The final copy of this thesis has been examined by the signatories, and we find that both the content and the form meet acceptable presentation standards of scholarly work in the above-mentioned discipline.

Objective: Relatively little is known about sources of variation in human breastmilk composition. Previous research posits that composition may be affected by the low oxygen environment found at high altitude (HA). Our objective is to compare the milk composition of a sample of HA Andeans to previously published data on HA Tibetans and low altitude (LA) Peruvians. Methods: Mid-feed milk samples were collected from 22 women and tested for lipid, carbohydrate and protein content in addition to dietary surveys and anthropometric measures. We examined relationships between maternal and infant characteristics (maternal BMI, adiposity, age, skinfold thickness, infant age) and milk composition using multiple regression analyses. A Welch-Satterthause t-test was used to compare summary measures from Andean sample and Tibetan samples. Results: Milk fat, sugar, and true protein content were correlated with different maternal and infant characteristics, with only waist to hip ratio negatively affecting fat and positively affecting sugar content in full models. Nuñoan and LA Peruvian milk exhibited significantly lower lactose levels (6.01 ± .89 g/100 mL and 6.38 ± 0.03 g/100 mL) than Tibetans (7.25 ± 0.35 g/100 mL). Fat content among Nuñoans was similar to that of Tibetans (4.5 ± 0.78 g/100 mL and 4.6 ± 1.51 g/100 mL, respectively), and higher than those of LA Peruvians. Conclusions: Both HA populations demonstrate relatively high fat content and both Peruvian samples exhibited relatively higher protein and lower lactose than Tibetan samples, suggesting that both population-based differences in feeding behavior and climatic differences may influence composition.

Background: Although breastfeeding is common in The Gambia, high rates of undernutrition in children under five highlight the importance of understanding drivers of particular feeding patterns in this environment, especially early introduction of non-breast milk foods (NBMFs). The country’s marked seasonality, which is associated with annual food insecurity and heavy maternal workload, may influence breastfeeding patterns; however, longitudinal assessments of infant diet in relationship to such factors are limited. We aimed to characterize infant breastfeeding patterns and timing of introduction of local complementary foods in a rural Gambian population across the first twelve months of life. Potential environmental and sociodemographic predictors of exclusive breastfeeding (EBF) duration were explored in order to identify factors that may influence infant feeding decisions in this population. Methods: Data from dietary questionnaires (administered every ten days until 12 months of age) collected as a part of the Hormonal and Epigenetic Regulators of Growth study (2013-2018) were used to calculate EBF duration in a subsample of 194 mother-infant pairs. Socioeconomic questionnaires and Principal Component Analysis were used to calculate household sociodemographic position (SEP). Multiple linear regression analyses were used to investigate potential predictors of EBF duration, including seasonality, SEP, and maternal and infant factors. Results: Mean age at introduction of food or liquid other than maternal milk was five months (±1.5). At twelve months, 98.7% of infants continue to receive some maternal milk. Being born in  May significantly predicted shorter EBF duration by -1.68 months (95% CIs: -2.52, -0.84mo; P<.0001). SEP, maternal parity, and infant sex were non-significant predictors of EBF duration. Conclusions: Maternal milk is a vital component of infant diet across the first twelve months of life in this population. Earlier introduction of NBMFs coincides with the annual period where maternal agricultural workload intensifies in this region, though additional investigation is warranted.