The validity of decision rules for selecting women with primary osteoporosis for bone mineral density testing.

The purpose of this study was to determine the validity of the Osteoporosis Risk Assessment Instrument (ORAI), Osteoporosis Self-Assessment Tool (OST) chart and equation, and a criterion based on body weight for identifying women with asymptomatic primary osteoporosis. Prospective recruitment and chart abstractions from family practices of three University affiliated hospitals were completed for women aged 45 years or more with baseline bone mineral density (BMD) testing results by dual energy X-ray absorptiometry. Those taking bone active medication other than hormone therapy, with prior fragility fracture or with risk factors for secondary osteoporosis were excluded. Women were categorized as being normal, osteopenic or osteoporotic by lowest BMD T-score at either the femoral neck or lumbar spine (L1-L4). Sensitivity, specificity and area under the receiver operating characteristic (ROC) curve to identify those with osteoporosis were determined for each decision rule. The positive predictive value (PPV) for detecting osteoporosis after using a second cut point to convert each decision rule into a risk index (low, moderate or high risk) was also determined. The sensitivity of the decision rules to identify women with osteoporosis ranged from 92% to 95% and specificity from 35% to 46%. The area under the ROC curves were significantly better for the ORAI (0.80), OST chart (0.82) and OST equation (0.82) compared with the body weight criterion (0.73). PPV for detecting osteoporosis ranged from 30% to 58% among women deemed at high risk. These data confirm the validity of the ORAI, the OST chart and the OST equation as screening tools for BMD testing. Further evidence is required to confirm the validity of the body weight criterion.