explanatory variable was the length of the clinic’s exposure to PHN in months. Other covariates included the number of patients in each clinic in each month, the site’s ownership status, % male patients, % patients with drug coverage, average age and average HCC risk scores, as well as month and year indicator variables. In total, seven linear regression models with clinic fixed effects were estimated.
Results For every 6-month incremental exposure to PHN, there was approximately $10 (P <0.05) reduction in mean PMPMP cost savings for a given primary care clinic. About 50% of this saving was attributable to reductions in inpatient cost, followed by reductions in professional fees and outpatient cost. The total cost saving was offset by an increases in prescription drug cost associated with PHN exposure.
Conclusions Our finding is consistent with the hypothesis that PCMH reduces avoidable acute care while potentially increasing patient adherence to prescription drugs.
