Predicting Future Osteoporosis-related Fracture Risk
Using Age, Bone Mineral Density, and Spine Fracture Status to Predict Fracture Risk
They used data from 2,761 patients who participated in the Canadian Multicentre Osteoporosis Study (CaMos)—a prospective study that followed a randomly selected, population-based community cohort of non-institutionalized men and women—to determine this.
A model considering age, BMD, and spine fracture status captured almost all of the predictive information provided by other models that included additional risk factors.
In a study published by the same researchers, the investigators confirmed their findings from the previous analysis by using data from study participants who had incomplete spine radiograph assessments and who were excluded from the previous study (1,447 women and 517 men).
Their results appeared in the paper “Age, bone mineral density, and spine fracture status for predicting fracture risk.” It was presented at the 2011 Annual Meeting of the American Society of Bone and Mineral Research.
Both prevalent and incident spine fractures were evaluated by quantitative morphometry of lateral spine radiographs at baseline—and again at 5 years. An annual questionnaire was mailed to participants to record incident non-vertebral fractures with validation.
The endpoint was a 5-year risk of an osteoporotic fracture defined as a:
- morphometric vertebral fracture
- non-vertebral fragility fracture (a fracture with minimal trauma)
Results for this study’s 1,964 participants were similar to the already published results for the 2,761 patients who participated in CaMos.
A logistic regression model—which included age, femoral neck BMD, and spine fracture status (yes/no)—revealed a gradient of risk/standard deviation (GR/SD) of 1.88.
For this model, almost all of the predictive information provided by a model with spine fracture status was captured—in addition to the World Health Organization (WHO) risk factors (GR/SD: 1.92).
In addition, this particular model provided greater predictive information than a model that only considered WHO risk factors (GR/SD: 1.74).
Researchers found that fracture risk was markedly impacted by spine fracture status at all ages and BMD levels—including the younger participants and those who had a higher BMD.
They repeated all analyses for the overall population (n=4,725); they found that the results were very similar.
These data confirm that morphometric spine fracture status—along with both age and BMD—predicted future fracture risk with greater simplicity and higher predictive accuracy than consideration of more complicated models, such as the WHO risk factors. Furthermore, researchers determined that spine fracture status provides useful predictive information regarding future fracture risk. They recommend a spine fracture status assessment for patients at risk of fracture.