Association for Public Policy Analysis & Management

Objective: Evidence on patient-centered medical homes (PCMHs) reducing emergency department (ED) visits or reducing costs is mixed. Much of the variation in findings may be symptomatic of the wide array of definitions by accrediting entities. Furthermore, while accrediting entities rely on provider-reported practice attributes, the patient experience may differ greatly from the provider’s intentions. This study aims to compare the concordance between the provider and the patient perspectives of PCMH attributes, and then identify whether or not one perspective has a greater impact on subsequent ED utilization and annual healthcare expenditures.

Methods: Data come from the 2015 Medical Expenditure Panel Survey (MEPS) Medical Organization Survey (MOS) and Household Component. The recently released MOS is the first nationally representative survey of medical practices to be matched with patient perspective, enabling a unique opportunity to identify perspective concordance. Using Cronbach alpha correlations, we developed indices to measure provider-reported patient-centeredness based on National Committee for Quality Assurance PCMH principles. Logit regressions were then estimated to identify concordance between provider-reported patient-centeredness and the patient’s perception of accessibility, team-based care, and care coordination. In order to identify how strongly both perspectives are associated with healthcare outcomes, we used a logit and log-linear regression to estimate likelihood of having an ED visit and expenditures, respectively. All regressions were adjusted for patient demographics and health status. This study included 4,233 matched patients who reported having a usual source of care other than the ED. All patients reported having visited that provider at least once within the prior 12 months.

Results: There was little agreement between provider indices for patient-centered access or team-based care with the patient-perspective. However, patients that reported having extended hours available were more likely to be patients of providers in the top tertile of patient-centered accessibility (p< 0.01). Provider-reported care coordination activities were associated with patients reporting that their provider asks about other treatments they may be receiving (p=0.06). While neither patient nor provider perspective of care coordination or team-based care affected ED utilization, the patient perspective (and not the provider perspective) was strongly associated with a lower probability of visiting an ED (-0.09, p=0.02). Similarly, the patient perspective of access is associated with decreased expenditures (-0.06, p=0.02), while the provider perspective has no effect.

Conclusions: Attributes or services that provider practices report offering is not equivalent to the services or attributes experienced by patients. In fact, little overlap exists. The patient experience of access is most strongly associated with a lower probability of using an ED and lower costs, while the provider-reported services have no association with either outcome. While PCMH accreditation is generally based on provider-reported characteristics and services, this may not translate to the patient experience of care. Policies and provider practices aiming to reduce costs and ED utilization should focus on the patient’s perspective of access to care as opposed to services practices report offering.