Association between Vitamin A and Anemia in Chinese Children: A Matched Case-Control Study

Vitamin A plays an important role in the development of anemia, providing a new perspective on improving anemia.

This study aimed to investigate the dose-response relationship between serum vitamin A and anemia risk, and evaluate its clinical applicability with adjustment for covariates.

A 1:3 matched case-control study (202 cases, 606 controls) was conducted, matching for sex and age (±2 y). Assess matching balance using the standardized mean difference. Anemia was defined according to the World Health Organization criteria. Unconditional logistic regression was used to assess the effect of vitamin A after adjusting for confounders. Variable adjustment was informed by a directed acyclic graph, with the primary model adjusting for age, sex, body mass index (BMI), and serum zinc (Zn). A restricted cubic spline (RCS) was applied to analyze the dose-response relationship. Subgroup analysis and interaction tests were used to examine the association across sexes. Model discrimination was evaluated using the receiver operating characteristic curve and its area under the curve (AUC). Model calibration was evaluated using calibration plots, the Hosmer-Lemeshow test, and the Brier score.

The case group exhibited significantly lower amounts of BMI, vitamin D, vitamin A, total cholesterol, Zn, and serum ferritin (all P < 0.05). After adjusting for age, sex, BMI, and Zn, each 0.1 μg/mL increase in serum vitamin A was associated with a 27% reduction in anemia risk [odds ratio (OR) = 0.73, 95% confidence interval (CI): 0.62, 0.87, P < 0.001]. RCS analysis confirmed a linear dose-response relationship (P for nonlinearity = 0.922). Sex-stratified analysis using unconditional logistic regression suggested a numerically stronger association in males (OR = 0.55, 95% CI: 0.38, 0.78, P < 0.001) than in females (OR = 0.82, 95% CI: 0.67, 1.00, P = 0.053); "sex × vitamin A" interaction was statistically significant (P = 0.019). The prediction model including vitamin A improved AUC to 0.63 (ΔAUC = 0.11, P < 0.001) and demonstrated good calibration (Brier score = 0.180; Hosmer-Lemeshow P = 0.636).

Serum vitamin A demonstrates a significant, independent, and linear inverse association with anemia risk in children and adolescents. This association was more pronounced in males, indicating a potential role for vitamin A in sex-specific anemia prevention strategies.