Estimation of the vitamin D (VD) status of pregnant Japanese women based on food intake and VD synthesis by solar UV-B radiation using a questionnaire and UV-B observations.

Although vitamin D (VD; serum 25 hydroxyvitamin D) deficiency (< 20 ng/mL) is widespread among Japanese women, the VD status among pregnant women is unknown. This study aimed to determine the VD status of pregnant Japanese women during different meteorological seasons and to determine the factors controlling VD status. A total of 309 pregnant Japanese women were recruited at 28 weeks of gestation at the gynecology department of a university hospital in Tokyo between August 2018 and October 2019. Blood samples were collected to measure serum 25(OH)D levels. Two questionnaires were completed: a brief self-administered dietary history questionnaire (BDHQ) and an outdoor exposure history questionnaire to determine skin sunlight exposure and the use of sunscreen. Among the recruited subjects, 268 were included in the statistical analysis. The average VD intake from food was 9.0 μg/day, the average VD synthesis from UV-B was 15.2 μg/day, and the average sum of VD intake and nominal VD synthesis was 24.1 μg/day; this exceeded the recommended 2011 Dietary Reference Intake for the USA and Canada (15.0 μg/day). However, the average serum 25(OH)D level (11.4 ng/mL) was very low, indicating widespread VD deficiency. Serum 25(OH)D and VD synthesis by solar UV-B were significantly correlated only during the high UV-B season. The 25(OH)D level was weakly correlated with the VD intake from food in all seasons. We obtained a statistically significant correlation between serum 25(OH)D level and VD intake from food using the BDHQ. We also obtained a statistically significant correlation between the serum 25(OH)D level and VD synthesis from solar UV-B exposure, especially during the high UV-B season. Our logistic regression analysis model predicted VD deficiency in 88.0% of subjects. Our method might be possible to be used to predict the VD status of pregnant Japanese women, although another validation cohort is needed to verify the ability of the estimation equation.