To estimate the prevalence of food insecurity (FI) among U.S. students and assess how sample characteristics correlate with the prevalence. Studies were identified via systematic searches in 4 electronic databases, reviewing websites of relevant organizations, and contacting authors of related studies. For inclusion, studies needed to have: 1) collected data and/or been published after 1995; 2) been written in English; 3) reported quantitative categorical outcomes of U.S. college students related to FI; 4) indicated method of evaluating FI; and 5) provided no intervention. Studies were independently screened for inclusion by 2 researchers. Data extractions for each study included design, sampling and recruitment strategies, sample size and demographics, institution type, FI assessment methods, and reported FI prevalence. Prevalence of FI among university students was estimated overall, by institution type, and by FI assessment method, using a random effects model. In total, 54 unique samples were meta-analyzed from 70 published reports. The majority were cross-sectional (96%) and used convenience or census sampling (67%) with sample sizes ranging from 49 to 26,131. The weighted prevalence of FI among college students was 37% (95% CI = 32%, 42%). Samples that included community college students resulted in significantly greater (P = 0.019) FI prevalence estimates when compared to samples with exclusively university students (44% [95% CI = 31%, 57%] versus 36% [95% CI = 31%, 41%]). The survey used to classify FI status was also significantly related to the resulting prevalence estimate (P < 0.001). The abbreviated 6-item USDA Food Security Survey Module resulted in the most liberal estimate (46%; 95% CI = 41%, 52%) while the 18-item household module was the most conservative (17%, 95%CI = 12%, 22%). About 37% of U.S. college students report experiences of FI. This estimate is impacted by academic institution type and survey tool used. Given that estimates of FI among students are used to garner support for interventions and policy changes, it is concerning that estimates are impacted by the survey tool used. Further studies evaluating the accuracy of FI estimates among this population are warranted. This work was supported by the USDA National Institute of Food and Agriculture grant no. ILLU-470-356.
No takes yet. Share an insight, caveat, or question.
Nikolaus et al. (2019) studied this question.