BACKGROUND AND OBJECTIVES: Personalized nutritional counseling about calcium intake during office encounters requires rapid estimation of calcium intake. We compared the accuracy of physician estimates to a validated calcium intake measure and characterized women whose intakes were incorrectly deemed inadequate by physicians. METHODS: As part of a controlled trial of brief, office-based calcium intake counseling of women, family physicians estimated calcium intake from patients' self-reported intake of dairy food/beverage intake and from their supplement use. We compared estimates to the Short Calcium Questionnaire (SCQ), a validated 7-day dietary recall measure completed by patients. Sensitivity/specificity of physician-estimated calcium intake was estimated by comparison with the SCQ. RESULTS: For 97 women, SCQ rated 32 (33%) as inadequate, 55 (57%) as adequate, and 10 (10%) as excessive. When compared to SCQ, the sensitivity of physician-estimated calcium intake inadequacy was 97% (95% confidence interval [CI]: 94%--100%), specificity was 51% (95% CI: 41%--61%), and positive predictive value was 49% (95% CI: 39%--59%). Women with underestimated intakes were more likely to report a family history of osteoporosis and take a daily multivitamin. The major source of physician underestimation of calcium intake was underestimate of dairy product contribution.CONCLUSIONS: More accurate estimates of dairy-based calcium intake will lead to greater specificity in identifying inadequate calcium intake