Background: As of 2017, 11.8% of US households reported being food insecure at some point during the year. Food insecurity has been linked to poor glycemic control and hypertension. Despite this, screening for food insecurity lags behind screening for other social determinants of health.1 Our team identified an opportunity as there was no standardized process in place for screening hospitalized adults for food insecurity at our institution.
Purpose: Our primary aim: develop and implement a sustainable interdisciplinary process to screen all hospitalized adults for food insecurity. In partnership with the Health Equity and Antiracism Technology Program, we designed and piloted a process to screen patients on one 26-bed medical unit.
Description: To accomplish this, our team worked with nursing staff, social workers, and medical providers on the unit to integrate food insecurity screening into the current admission workflow. Upon admission to the unit, nurses utilized a validated two-question screening tool from the U.S. Household Food Security Scale which was incorporated into the admission questionnaire in the electronic medical record. If a patient screened positive to either one or both questions, nursing would submit a consult to the unit social worker. The social worker would then provide the patient with local food resources to utilize at their discretion. Baseline data from the unit during a one-month period showed only one patient (of an average of 114 patient encounters per month) was identified as food insecure. Two weeks prior to pilot initiation, daily education was provided to nursing staff during unit safety huddles. Appointed nurse champions educated nursing staff on the screening workflow and provided interactive simulations of potential patient interactions.The pilot began on June 7, 2021, and after 22 weeks, 312 patients had been screened. This amounted to 61.5% of all patient encounters on the unit with 15 patients, or 5.0% of screened patients, identified as food insecure, a substantial improvement from baseline. PDSA cycles guided workflow adjustments based on nursing feedback and monitoring of adherence to the screening process. Limitations identified in the process included inability to screen patients who were non-verbal, encephalopathic, or who had other medical conditions impacting cognition at the time of unit admission. These limitations may have resulted in a lower percentage of identified food insecure patients compared to local rate of food insecurity of 12.5%.
Conclusions: As a hospitalist, the inpatient setting provides an opportunity to increase identification of food insecure patients and connect these patients to food resources. Our pilot study has led to a sustainable and substantial increase in food insecurity screening and as a result an increase in identification of food insecure patients. Future goals are to expand this process to other medical units at our hospital and to follow up with patients who were connected with resources to evaluate whether the resources had been utilized.