Factors predicting osteoporosis treatment initiation in a regionally based cohort

Abstract

Osteoporosis treatment initiation was assessed during the year after baseline BMD testing in 8,689 previously untreated women. Treatment initiation increased progressively as BMD T-scores decreased, but there was a gradient response rather than step increases at conventional T-score intervention thresholds. Bone mineral density (BMD) testing is used to identify those at high fracture risk and guide osteoporosis treatment (OTx) initiation. Clinical guidelines have used the World Health Organization T-score diagnostic cutoffs as thresholds for treatment intervention. Our objective was to assess whether OTx initiation tracks these T-score cutoffs. Eight thousand six hundred and eighty-nine women age > or = 50 years who had not been dispensed any OTx medication in the year prior to baseline BMD were identified from a regionally based database in the Province of Manitoba, Canada, and OTx initiation rates were analyzed. Forty-four percent of women were dispensed OTx in the year after BMD. OTx initiation increased progressively as BMD T-scores decreased (8.2% normal, 41.0% osteopenic, 78.5% osteoporotic, p-for-trend < 0.0001). There was a gradient response to OTx initiation, rather than step increases at conventional T-score intervention thresholds. BMD was strongly associated with OTx (p < 0.0001) while age, weight, and fracture in the last year were not. Physicians rely heavily on BMD T-score to decide on OTx initiation. Although guidelines suggest using clinical risk factors to guide decision making, we did not see evidence of this. More explicit methods of reporting fracture risk may help physicians select patients who are likely to derive the largest benefit from OTx.