Feasibility, acceptability, and adoption of an inpatient tobacco treatment service at a safety-net hospital: A mixed-methods study

Rationale: Hospitalization is an opportunity to engage smokers who may not seek tobacco treatment. Our safety-net hospital developed and implemented an inpatient intervention consisting of an "opt-out" electronic health record-based Best Practice Alert (BPA)1order-set, designed to trigger referral to the Tobacco Treatment Consult (TTC) service (a team staffed by tobacco treatment specialists) for all hospitalized smokers, regardless of motivation to quit. Objectives: We performed a sequential explanatory mixedmethods study to evaluate the feasibility, acceptability, and adoption of the TTC service. Methods: Among all admissions of adult "current smokers" between July 2016 and June 2017, we calculated the percentage of patients whose clinicians accepted the order-set (through a simple "order" click), thus generating the TTC referral. We then determined the extent of clinician follow-through of TTC recommendations for prescribing nicotine replacement therapy among 1,651 consecutive smokers seen by the TTC service. Finally, we conducted qualitative interviews with inpatient clinicians (n = 25) to understand their rationale for adoption or nonuse of the TTC intervention, including perceived usefulness, barriers to adoption, and strategies to improve the utility of the service. Results: Clinicians accepted the TTC order-set for 4,100 out of 6,598 "current smokers" (62.1%) for whom the BPA fired, typically after initially deferring the BPA. Rates of acceptance significantly differed across clinical services (range: 8% [obstetricsgynecology] to 82.2% [cardiology]; P , 0.00001). A chart review showed that 43.5% (719/1,651) of the patients seen by the TTC service desired outpatient nicotine replacement therapy, but only half of these patients (48.8%; 351/719) received a discharge prescription from the inpatient team. Clinicians expressed that they valued the TTC service, but that BPA fatigue, time constraints, competing priorities, and poor communication with the TTC service were barriers to using the service and following recommendations. Clinicians suggested strategies to address barriers to the use of tobacco treatment interventions during hospitalization and after discharge. Conclusions: Implementing a large-scale "opt-out" tobacco treatment service for hospitalized smokers at a safety-net hospital is feasible and acceptable, but suffers from inconsistent adoption due to a variety of clinician barriers. System-level changes are needed to increase uptake and sustain inpatient tobacco treatment interventions to promote smoking cessation.