The Relation between Food Security, Iron Status, and Depressive Symptoms in Women of Reproductive Age: a Cross-Section Analysis from NHANES 2005 – 2010
Open Access
- Author:
- Ciulei, Mihaela A.
- Graduate Program:
- Nutritional Sciences
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- September 29, 2020
- Committee Members:
- Laura E Murray-Kolb, Dissertation Advisor/Co-Advisor
Laura E Murray-Kolb, Committee Chair/Co-Chair
Connie Jo Rogers, Committee Member
Douglas Michael Teti, Outside Member
Laura E Murray-Kolb, Program Head/Chair
Penny Margaret Kris-Etherton, Committee Member
Namanjeet Ahluwalia, Special Member - Keywords:
- food security
depressive symptoms
iron status
women of reproductive age
non-pregnant
nationally representative
NHANES
United States - Abstract:
- Iron deficiency (ID) is the most common single-nutrient deficiency in the US, where it affects approximately 10% of non-pregnant women of reproductive age (WRA). This group of women is also at high risk for food insecurity and depressive symptoms (DS) when compared to their male counterparts. Little is known on the relation between food security (FS) and iron status and between iron status and DS. This dissertation explores the following relations: 1) whether higher food insecurity (FIS) is related to worse iron status (with a subaim to examine whether SNAP or WIC participation alters the relation between FS and iron status (using only those eligible for SNAP or WIC participation)), 2) whether worse iron status is related to higher DS, and 3) whether FIS is related to worse DS. These aims were investigated using the cross-sectional data from the National Health and Nutrition Examination Survey (NHANES) 2005 – 2010. We included 2507 non-pregnant women (20-44 years) with complete data for iron biomarkers, DS (Patient Health Questionnaire-9), FS (Household Food Security Survey Module), BMI, sociodemographic variables, and antidepressant use. Ferritin (Ft), transferrin receptor (TfR), body iron (BodyFe) and hemoglobin (Hb) were used to classify individuals as iron sufficient (IS), ID, and ID anemic (IDA). ANCOVA, simple and multinomial logistic regression, and negative binomial (NB) were run, accounting for covariates. In the total sample, 8.6% of women were anemic, the prevalence of ID ranged from 7.7-15.3% and the prevalence of IDA among those who were anemic ranged from 42.0-61.6%, based on the iron biomarker used. Adjusted models revealed: Aim 1) WRA with marginal or very low FS were less likely to be ID compared with those who were fully FS. However, those with low FS were 1.69 (95% CI: 1.06-2.71) more likely to be anemic than their fully FS counterparts. When limited to just those with a poverty to income ratio (PIR) ≤ 1.85, food insecure WRA were 1.81 (95% CI: 1.07-3.07) times more likely to be ID compared to their FS counterparts. The relation between FS and iron status was not modified by SNAP or WIC participation. However, women who were fully FS and participated in SNAP had higher mean Ft and Hb concentrations than those who were fully FS but did not participate in SNAP (p=0.045 and p=0.01). Whereas if they were food insecure, WIC participants had lower mean BodyFe concentrations than non-participants (p=0.04) and if they were of very low FS status, WIC participants had lower mean Ft and BodyFe concentrations than WIC non-participants (p=0.02 and p=0.01). Aim 2) ID (low Ft and TfR or low BodyFe) rather than IS women were 1.95- (95% CI:1.03-3.70) and 1.84- (95% CI:1.09-3.10) times more likely to show moderately severe rather than none/mild DS. Further, the DS scores were 1.20- and 1.23 times higher for ID (TfR, Ft and TfR) rather than for IS WRA, p=0.04 and p=0.02. Low-income WRA with ID (TfR) rather than IS were 1.64 (95% CI:1.03-2.61) times more likely to have moderate to severe rather than none/mild DS, and the DS score were 1.29- and 1.27 times higher for ID (TfR, Ft and TfR) rather than for IS WRA, p=0.03 and p=0.04, respectively. (Aim 3) Women in the total sample, those with PIR < 1.85, and those with PIR > 1.85, who were food insecure rather than FS had 2.67-(95% CI: 1.89-3.76), 2.49-(95% CI: 1.68-3.70), and 4.16-(95% CI: 2.02-8.59) times higher odds, respectively, for moderately severe to severe rather than none/mild DS. Further, food insecure WRA from the total sample, PIR < 1.85, and PIR > 1.85 had 76%, 70%, and 92%, respectively higher DS scores than FS ones. In conclusion, this study contributes to the limited evidence on FS, iron status, and DS by using a US nationally representative sample of non-pregnant WRA. Our findings indicate that food insecure women have higher odds of ID and DS than FS women, and ID women have higher odds and scores of DS when compared to IS women. These findings were more prevalent among those with the lowest PIR and who are eligible for welfare.