Search

Objective Clinical pharmacists use population health methods to generate chronic disease management referrals for patients with uncontrolled chronic conditions. The purpose of this study was to compare primary care providers’ (PCPs) referral responses for 4 pharmacist-managed indications and to identify provider and patient characteristics that are predictive of PCP response. Design Retrospective cohort study. Setting This study occurred in an academic internal medicine clinic. Participants Clinical pharmacy referrals generated through a population health approach between 2012 and 2016 for hypertension, chronic pain, depression, and benzodiazepine management were included. Main outcome measures Proportion of referrals accepted, left pending, or rejected and influencing provider and patient characteristics. Results Of 1769 referrals generated, PCPs accepted 869 (49%), left pending 300 (17%), and rejected 600 (34%). Compared with referrals for hypertension, benzodiazepine management, and depression, chronic pain referrals had the lowest likelihood of rejection (odds ratio [OR] 0.31; 95% CI 0.19–0.49). Depression referrals had an equal likelihood of being accepted or rejected (OR 1.04; 95% CI 0.66–1.64). Provider characteristics were not significantly associated with referral response, but residents were more likely to accept referrals. Patient characteristics associated with lower referral rejection included black race (OR 0.39; 95% CI 0.18–0.87), higher systolic blood pressure (OR 0.98; 95% CI 0.97–0.99), and missed visits (OR 0.24; 95% CI 0.07–0.81). Conclusion The majority of referrals for clinical pharmacists in primary care settings were responded to, varying mostly between acceptance and rejection. There was variability in referral acceptance across indications, and some patient characteristics were associated with increased referral acceptance.