The total intra-individual variation of fasting urinary calcium- and hydroxyproline-creatinine ratios were determined both as outpatients and as inpatients in 46 patients with metabolic bone disorders, including 20 women with osteoporosis. The intra-individual variation of both measurements was reduced considerably by collecting samples in the ward rather than in the outpatient setting, although this was not the case for a sub-group of osteoporotic women. The large intra-individual variation observed for calcium and hydroxyproline (30-40% in osteoporotic women) can give rise to misclassification in screening for fast bone losers. The very large critical difference required (109% and 94%, respectively, for calcium and hydroxyproline) for two results to reflect a significant change, is a major limitation in the use of these analytes for monitoring bone resorption in menopausal and osteoporotic women.