Joseph Rigdon1, Seth A Berkowitz2, Hilary K Seligman3, Sanjay Basu4. 1. Quantitative Sciences Unit, Stanford University School of Medicine, 1070 Arastradero Rd #3C3104 MC 5559, Palo Alto, CA 94304, United States. Electronic address: jrigdon@stanford.edu. 2. General Internal Medicine and Diabetes, Massachusetts General Hospital, 50 Staniford St., 9th Floor, Boston, MA 02114, United States. Electronic address: SABerkowitz@mgh.harvard.edu. 3. Division of General Internal Medicine, University of California San Francisco, 1001 Potrero Ave, San Francisco, CA 94110, United States. Electronic address: Hilary.Seligman@ucsf.edu. 4. Stanford University School of Medicine, 1070 Arastradero Rd, Palo Alto, CA 94304, United States. Electronic address: basus@stanford.edu.
Abstract
OBJECTIVE: To evaluate the association between participation in the Supplemental Nutrition Assistance Program (SNAP) and body mass index (BMI) in the presence of unmeasured confounding. METHODS: We applied new matching methods to determine whether previous reports of associations between SNAP participation and BMI were robust to unmeasured confounders. We applied near-far matching, which strengthens standard matching by combining it with instrumental variables analysis, to the nationally-representative National Household Food Acquisition and Purchasing Survey (FoodAPS, N = 10,360, years 2012-13). RESULTS: In ordinary least squares regressions controlling for individual demographic and socioeconomic characteristics, SNAP was associated with increased BMI (+1.23 kg/m2, 95% CI: 0.84, 1.63). While propensity-score-based analysis replicated this finding, using instrumental variables analysis and particularly near-far matching to strengthen the instruments' discriminatory power revealed the association between SNAP and BMI was likely confounded by unmeasured covariates (+0.21 kg/m2, 95% CI: -3.88, 4.29). CONCLUSIONS: Previous reports of an association between SNAP and obesity should be viewed with caution, and use of near-far matching may assist similar assessments of health effects of social programs.
OBJECTIVE: To evaluate the association between participation in the Supplemental Nutrition Assistance Program (SNAP) and body mass index (BMI) in the presence of unmeasured confounding. METHODS: We applied new matching methods to determine whether previous reports of associations between SNAP participation and BMI were robust to unmeasured confounders. We applied near-far matching, which strengthens standard matching by combining it with instrumental variables analysis, to the nationally-representative National Household Food Acquisition and Purchasing Survey (FoodAPS, N = 10,360, years 2012-13). RESULTS: In ordinary least squares regressions controlling for individual demographic and socioeconomic characteristics, SNAP was associated with increased BMI (+1.23 kg/m2, 95% CI: 0.84, 1.63). While propensity-score-based analysis replicated this finding, using instrumental variables analysis and particularly near-far matching to strengthen the instruments' discriminatory power revealed the association between SNAP and BMI was likely confounded by unmeasured covariates (+0.21 kg/m2, 95% CI: -3.88, 4.29). CONCLUSIONS: Previous reports of an association between SNAP and obesity should be viewed with caution, and use of near-far matching may assist similar assessments of health effects of social programs.
Authors: Cindy W Leung; Susan J Blumenthal; Elena E Hoffnagle; Helen H Jensen; Susan B Foerster; Marion Nestle; Lilian W Y Cheung; Dariush Mozaffarian; Walter C Willett Journal: Pediatrics Date: 2013-02-25 Impact factor: 7.124
Authors: Eva Tseng; Allen Zhang; Oluwaseun Shogbesan; Kimberly A Gudzune; Renee F Wilson; Hadi Kharrazi; Lawrence J Cheskin; Eric B Bass; Wendy L Bennett Journal: J Gen Intern Med Date: 2018-09-11 Impact factor: 5.128
Authors: Seth A Berkowitz; Jean Terranova; Liisa Randall; Kevin Cranston; David B Waters; John Hsu Journal: JAMA Intern Med Date: 2019-06-01 Impact factor: 21.873