Validation of the OSTA index for discriminating between high and low probability of femoral neck and lumbar spine osteoporosis among Thai postmenopausal women

Abstract

Rationale: Bone mineral density (BMD) measurement of postmenopausal women is needed to diagnose osteoporosis. This is an expensive procedure and available in only a few hospitals in Thailand. The Osteoporosis Self-Assessment Tool for Asians (OSTA) index, based on age and weight, has been developed for screening of postmenopausal Asian women to identify women for whom there is little advantage in undergoing bone densitometry. This index was developed for prediction of neck of the femur osteoporosis and did not include spinal osteoporosis, which is also common. Furthermore, the index was based on data from a sample of mainly ethnic Chinese. There is evidence that the BMD of Thai women is significantly higher for the same age and weight than that of women of several other Asian ethnics, thus there is a need to validate the appropriateness of the OSTA index for both femur and spinal osteoporosis among the Thai population.

Objective: To determine the properties of the OSTA index as a screening tool among postmenopausal women in southern Thailand.

Patients: Three-hundred-and-eighty-eight postmenopausal women, with no history of disease or use of drugs associated with secondary osteoporosis and no history of treatment for osteoporosis, attending the postmenopausal clinic or internal medicine out-patient department of Songklanagarind Hospital, a teaching hospital in southern Thailand, between November 2000 and April 2002.

Method: BMD t-scores of the neck of the femur and lumbar spine were collected retrospectively and a diagnosis of osteoporosis made according to WHO criteria. Sensitivity and specificity and their 95% confidence limits were calculatedfor the dichotomized OSTA index.

Results: Thirty-one percent of the women were detected as having osteoporosis, comprising neck of the femur (12 percent) and lumbar spine (31 percent). Twenty nine women (7.5 percent) had evidence of spinal fracture. The OSTA index at the standard cut-point of < or = -1 had a sensitivity and specificity of 0.93 (95% CI: 0.82 - 0.99) and 0.61 (95% CI: 0.56-0.66) respectively for neck of the femur but only 0.80 (95% CI: 0.72-0.87) and 0.70 (95% CI: 0.64-0.75) respectively for lumbar spine. Raising the cut-point to < or = 0 reduced the high false negative rate (0.20) in prediction of lumbar spine osteoporosis to 0.07 (95% CI: 0.03-0.13) and identified 27 percent (95% CI: 23-32 percent) of all women at low risk of osteoporosis at either site. These women may not need to undergo BMD measurement.

Conclusion: The standard cut-point of the OSTA index could identify most Thai postmenopausal women with osteoporosis of the neck of the femur. However, to improve the sensitivity of detection of osteoporosis of the lumbar spine, the cut-point of < or = 0 may be more appropriate.