Implementing regional care coordination through patient trust, health literacy, and experience: a path analysis of reverse referral preference

Abstract

Abstract Background “Reverse referral” refers to transferring outpatient care from a large hospital to a community clinic, small hospital, or primary care physician when specialized hospital care is no longer needed. It is key to functional differentiation among medical institutions in Japan and requires strategies tailored to individual patients’ characteristics. This study examined whether trust in physicians, health literacy (HL), and patient experience (PX) are associated with patients’ preferences for reverse referrals, using the previously developed 19-item Patient Preference Questionnaire for Reverse Referral (PQR-19). Methods An online cross-sectional survey was administered to adults aged ≥ 20 years who had been hospitalized at a large hospital (≥400 beds) within the past 6 months and were currently receiving outpatient care. Patient reverse referral preference, trust in physicians, HL, and PX were measured using the PQR-19, the general physician version of the Abbreviated Wake Forest Physician Trust Scale (A-WFPTS), the 14-item Health Literacy Scale (HLS-14), and the Consumer Assessment of Healthcare Providers and Systems Clinician and Group Survey (CG-CAHPS) global rating, respectively; higher scores indicated more favorable reverse referral preference, greater trust in physicians, higher HL, and better PX. Multiple regression analyses were performed, adjusting for demographic variables, to examine relationships between these constructs. Additionally, a hypothesized path model was tested to assess model fit. Results A total of 293 participants completed the survey. Multiple regression analyses indicated that A-WFPTS and HLS-14 were independent predictors of PQR-19, with 10-point increases corresponding to 0.06- and 0.26-point increases, respectively. CG-CAHPS score was associated with both A-WFPTS and HLS-14, with 10-point increases corresponding to 1.66- and 1.26-point increases, respectively. CG-CAHPS was not independently associated with PQR-19 when A-WFPTS and HLS-14 were included in the model, but was positively associated with PQR-19 in a model that omitted these hypothesized mediators. Path analysis demonstrated a good model fit, supporting relationships between reverse referral preference, trust in physicians, HL, and PX. Conclusions Patients’ reverse referral preferences were associated with trust in physicians and HL, both of which were linked to PX. Enhancing patients’ trust in physicians, HL, and PX may be an effective means to promote reverse referrals.