In Ethiopia, women’s dietary diversity is low, primarily due to poor food availability and access, both at home and market level. The present study aimed to describe market access using a new definition called market food diversity (MFD) and estimate the impact of MFD, crop and livestock diversity on dietary diversity among women enrolled in the Agriculture to Nutrition (ATONU) trial.

Baseline cross-sectional data collected from November 2016 to January 2017 were used for the analysis. Availability of foods in markets was assessed at the village level and categorized into nine food groups similar to the dietary diversity index for women. Bivariate and multivariate mixed-effects regression analyses were conducted, adjusted for clustering at the village level.

Chicken-producing farmers in rural Ethiopia.

Women (n 2117) aged 15–49 years.

Overall, less than 6 % of women met the minimum dietary diversity (≥5 food groups) and the most commonly consumed food groups were staples and legumes. Median MFD was 4 food groups (interquartile range: 2–8). Multivariate models indicated that women’s dietary diversity differed by livestock diversity, food crop diversity and agroecology, with significant interaction effects between agroecology and MFD.

Women’s dietary diversity is poor in Ethiopia. Local markets are variable in food availability across seasons and agroecological zones. The MFD indicator captures this variability, and women who have access to higher MFD in the highland agroecological zone have better dietary diversity. Thus, MFD has the potential to mitigate the effects of environment on women’s dietary diversity.

We examined the relationships between exclusive breast-feeding and the risks of respiratory, diarrhoea and nutritional morbidities during the first 2 years of life among children born to women infected with HIV-1.

We prospectively determined the incidence of respiratory illnesses, diarrhoea, fever, hospitalizations, outpatient visits and nutritional morbidities. Generalized estimating equations were used to estimate the relative risks for morbidity episodes and Cox proportional hazards models to estimate the incidence rate ratios of nutritional morbidities.

Dar es Salaam, Tanzania.

The sample consisted of 666 children born to HIV-infected women.

The 666 children were followed for 2 years. Exclusive breast-feeding was associated with lower risk for cough (rate ratio (RR) = 0·49, 95 % CI 0·41, 0·60, P < 0·0001), cough and fever (RR = 0·44, 95 % CI 0·32, 0·60, P < 0·0001) and cough and difficulty breathing or refusal to feed (RR = 0·31, 95 % CI 0·18, 0·55, P < 0·0001). Exclusive breast-feeding was also associated with lower risk of acute diarrhoea, watery diarrhoea, dysentery, fever and outpatient visits during the first 6 months of life, but showed no effect at 6–24 months of life. Exclusive breast-feeding did not significantly reduce the risks of nutritional morbidities during the first 2 years of life.

Exclusive breast-feeding is strongly associated with reductions in the risk of respiratory and diarrhoea morbidities during the first 6 months of life among children born to HIV-infected women.