Calcium and body fat in peripubertal girls: cross-sectional and longitudinal observations

The objective was to investigate whether calcium intake is independently associated with body fat in peripubertal girls.

A total of 45 healthy premenarcheal girls (initially 10.5 +/- 0.6 years of age) completed a 2-year prospective observational study. Percent body fat and trunk fat (by DXA), height, weight, maturational stage, and eating attitudes (children's Eating Attitudes Test [EAT]) were measured at baseline and at 1 and 2 years. Physical activity (by questionnaire) and calcium intake (by calcium-specific food frequency questionnaire and 3-day food records) were assessed at 6-month intervals.

Girls with 2-year mean calcium intake below and above the median had similar age, height, lean mass, and maturational stage at baseline, but girls below the median had significantly higher baseline percentage body fat (29.3 +/- 10.3% vs. 22.0 +/- 6.8%, p < 0.01) and trunk fat (24.2 +/- 10.6% vs. 15.8 +/- 6.8%, p < 0.01). However, differences were no longer significant when covariates (most notably children's EAT dieting score) were considered. Regression analysis revealed that dieting score was a consistent positive predictor of percentage body and trunk fat at all cross-sectional time-points, accounting for >20% of the variance, but did not predict 2-year change in percentage fat. Calcium intake did not enter longitudinal regression equations for 2-year change in percentage fat.

In this group of girls, an inverse cross-sectional association between calcium intake and body fat appeared to result from avoidance of foods high in calcium by girls who were concerned about their body weight or shape. Calcium intake was not associated with change in fat over time.