ABSTRACT
Background: Due to an aging population, osteoporosis has become an increasingly prevalent metabolic bone disorder that is largely undiagnosed worldwide because of inaccessible and expensive DXA machines. The Chapman bone algorithm (CBA), a mathematical treatment that enables osteoporosis determination by using simply-assayed bone metabolites from blood serum, has been previously presented as a cheaper and feasible alternative for analyzing bone health. The CBA has a sensitivity of 1.0 and a specificity of 0.83, with an area under the Receiver Operating Characteristic curve of 0.93. Our goal was to utilize existing data from primary literature sources to determine if the CBA could be applied with similar or equal fidelity.Methods: We obtained mean values from analyses of serum Osteocalcin (s-OC) and serum Pyridinoline (s-PYD) markers in conjunction with patient age from various large-sample data sets available in primary literature.Results: Following analyses of aggregated mean values from the literature, we found that 60% of studies predicted the presence or absence of osteoporosis with the same degree of accuracy between FRAX and CBA methods. Osteoporosis was defined as having a t-score of <-2.5 (FRAX) or surpassing the threshold p-value of >0.035 (CBA).Conclusions: We expected higher agreement between the FRAX scores and our CBA, but this may be due to the aggregated nature of the data. Our findings indicated the need to advance the CBA in analyzing larger-scale primary data sets, underscoring the importance of raw data analysis, to determine the full efficacy of this diagnostic tool.