Comparing impact of a holistic patient centered navigation model to an online navigation service on health care utilization

<h3>Context</h3> Social prescribing has been linked to better access, improved health and wellbeing, and decreased health care utilization (HCU). However, very few randomized controlled trials have been conducted and have not assessed HCU. <h3>Objective</h3> We aim to compare the effects of two social prescribing (SP) lay navigation models on HCU in a randomized controlled trial. <h3>Population</h3> Primary care providers in 12 Ottawa and Sudbury (Ontario, Canada) practices referred their patients with health or social needs to access needed SP services. <h3>Study design/intervention</h3> These Patients were randomized to either receive the Access to Resources in the Community (ARC) SP navigation service; a holistic, patient-centered navigation services, or the provincially funded Ontario-211 online/telephone information and remote navigation SP services. <h3>Dataset</h3> Of the 326 enrolled patients, 150 consented and had their data successfully linked to health administrative data housed at ICES (ARC=83, Ontario-211: 67). <h3>Outcome Measures</h3> We compared the pre (Year -1/-2) - post (Year 0 and Year +1) differences in the number of outpatient and primary care visits in the ARC and 211 arms using linear regressions. We compared the odds ratio of patients having &gt;1 emergency department (ER) visit and &gt;1 hospitalization in the post-intervention years (Year 0 and Year +1) in the ARC and 211 arms using logistic regressions, while adjusting for pre-intervention HCU. Regression models were all adjusted with sociodemographic covariates. <h3>Results</h3> The adjusted 211-ARC difference (95% confidence interval (CI)) in outpatient visits was 1.0 (-1.6, 3.5) in Year 0 and 2.4 (-0.3, 5.1) in Year +1, and in primary visits was 0.7 (-0.6, 1.9) in Year 0 and 1.0 (-0.6, 2.6) Year +1; both in favour of the ARC. The odds ratio (95 CI) of ARC relative to 211 for ER visits was 0.7 (0.3, 1.5) in Year 0 and 1.7 (0.7, 4.1) in Year +1, and for hospitalizations was 1.2 (0.4, 3.3) in Year 0 and 1.8 (0.5, 6.8) in Year +1 in favour of ARC. <h3>Conclusions</h3> There was a trend for reduced HCU for patients in the ARC arm, although these results were not statistically significant. This study suggests that the ARC holistic patient navigation approach may be beneficial in reducing HCU and warrants further investigation with larger sample sizes.