3184 — Which facilitators and barriers have the strongest correlations with implementation of a patient-centered medical home in VA?

Helfrich CD, Seattle-Denver COIN; Gale RC, Palo Alto COIN; Sylling PW, Analytics and Business Intelligence; Mohr D, Bedford-Boston COIN; Stockdale SE, Los Angeles COIN; Brown EJ, University of Pennsylvania; Asch SM, Palo Alto COIN; Nelson KM, Seattle-Denver COIN; Fihn SD, Analytics and Business Intelligence; Meredith LS, RAND

Objectives:
The VA patient-centered medical home (PCMH) model (Patient Aligned Care Teams, PACT) may be associated with significant implementation challenges, and the conditions and resources that foster or hinder PACT implementation are not clear. To better understand how to disseminate PCMH, we assessed the associations of facilitators and barriers with PACT implementation.

Methods:
We performed a cross-sectional analysis of the odds of employees from high (n = 77) and medium-implementation (n = 749) PACT clinics relative to low-implementation (n = 87) clinics reporting specific barriers and facilitators. We assessed PACT implementation with the 2012 PACT Implementation Progress Index, a previously validated, clinic-level index covering 8 PCMH domains. Data on availability and helpfulness of 10 facilitators of PACT implementation and 19 barriers to delivery of patient-centered care were from a 2012 national primary-care employee survey (n = 4,819). We adjusted for respondent and clinic characteristics.

Results:
The most prevalent barrier was volume of electronic clinical reminders (54% reported "limits a great deal") and the highest rated facilitator was teamlet huddles (51% reported "very helpful"). Respondents at high-implementation clinics, relative to low-implementation clinics, had higher odds of reporting 5 facilitators, especially involvement in teamlet huddles (OR = 3.75, p = .01), while respondents at medium-implementation facilities had higher odds of reporting 3 facilitators especially the availability of disease registries (OR = 1.59, p = .04). Respondents at high-implementation clinics had higher odds of rating 4 facilitators as very helpful, especially local education sessions (OR = 1.99, p = .02). Relative to low-implementation clinics, high-implementation clinics were better on 9 barriers, especially recruiting and retaining providers (OR = .41, p < .01), and medium-implementation clinics were better on 7 barriers, especially recruiting and retaining non-provider clinicians (OR = .59, p = .01).

Implications:
Respondents at clinics with the highest level of PACT implementation were more likely to report teamlet huddles and rated local PACT training sessions as more helpful, while less likely to report recruitment and retention barriers. This suggests some clinics may have found local solutions to common problems such as obtaining PACT training, and hiring providers.

Impacts:
These findings contribute to the PCMH literature, by testing the associations of specific barriers and facilitators with a nuanced, validated measure of medical home implementation in a large population of clinics.