Food Insecurity and Odds of High Allostatic Load in Puerto Rican Adults: The Role of Participation in the Supplemental Nutrition Assistance Program During 5 Years of Follow-Up

• Published in: Psychosomatic medicine 2018-10, Vol.80 (8), p.733-741
• Main Authors: McClain, Amanda C, Xiao, Rui S, Gao, Xiang, Tucker, Katherine L, Falcon, Luis M, Mattei, Josiemer
• Format: Article
• Language: English
• Subjects: Aged; Allostasis - physiology; Boston - ethnology; Chronic illnesses; Confidence intervals; Data processing; Disease; Female; Follow-Up Studies; Food Assistance - statistics & numerical data; Food security; Food stamps; Food Supply; Healthy food; Humans; Inflammation; Insecurity; Longitudinal Studies; Male; Middle Aged; Moderation; Nutrition; Participation; Poverty - ethnology; Public assistance programs; Puerto Ricans; Puerto Rico - ethnology; Sociodemographics
• ISSN: 0033-3174; 1534-7796
• DOI: 10.1097/PSY.0000000000000628

Limited evidence demonstrates pathways linking food insecurity (FI) to chronic disease. Allostatic load (AL) may elucidate potential pathways, capturing both primary (neuroendocrine, inflammation) and secondary (metabolic, cardiovascular) physiological disturbances. We examined the longitudinal association of FI with 5-year AL and primary and secondary subsystem dysregulation and tested moderation by Supplemental Nutrition Assistance Program (SNAP) participation. We analyzed data from the longitudinal Boston Puerto Rican Health Study among 733 adults aged 45 to 75 years. Participants categorized as food insecure (assessed by US survey module) experienced FI at baseline and/or year 5. AL score comprised 11 biological components (5 primary, 6 secondary). We classified participants as having high scores for AL (≥6 dysregulated components), primary system (≥3), and secondary system (≥4). Multivariate models estimated odds ratios (OR), adjusting for baseline AL, sociodemographic, cultural, and behavioral characteristics. By study end, 33.8% had experienced FI, 65.5% had participated in SNAP, and 37.5% had high AL. In adjusted models, FI was not associated with AL (OR [95% confidence intervals] = 1.07 [0.70-1.64]) or secondary system (0.82 [0.48-1.40]) scores, but was associated with high primary system scores (1.71 [1.25-2.36]). SNAP participation seemed to moderate the FI-primary system relationship (p = .06); food-insecure participants never receiving SNAP (mean (SE) = 2.06 (0.14)) had higher scores than food-secure participants receiving (1.72 (0.06], p = .02) or never receiving SNAP (1.64 (0.10), p = .01) and food-insecure participants receiving SNAP (1.80 (0.07), p = .08). FI is associated with dysregulated components of the primary AL system, and this relationship may be stronger for those not receiving SNAP. Research is needed in additional populations to test AL as a plausible pathway connecting FI to chronic disease and SNAP as a moderator.