Effects of Food Insecurity on Nutritional Outcomes during a Pragmatic Weight Loss Trial in an Underserved Adult Population: PROPEL - A Cluster-Randomized Trial

Abstract

Listen

Translate article

ObjectivesPersons with food insecurity (FI) have been shown to have lower fruit and vegetable (F/V) intake and higher fat intake compared to individuals with food security (FS). However, currently there are limited data as to whether diet can be improved during weight loss interventions in persons with FI. MethodsPatients with obesity (BMI 30 - 50 kg/m2) were recruited into the Promoting Successful Weight Loss in Primary Care in Louisiana (PROPEL) trial, which randomized 18 clinics to either an intensive lifestyle intervention (ILI) or usual care (UC). At baseline, FI was determined by the 6-Item Food Security Survey. At baseline and Month (M) 6, 12, and 24, F/V intake was determined by a brief screener and fat intake was determined by the % Energy from Fat Screener. The outcomes were analyzed by repeated-measures linear mixed-effects multilevel models, which included random cluster (clinic) effects and were stratified by FI. In addition to group, time, and their interaction, the models included age, sex, race, education, income, health literacy, and marital status as covariates. ResultsA total of 803 patients were recruited, and 30.8% reported FI. About 84% were female, 67.2% black, 26.1% received Medicaid, and 65.5% made less than $40,000. At baseline, patients with FI reported lower F/V intake than patients with FS (2.04 ± 0.11 vs. 2.32 ± 0.07 servings/day [d]; mean ± SEM), but no differences in fat intake were seen (35.30 ± 0.41 vs. 35.34 ± 0.27%). In patients with FI, the ILI only improved F/V intake compared to UC at M 24 (0.70 ± 0.28 servings/d). In patients with FS, F/V intake was improved in the ILI vs. UC groups at M 6 (0.55 ± 0.16), 12 (0.39 ± 0.16), and 24 (0.39 ± 0.17 servings/d). Fat intake was not improved in ILI patients with FI, but patients with FS decreased fat intake at all time points compared to UC (M6 –2.79 ± 0.91, M12 –2.51 ± 0.92, and M24 –2.55 ± 0.92%, respectively). ConclusionsOn average, patients with FI in the ILI improved F/V intake, but the high variability (i.e., heterogeneity) in this nutritional outcome prevented significant differences between the ILI and UC groups. Patients with FI failed to decrease fat intake in response to a weight loss intervention. For persons with FI to achieve greater weight loss, targeting nutritional outcomes may be an important consideration. Funding SourcesPCORI OB-1402–10,977, NIH U54 GM104940 and P30 DK072476.