Vitamin A is an essential micronutrient, especially during pregnancy. We aimed to assess the prevalence of vitamin A deficiency in Brazilian women of childbearing age. We conducted a systematic review with meta-analysis of studies that assessed vitamin A deficiency in women of childbearing age following the registered protocol (CRD42020171856). Independent peer researchers selected the studies retrieved from MEDLINE, Embase, Scopus and other sources. Data from the eligible studies were extracted in pairs and assessed for methodological quality. The prevalence of vitamin A deficiency (< 0·70 µmol/l or <0·20 µg/dl) and 95 % CI was combined by meta-analysis, and heterogeneity was estimated by I2. Out of 3610 screened records, thirty-two studies were included, which assessed 12 577 women from 1965 to 2017, mostly in maternity hospitals. Main limitations of the studies were in sample frame (30/32) and sampling method (29/32). Deficiency occurred in 13 % (95 % CI 9·4, 17·2 %; I² = 97 %) of all women and was higher in pregnant women (16·1 %; 95 % CI 5·6, 30·6 %; I² = 98 %) than non-pregnant women (12·3 %; 95 % CI 8·4, 16·8 %; I² = 96 %). The prevalence increased according to the decade, from 9·5 % (95 % CI 1·9–21·6 %; I² = 98 %) up to 1990, 10·8 % (95 % CI 7·9, 14·2 %; I² = 86 %) in the 2000s and 17·8 % (95 % CI 8·7, 29·0 %; I² = 98 %) in the 2010s. Over 10 % of Brazilian women in childbearing age were deficient in vitamin A. Higher prevalence was observed in pregnant women, and deficiency seemed to be increasing over the decades. Low representativeness of the studies, mainly based on convenience sampling that included pregnant, postpartum, lactating and non-pregnant women, as well as high heterogeneity, limits the findings.
Keywords: Meta-analysis; Prevalence; Systematic review; Vitamin A deficiency; Women.