Contextualizing Findings from an Updated Veterans Health Administration Food Insecurity Clinical Screener

Context: The Veterans Health Administration (VHA) implemented universal screening for food insecurity in 2017. In April 2021, VHA switched from a single non-validated question to a validated two-item screener.

Objective: To determine if this change impacted identified prevalence or correlates of food insecurity among Veterans screened.

Study Design and Analysis: Repeated cross-sectional study; multivariable logistic regression.

Dataset: VHA electronic health record.

Population Studied: All Veterans screened for food insecurity between March 2020-April 2022.

Outcome Measures: 1) prevalence of food insecurity between March 2020-March 2021 (old screener) and April 2021-April 2022 (new screener) across VA facilities nationally; 2) sociodemographic, medical, and psychosocial characteristics associated with a positive screen during each time period.

Results: In the final 14 months of the old screener, 2,409,493 Veterans were screened for food insecurity, of whom 25,099 screened positive (1.0%); 2,827,319 Veterans were screened in the first 14 months of the new screener, of whom 56,967 screened positive (2.0%). Characteristics of those screened were comparable across study periods (SMDs all <0.05). For both screeners, food insecurity was significantly associated with identifying as American Indian/Alaskan Native, Non-Hispanic Black, Hispanic, and Native Hawaiian/Pacific Islander; non-married/partnered; low-income; housing instability, and mental health and trauma-related comorbidities including depression, PTSD, history of military sexual trauma, and intimate partner violence (IPV). The new screener, however, was more likely to identify as food insecure women (2.9% with the new screener vs 1.3% with the old screener) and those with a recent history of IPV (6.3% vs 3.0%).

Conclusions: The rate of reported food insecurity with the new screener doubled compared with the old screener. Although the absolute percentage of reported food insecurity remained low, ~26,000 additional Veterans screened positive in the first year of the new screener who may not have otherwise been identified. Findings suggest the new screener may be more likely to identify food insecurity among certain Veteran populations including women and those experiencing IPV. Accurately identifying food insecurity is a critical to connecting Veterans with needed services, developing tailored interventions for those at highest risk, and informing future resource allocation to improve health equity.