Reproducibility and validity of a food frequency questionnaire to estimate calcium intake in Puerto Ricans

The objective of the study was to assess the reproducibility and validity of a short semiquantitative calcium specific FFQ in a convenience Puerto Rican sample.Reproducibility was tested by the difference between calcium intakes from the FFQ completed twice, with a one month difference; validity was assessed against 6-day diet records (DR) in a sample of students, employees and faculty of the Medical Sciences Campus of the University of Puerto Rico. The statistical analyses performed were Pearson correlations, paired t-test, Bland-Altman plots, cross-classification analysis, Willett's surrogate categories, specificity, sensitivity and predictive value. A total of 93 subjects were included in the study (mean age 28.8 +/- 10.1 years, BMI 25.7 +/- 6.1 kg/m2). Mean estimated calcium intakes were 846.4 +/- 356.9 mg/day for the first FFQ, 801.1 +/- 423.7 mg/day for the repeated FFQ and 845.4 +/- 377.5 mg/day for 6-day DR. No statistical difference was observed between the means by paired t-test (p > 0.05), with significant correlations between means (p < 0.05). The FFQ was found to have a specificity of 65.2%, a sensitivity of 73.5%, a positive predictive value of 86.2% and a negative predictive value of 45.5%. In conclusion, the short FFQ is a valid tool that could be used in the future to assess calcium intake in epidemiological surveys in this group.