Identification of Factors Influencing the Intervention Thresholds for Treatment of Osteoporosis Based on 10-Year Absolute Fracture Risks

Abstract

There is an emerging consensus among practitioners and researchers in the field of osteoporosis that an expression of fracture risk using absolute risk estimation is preferable to risk categorization based on bone mineral density (BMD) alone (1e4). In 1994, the World Health Organization (WHO) provided an operational definition for osteoporosis in postmenopausal Caucasian women as a T-score less than or equal to 2.5 (5). Subsequent clinical practice guidelines incorporated the WHO diagnostic category as the pivotal variable in the decisionmaking process for individual patients (6e10). However, it is now recognized that BMD is only one of a multitude of risk factors for fracture, and clinical guidelines have tried in various ways to incorporate these other factors in the assessment of fracture risk. Recent discussions in the literature have also promoted a fundamental shift to the use of absolute fracture risk determination, rather than risk categories as conferred by the WHO T-score categorization (11e16). The reference work in this field is based on Sweden population register of fractures. In 2001, a mathematical model was devised by Kanis et al, which combined pooled estimates of relative risk from a meta-analysis (17) with U.S. normative BMD data and Swedish fracture incidence records to provide absolute risk estimates (1). Other studies have estimated 10-yr or lifetime absolute risk (probability) of fracture in different populations using different mathematical modeling or statistical approaches (18e21). We also have estimated 10-yr probability of fracture for the participants of East Anglian branch of the European Prospective