Purpose: The orthopaedic unit at a university teaching hospital hired an osteoporosis (OP) coordinator to manage a collaborative program to identify fragility fracture patients and arrange for investigation and treatment of OP, and patient education. This analysis elaborates on a previously reported evaluation of the cost-effectiveness of the coordinator in avoiding inpatient hospitalizations due to further hip fractures. We apply the expected value of perfect information (EVPI) framework to identify and propose areas of future research.

Methods: A 1-year decision analytic model was developed based on patient-level data from the first year of the program and data from the literature. The model calculates the annual incidence of a further hip fracture dependent on type of index fracture (hip, wrist, humerus, other), attribution to OP, age and gender. Referral uptake, initiation of OP treatment and compliance modified the incidence of further hip fractures in the presence of a coordinator. Deterministic and Bayesian modeling techniques were used to evaluate the cost-effectiveness and to compare implications for decision making. Furthermore, an expected value of perfect information (EVPI) analysis was performed to identify those parameters where more research would be most valuable.

Results: Deterministic cost-effectiveness analysis showed that a coordinator who manages 500 patients yearly would reduce further hip fractures from 30 to 21, a cost saving of $104,000. Bayesian analysis indicates that the probability of the coordinator being a cost-effective intervention is exceeding 80% if willingness to pay (WTP) is C$16,000/per hip fracture avoided (C$23,000/QALY) or greater. Depending on the WTP threshold, the population EVPI ranges from C$4,000 to C$90,000 for a cohort of 500 patients for one year. More research is most valuable for the treatment costs of further hip fractures (up to C$90,000 at a WTP threshold of C$15,000). The EVPI for all other parameter is below C$20,000 at any WTP threshold.

Conclusion: Employment of a coordinator to manage fragility fracture patients may reduce further hip fractures and was a dominant strategy in the deterministic analysis. The Bayesian analysis showed a high probability of cost-effectiveness for this intervention from the hospital's perspective if WTP exceeds C$16,000/per hip fracture avoided. However, if WTP is low, more research on treatment costs of further hip fractures is warranted.