B2-4: A Vertebral Fracture Assessment (VFA) Performance Algorithm Improves Appropriate Utilization Among Those Referred for Bone Density Tests (DXA)

  • September 2013,
  • 138.2;
  • DOI: https://doi.org/10.3121/cmr.2013.1176.b2-4

Abstract

Background/Aims Densitometric lateral spine imaging (vertebral fracture assessment, or VFA) identifies prevalent vertebral fracture and improves fracture risk estimation, but is under-utilized in clinical practice. We created an algorithm to be used by bone density (DXA) technologists to identify those with moderate or higher pre-test probability of prevalent vertebral fracture who should have VFA at the time of a DXA test. Our objectives were to: a) assess changes of VFA utilization after implementation of the algorithm at a large rural multispecialty community health care organization and an academic health center; and b) assess the association of VFA results with prescription of fracture prevention medication.

Methods We devised a physician DXA order option that VFA also be performed for those patients whose worst T-score (spine, femoral neck, or total hip) is between −1.5 and −2.5 PLUS age 65 years or more OR height loss 1.5 inches or more OR current glucocorticoid use. The proportion for those with an indication for VFA who had one done before and after introduction of the performance algorithm was compared with chi2 statistic. Manual medical record review was done to assess results of VFA, and logistic regression was used to estimate the multivariable-adjusted association of VFA results (positive for vertebral fracture vs. negative) with subsequent physician prescribing of fracture prevention therapy.

Results After introduction of the DXA/VFA order option, 36% (331 of 969) and 87% (438 of 452) of those with a VFA indication at the two institutions had VFA performed (P-value <0.001 compared to before introduction). Those with a VFA positive for prevalent vertebral fracture had an odds ratio of 3.2 (95% C.I. 2.1 – 3.1) of starting fracture prevention medication compared to those with a VFA negative for vertebral fracture, adjusted for age, sex, prior clinical fracture, and current glucocorticoid use.

Conclusions DXA technologists can successfully use an algorithm to identify those for whom VFA is indicated, provided the appropriate conditional order for VFA is part of the DXA order. Documentation of prevalent vertebral fracture appropriately increases utilization of fracture prevention therapy in a substantial subset of those who do not otherwise have an indication for therapy.

Loading
  • Share
  • Bookmark this Article