Efficient patient identification strategies for women with osteoporosis

The purpose of this analysis was to identify efficient (highest sensitivity at each level of cost) strategies to detect osteoporosis in postmenopausal women. Our study sample consisted of 392 women (age greater than or equal to 50 yr) who were retirees or active employees of a corporation. The Simpl e Calculated Osteoporosis Risk Estimation (SCORE(TM)) questionnaire was com pleted, and bone mineral density levels were collected at the forearm using peripheral dual X-ray absorptiometry (pDXA), and at the femoral neck and l umbar spine using central DXA. Osteoporotic women were those with a T-score of -2.5 or less at any one of the three skeletal sites tested. Assumed cos ts were $5 for SCORE, $35 for pDXA, $120 for DXA at either the hip or spine , and $200 for DXA at both the hip and spine. The analysis indicated that t he current "gold standard" is inefficient relative to other strategies inve stigated. By comparison, a tiered strategy consisting of SCORE, pDXA, and t hen selective use of DXA at both the hip and spine identified 90% of the wo men with osteoporosis at a cost of only $106 per woman tested. In choosing among the efficient strategies, decision makers must determine the extent t o which they are willing to trade off higher program cost for greater sensi tivity.