Noncommunicable diseases, including cardiovascular diseases, diabetes, and others, are the leading cause globally of premature mortality (death before the age of 70), with the majority of deaths occurring in low- and lower-middle-income countries. Two prominent conditions contributing to noncommunicable disease risk include overweight or obesity and hypertension. Worldwide, >40% of adults and almost 6% of children experience overweight and obesity, and >15% of adults have raised blood pressure. Alongside rising prevalence of overweight or obesity and hypertension, micronutrient deficiencies persist, especially in low- and lower-middle-income countries. The United Nations’ Sustainable Development Goal (SDG) 2.2 calls for an end to malnutrition in all its forms by 2030, and SDG 3.4 aims to reduce premature deaths from noncommunicable diseases by one-third by 2030 (relative to 2015 levels). However, progress towards these goals is slow, with most countries worldwide off track. Given the potential for conditions of both undernutrition and overnutrition to increase the risk of developing noncommunicable diseases, a greater understanding of the relationships between selected contributors to noncommunicable disease risk, and how they are measured, are needed to reach the SDG targets. Inflammation is the body’s physiological response to injury, illness, infection, or environmental insult, and is characterized by the presence of pro-inflammatory cytokines and acute-phase proteins, such as C-reactive protein (CRP) and α-1-acid glycoprotein (AGP). In populations in LIC and LMIC, inflammation may be more likely associated with illnesses such as diarrhea or helminth infections, while populations in upper-middle and high-income countries may more commonly experience the chronic inflammation associated with obesity. Inflammation confounds the assessment of the micronutrients iron and vitamin A, as the biomarkers commonly used to measure iron and vitamin A status are also acute phase proteins. In the presence of inflammation, serum or plasma concentrations of ferritin transiently decrease, and serum or plasma concentrations of retinol and retinol-binding protein (RBP) increase. In Chapters 2 and 3, I explored the extent to which adiposity-related inflammation may influence ferritin and retinol or RBP interpretation.
In Chapter 2, I describe relationships between weight status, inflammation, and ferritin among non-pregnant women of reproductive age (15-49 years, WRA) and preschool-age children (6-59 months, PSC) with normal weight to overweight or obesity in differing geographic settings. Cross-sectional data were separately analyzed from n=18 surveys (WRA) and n=25 surveys (PSC) from the Biomarkers Reflecting Inflammation and Nutritional Determinants of Anemia (BRINDA) project, excluding observations with underweight, wasting, pregnancy, or malaria. Relationships were assessed between BMI (WRA) or BMI-for-age z-score (BAZ, PSC), inflammatory biomarkers CRP and/or AGP, and ferritin by linear regression, and potential mediation by CRP and/or AGP in relationships between BMI or BAZ and ferritin with structural equation modeling. Regression and mediation models accounted for complex survey designs, and results were grouped by World Bank income classifications. In 5 of 6 surveys among WRA from upper-middle and high-income countries, ferritin was significantly positively associated with BMI, and this relationship was partially (or fully in the survey from the United States) mediated by CRP and/or AGP. Mediation was present in 4 of 12 surveys for WRA in low- and lower-middle income countries. Among PSC, ferritin was positively associated with CRP and/or AGP in all surveys, but there were no significant CRP- or AGP-mediated relationships between ferritin and BAZ, except a negative relationship in the Philippines. I concluded that where overweight and obesity are common among WRA, measurement of inflammatory biomarkers and their use in interpreting ferritin may improve iron status assessment. While these relationships were inconsistent among PSC, inflammation was common and should be measured to interpret iron status.
In Chapter 3, I conducted similar analyses to examine relationships between weight status, inflammation, and retinol or RBP among WRA and PSC with normal weight to overweight or obesity. BRINDA data from n=13 surveys (WRA) and n=22 surveys (PSC) were separately analyzed, excluding observations with underweight, wasting, pregnancy, or malaria. Relationships were assessed between BMI (WRA) or BAZ (PSC), CRP and/or AGP, and retinol or RBP by linear regression, and potential mediation by CRP and/or AGP in relationships between BMI or BAZ and retinol or RBP with structural equation modeling. All regression and mediation models accounted for complex survey designs. Among WRA, greater BMI was positively associated with retinol or RBP in 5 of 13 surveys, BMI was positively associated with CRP and/or AGP in 10 of 13 surveys, but associations between biomarkers of inflammation and retinol or RBP were inconsistent. Among PSC, BMI was not associated with retinol, RBP, CRP, or AGP, but biomarkers of inflammation were consistently negatively associated with retinol or RBP. In 3 of 13 surveys among WRA and 1 of 22 surveys among PSC, inflammation partially mediated the relationship between BMI or BAZ and retinol or RBP, however the direction of association varied. I concluded that in these surveys, inflammation associated with overweight and obesity does not appear to impact vitamin A assessment when measured with retinol or RBP; however, inflammation should continue be measured to interpret vitamin A status among PSC.
Chapter 4 explores salt consumption in Ghana, where salt consumption ranges 6-12 g/d, and salt consumption ≥5 g/d is associated with increased risk of noncommunicable diseases. To develop salt reduction strategies that are relevant to this context, understanding salt usage and consumption patterns is necessary. My objectives for this chapter were to: 1) estimate consumption of salt, including salt from bouillon, among households, women, and children, and compare to global recommendations; 2) estimate the proportion of salt consumed from bouillon; and 3) identify factors, including knowledge, attitudes, and practices (KAP), associated with household salt consumption in 2 districts in Northern Region, Ghana. Employing mixed-methods methodology, households were enrolled from 14 urban and 14 rural clusters from Tolon and Kumbungu districts in a pilot survey and focus group discussions (FGDs, n=20). Using the Fortification Assessment Coverage Toolkit, households (n=369) reported most recent purchases of discretionary salt (DS, ‘table salt’) and bouillon cubes. From purchase data, median (IQR) household consumption (g/d) of DS and total salt (TS, DS + salt from bouillon, assumed to be 55% salt) were calculated, including the proportion of salt from bouillon. DS and TS consumption for women (15-49 y) and children (2-5 y) were estimated with the Adult Male Equivalent method and compared to global recommendations. Salt intake from urinary sodium excretion was predicted with the INTERSALT equation (women only). Associations between DS and TS consumption and household and individual (women’s) characteristics, including KAP, were tested with mixed effects ANOVA. Minimally-adjusted and multivariable models included district, setting (urban/rural), household size, and participant type (non-lactating or lactating woman) as fixed effects, and the random effect of cluster. Qualitative themes were generated from FGDs using the Framework Method. From reported household purchase data, estimated consumption of DS and TS appeared to exceed global recommendations for many children (TS: 2.9 [1.9, 5.2] g/d) and the majority of women (TS: 6.0 [4.0, 10.2] g/d). Women’s mean urinary sodium excretion also suggested high sodium exposure (7.1 g/d). Bouillon contributed <25% to households’ daily TS consumption. Household salt consumption was greater among households in 3rd-5th (highest) asset quintiles and those with severe food insecurity. Few other characteristics were associated with household salt consumption. Salient qualitative themes included salt’s ubiquity as a seasoning, and how intra-household dynamics, taste preferences, and perceptions about salt and health shaped salt usage and consumption. These results suggest that salt consumption among women and children in this area exceeds recommendations; food prepared outside the home may further contribute to salt consumption. Salt reduction interventions may be warranted in this context.
Together, these studies broaden our understanding of how measuring indicators of iron and vitamin A status relate to noncommunicable disease risk assessment in different global settings, which will aid global nutrition status surveillance efforts. Also, the salt consumption results from Ghana will help inform nutrition and policy discussions related to salt in Ghana, including the development of salt-reduction behavior change communication strategies.
