Vitamin-D status is not a confounder of the relationship between zinc and diarrhoea: a study in 6–24-month-old underweight and normal-weight children of urban Bangladesh

The role of micronutrients particularly zinc in childhood diarrhoea is well established. Immunomodulatory functions of vitamin-D in diarrhoea and its role in the effect of other micronutrients are not well understood. This study aimed to investigate whether vitamin-D directly associated or confounded the association between other micronutrient status and diarrhoeal incidence and severity in 6–24-month underweight and normal-weight children in urban Bangladesh. Multivariable generalised estimating equations were used to estimate incidence rate ratios for incidence (Poisson) and severity (binomial) of diarrhoea on cohorts of 446 normal-weight and 466 underweight children. Outcomes of interest included incidence and severity of diarrhoea, measured daily during a follow-up period of 5 months. The exposure of interest was vitamin-D status at enrolment. Normal-weight and underweight children contributed 62 117 and 62 967 day observation, with 14.2 and 12.8 days/child/year of diarrhoea, respectively. None of the models showed significant associations of vitamin-D status with diarrhoeal morbidity. In the final model, zinc-insufficient normal-weight children had 1.3 times more days of diarrhoea than sufficient children (P<0.05). Again zinc insufficiency and mother's education (1–5 and >5 years) had 1.8 and 2.3 times more risk of severe diarrhoea. In underweight children, older age and female had 24–63 and 17% fewer days of diarrhoea and 52–54 and 31% fewer chances of severe diarrhoea. Vitamin-D status was not associated with incidence and severity of diarrhoea in study children. Role of zinc in diarrhoea was only evident in normal-weight children. Our findings demonstrate that vitamin-D is not a confounder of the relationship between zinc and diarrhoea.