oral Preconception Dietary Adherence to the 2010 Healthy Eating Index and Birth Outcomes Among Hispanic/Latina Women in the HCHS/SOL Study

Abstract

Epidemiologic evidence suggests that a mother’s metabolic environment during the periconception period can adverselyaffect her offspring’s health, not only at birth, but subsequently in childhood and beyond. We assessed overall diet qualityduring the preconception period and its association with birth outcomes, not previously explored in a U.S. Hispanic/Latinapopulation. The Hispanic Community Health Study/Study of Latinos (HCHS/SOL) is the largest community-based cohort studyof U.S. Hispanic/Latino adults aged 18-74 years (n=16,415, including 3,801 women 18-44y). The baseline clinic examination wasconducted in 2008-2011, with yearly telephone follow-up assessments, and an ongoing second clinic examination (2014-2017)which includes ascertainment of birth outcomes. Diet assessment included up to two 24-hour recalls conducted at the baselineexamination. In the full cohort, we found that individuals of Cuban and Puerto Rican heritage had relatively higher intakesof foods and nutrients (e.g., total fat, saturated fat, sodium, refined carbohydrates, and red meats) that were associated with anelevated risk of chronic disease, and lower intakes of foods and nutrients (e.g., fiber, folate, fish and fruits) that were associatedwith a lower risk of chronic disease.1,2 This preliminary analysis describes results from the first 463 singleton live- born infantsdelivered after the baseline exam among 2,287 women (<45 yrs) who attended the second clinic visit by May 2017. The HealthyEating Index (HEI), a measure of diet quality, was used to assess adherence to the U.S. dietary guidelines with higher valuesindicating a healthier diet (range 0- 100). We hypothesized that lower HEI scores would be associated with a higher birth weightzscore and higher birth weight for gestational age (GA) percentile and that these associations would vary by preconception BMI(measured at the baseline visit). Complex survey linear regression was used to estimate the association between HEI scores (asboth a continuous variable and categorized into tertiles) and birth weight z-score and birth weight for GA percentiles overall andby preconception BMI. On average women were 26 years of age and 37.5% were under/normal weight, 30.8% were overweightand 31.8% were obese at the baseline visit. Overall, the mean HEI score was 56.4 and by BMI were 55.6 and 57.5 for under/normal weight, and overweight/obese, respectively. The median birth weight z-score was 0.8 for all, and 0.7, 0.7 and 0.9 for thefirst, second and third HEI tertile, respectively. This corresponded to a birth weight for GA percentile of 79.6 for all, and 77.1,77.3 and 81.3 for the first, second and third tertile. A significant interaction between HEI score and preconception BMI wasdetected (p=0.01). Linear regression results for the association between adherence to the 2010 HEI and birth outcomes by BMI,adjusting for several covariates, show that for under/normal weight women, a higher HEI score, indicative of a better-qualitydiet, is associated with higher birth weight outcomes. Whereas for overweight and obese women, the association, while nonsignificant,was in the reverse direction with higher scores being associated with lower birthweight z-score and percentile. Insummary, overall diet quality, as measured by the HEI, in the preconception period appears to influence birth weight outcomesin the HCHS cohort.