Validation and reproducibility of a calcium frequency of consumption questionnaire for 30 through 50 year-old women

Abstract

Objectives: To determine the validity and reproducibility of a food frequency questionnaire (FFQ) in order to assess calcium intake in women. Design: Longitudinal and prospective study. Location: Homes of Nutrition EAP students’ family environment in Lima. Participants: Women volunteers 30 to 50 year-old. Methods: Seventy-eight women volunteers 30 to 50 year-old were trained to keep food records (REC) for 7 days in two distinct periods four months apart. In the second period a FFQ survey with 24 food and preparations in normal portions was applied. Correlation was determined with Spearman rho and Kappa coefficient, as well as FFQ sensitivity and specificity; we present a Bland and Altman graphic. Main outcome measures: Milligrams of calcium consumed by REC and FFQ. Results: Mean calcium intake for 14 days obtained by 24 h record was 415 mg ± 130.4 mg for FFQ1 and 490 mg ± 338.3 mg for FFQ2. Kappa coefficient was 0.247, low, but significant; 98.7% and 76.9% of women had inadequate intake based on 14 records of 24 h and FFQ, respectively. Sensitivity was 77.9%, specificity 100% and positive predictive value 100%. Conclusions: Correlation coefficients obtained were low, but due to the high incidence of inadequate calcium ingestion the format employed had the ability to detect people with poor consumption.