Low maternal vitamin D status in pregnancy increases the risk of childhood obesity

BACKGROUND

Vitamin D may modulate adipogenesis. However, limited studies have investigated the effect of maternal vitamin D during pregnancy on offspring adiposity or cardiometabolic parameters with inconclusive results.

OBJECTIVES

The objective of this study is to examine the association of maternal 25(OH)-vitamin D [25(OH)D] status with offspring obesity and cardiometabolic characteristics in 532 mother-child pairs from the prospective pregnancy cohort Rhea in Crete, Greece.

METHODS

Maternal 25(OH)D concentrations were measured at the first prenatal visit (mean: 14 weeks, SD: 4). Child outcomes included body mass index standard deviation score, waist circumference, skin-fold thickness, blood pressure and serum lipids at ages 4 and 6 years. Body fat percentage was also measured at 6 years. Body mass index growth trajectories from birth to 6 years were estimated by mixed effects models with fractional polynomials of age. Adjusted associations were obtained via multivariable linear regression analyses.

RESULTS

About two-thirds of participating mothers had 25(OH)D concentrations <50 nmol L^-1 . Offspring of women in the low 25(OH)D tertile (<37.7 nmol L^-1 ) had higher body mass index standard deviation score (β 0.20, 95% CI: 0.03, 0.37), and waist circumference (β 0.87 95% CI: 0.12, 1.63) at preschool age, compared with the offspring of women with higher 25(OH)D measurements (≥37.7 nmol L^-1 ), on covariate-adjusted analyses. The observed relationships persisted at age 6 years. We found no association between maternal 25(OH)D concentrations and offspring blood pressure or serum lipids at both time points.

CONCLUSIONS

Exposure to very low 25(OH)D concentrations in utero may increase childhood adiposity indices. Given that vitamin D is a modifiable risk factor, our findings may have important public health implications.

Reliability of drug dependents' self-reports

Among 936 demands for treatment by drug users in Greece in 7 months in 1994, 78 subjects making at least two demands were identified by code numbers. The median time between demands was 50 days (range 2-219). The data obtained by interview at the first two demands were compared to estimate the reliability of the interview schedule of the First Treatment Demand protocol of the Pompidou Group of the Council of Europe. Subjects were 84.6% male, with a median age of 27 years (range 14-43) and median duration of use 11 years (range 2-27). The primary substance of use was mainly heroin (88.5%). The percentage of agreement between interviews was highest for lifetime history of injecting (100%), urban residence (98.7%) and mode of use of heroin (94.3%). Percentages of agreement were close to 90% for most other items of socio-demographic data and drug use history, whether or not the correct response to the item could logically change between interviews. It is concluded that reliability of the data is around 90% and short-term behaviour in this population is rather stable. Only employment status (52.9% agreement) and secondary substance of abuse (25.5%) appeared to be exceptions. It is noteworthy that HIV and hepatitis serostatuses were reported less reliably than most other items.