Development and validation of an eight-item calcium screener to assess daily calcium intake of patients with osteoporosis in clinical practice.

Abstract

To validate a short food frequency questionnaire (screener) estimating daily average calcium intake from dietary sources to guide calcium supplementation of patients with osteoporosis in clinical practice. An eight-item calcium screener was developed based on existing literature, food consumption data and expert opinion. Convergent validity was determined by comparison with 3-day food records using mean difference, Spearman's correlation coefficients (SCC) and Bland-Altman analysis. Test-retest reliability was assessed by SCC and intraclass correlation coefficients (ICC). We calculated sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) to identify patients requiring calcium supplementation (<1200 mg dietary calcium intake/day). Fifty-two patients filled out the eight-item calcium screener and the 3-day Food record (mean age of 66.8 ± 12.9 (SD)) and 38 patients filled out the screener twice for reliability analysis (mean age of 65.8 ± 12.8 (SD)). Dietary calcium intake between the calcium screener and food records showed a strong correlation (N = 52 patients, SCC = 0.53, p ≤ 0.001) and mean difference of 21 mg (p = 0.70). Bland-Altman analysis showed agreement within 95% confidence intervals for 49/52 comparisons (94%). Test-retest reliability of the calcium screener was excellent (SCC = 0.96, p ≤ 0.001; ICC = 0.99, p ≤ 0.001). The calcium screener shows good convergent validity, reliability and feasibility to estimate daily calcium intake of patients with osteoporosis in routine clinical practice.