RESUMO
Importance: In the Supplemental Nutrition Assistance Program (SNAP), families may temporarily lose benefits for which they are still eligible because of administrative issues. This lapse in benefits, referred to as churning, increases the risk of food insecurity for families, which is linked with poorer health. Objectives: To examine the rate of churning among SNAP participants with young children and evaluate the association of administrative policy changes with churning risk. Design, Setting, and Participants: A cross-sectional study of recertifications among 9735 SNAP-participating households with at least 1 child younger than 6 years from May to November 2019 tested whether there were sociodemographic differences in churning risk. An interrupted time series analysis of recertifications among 70â¯799 households from January 1, 2014, to December 31, 2019, tested whether state-level administrative policy changes were associated with churning risk. Data were analyzed between February and November 2021. Exposures: Three Massachusetts SNAP administrative policy changes. Main Outcomes and Measures: Churn episodes, defined as losing SNAP benefits after a deadline for eligibility recertification followed by receiving benefits again within 30 days, measured using state SNAP administrative data. Results: In 2019, a total of 9752 recertification events occurred among 9735 households (3841 [39.4%] Hispanic, 2138 [21.9%] non-Hispanic Black, and 3533 [36.2%] White) with at least 1 child younger than 6 years participating in SNAP. Of these households, 3984 (40.9%) experienced SNAP churning for a period of 1 to 30 days because of missed recertification forms in any given month. Churning was more common among Hispanic households (by 3.9 percentage points; 95% CI, 1.2-6.6 percentage points) than non-Hispanic White households and more common among households with more than 1 child younger than 6 years (by 5.6 percentage points; 95% CI, 3.0-8.1 percentage points) compared with smaller households. Among households that churned, non-Hispanic White households lost a mean of $10.51 (95% CI, $10.16-$10.86) in benefits per household member, and non-Hispanic Black households lost an additional $1.10 (95% CI, $0.47-$1.74) per household member. On implementation of a task-based, first-available-caseworker model, mean churn episode length was significantly shorter immediately on introduction of the model (-0.30 days per each month after the policy change; 95% CI, -0.52 to -0.07 days) and over time (-0.23 days per each month after the policy change; 95% CI, -0.26 to -0.20 days). At the addition of a second policy change to expand participants' access to simplified reporting requirements, a decrease was seen in the trend in the probability of a churn episode by -0.74 percentage points per month (95% CI, -0.87 to -0.62 percentage points). A third policy to expand online services for submitting recertification paperwork was not associated with a change in churn risk over time. Conclusions and Relevance: Churning in SNAP is common among households with young children and is disproportionately experienced by Black and Hispanic households and lower-income households. Administrative changes to simplify caseworkers' workloads and streamline eligibility reporting for participants may reduce churn.
