Predicting Primary Care Use Among Patients in a Large Integrated Health System: The Role of Patient Experience Measures.

Wong ES, Maciejewski ML, Hebert PL, Reddy A, Liu CF. Predicting Primary Care Use Among Patients in a Large Integrated Health System: The Role of Patient Experience Measures. Medical care. 2019 Aug 1; 57(8):608-614.

Related HSR&D Project(s)

• CDA 13-024 – Economic Impact of Dual Use and Patient Choice in Primary Care
• RCS 10-391 – Research Career Scientist

Dimensions for VA is a web-based tool available to VA staff that enables detailed searches of published research and research projects.

If you have VA-Intranet access, click here for more information vaww.hsrd.research.va.gov/dimensions/

VA staff not currently on the VA network can access Dimensions by registering for an account using their VA email address.    Search Dimensions for VA for this citation
* Don't have VA-internal network access or a VA email address? Try searching the free-to-the-public version of Dimensions

Search for Abstract from PubMed

---

Abstract:

OBJECTIVE: Most Veterans Affairs (VA) Health Care System enrollees age 65+ also have the option of obtaining care through Medicare. Reliance upon VA varies widely and there is a need to optimize its prediction in an era of expanding choice for veterans to obtain care within or outside of VA. We examined whether survey-based patient-reported experiences improved prediction of VA reliance. METHODS: VA and Medicare claims in 2013 were linked to construct VA reliance (proportion of all face-to-face primary care visits), which was dichotomized ( = 1 if reliance > 50%). We predicted reliance in 83,143 Medicare-eligible veterans as a function of 61 baseline characteristics in 2012 from claims and the 2012 Survey of Healthcare Experiences of Patients. We estimated predictive performance using the cross-validated area under the receiver operating characteristic (AUROC) curve, and assessed variable importance using the Shapley value decomposition. RESULTS: In 2012, 68.9% were mostly VA reliant. The AUROC for the model including claims-based predictors was 0.882. Adding patient experience variables increased AUROC to 0.890. The pseudo R for the full model was 0.400. Baseline reliance and patient experiences accounted for 72.0% and 11.1% of the explained variation in reliance. Patient experiences related to the accessibility of outpatient services were among the most influential predictors of reliance. CONCLUSION: The addition of patient experience variables slightly increased predictive performance. Understanding the relative importance of patient experience factors is critical for informing what VA reform efforts should be prioritized following the passage of the 2018 MISSION Act.