Journal of NeuroEngineering and Rehabilitation

that O&P services can prevent falls, reduce downstream clinical manifestations such as the development of dia- betes, and lead to long-term savings in health care spend- ing, patients can face significant barriers to access. Varying cost pressures caused Medicare prosthetic payments to decline by 6% between 2010 and 2014, and Medicare beneficiary access to more advanced prosthetics declined even more steeply, by approximately 36% over that same period [ 22 ] . In 2015, Medicare contractors proposed a new Local Coverage Determination (LCD) which would have further restricted access to more advanced devices, asserting, for example, that any Medicare beneficiary who had received a walker, wheel chair, crutches or cane would be automatically excluded from eligibility for more ad- vanced devices. This proposed LCD prompted such con- troversy that the entire matter was referred to study, which has continued for nearly two years without any published conclusions. In the interim, the RAND Corpor- ation has issued a new report underscoring the economic value of advanced technologies for amputees [ 23 ] . Our study suggest that lower extremity orthoses, spinal orthoses and lower extremity prostheses have the potential to increase quality of life and reduce facility-based care for applicable Medicare beneficiar- ies. Similarly, these results suggest that orthotic and prosthetic services provide value to the Medicare program, as well as to the patient. In orthotics, there is a clear savings margin for the treated study group patients. In prosthetics, the cost of the services, including the higher initial cost of the prosthesis itself, is completely amortized through reduced acute care hospitalizations and facility-based care. One clinical example of this is the situation where micro- processor knees have been shown to improve patient safety in patients with transfemoral amputation by reducing stumble and fall events [ 11 ] . Limitations One limitation of the methodology was reliance on administrative data as opposed to clinical data recorded in the medical records. While the dataset included all fee-for-service health care utilization and payments, more detailed clinical indicators, such as functional status, were not available from the administrative data. Propensity score matching relied on all recorded patient demographic and clinical characteristics in an attempt to control for observ- able selection bias among those who received orthotic/prosthetic services compared to those who did not. More medical information could perhaps improve the selection of matched pairs. Another limitation of the claims data was the lack of Medicare Advantage discharges and Medicaid long term care-related expenses for dually eligible patients. The relationship of the Medicare to Medicaid payment sys- tems is problematic for analyses that involve episodes of care, as the exclusion of Medicaid claims for dually eli- gible patients prohibits identification of patients who re- ceive care in long-term care facilities as compared to the community. With additional data, reduction in long-term care facility use may have been determined to be another important outcome variable for the study group. Conclusion The results of this study generally echo those of the prior study, with some fluctuation in the cost difference between the study and comparison groups in specific subcategories of expenditures. Study group patients receiving lower extremity and spinal orthoses had significantly lower total episode spending than did the non-treated beneficiaries in the comparison group, despite having more therapy visits. Study group patients receiving lower extremity prostheses had average Medicare payments across all care settings that were slightly lower than the comparison group and the prosthetic cost was fully amortized within 15 months due to a reduction of care in other settings. Among other iden- tified benefits to prosthetic use, prosthesis users had a significantly lower hospitalization rate than comparison group patients further resulting in lower Medicare payments for acute care hospitalizations. Across all analyses, the results cumulatively suggest that orthotic and prosthetic services provide value to the Medicare program, and potentially to other payers, as well as to the patient. Endnotes 1 Codes used to identify the etiological diagnoses of interest for the orthotic and prosthetic models are in- cluded in a separate technical methodology available from the authors. Abbreviations ACA: Affordable Care Act; CMS: Centers for Medicare & Medicaid Services; IRF: Inpatient Rehabilitation Facility; LCD: Local Coverage Determination; O&P: Orthotics and prosthetics; RCT: Randomized controlled trial Acknowledgements The authors wish to acknowledge Tom Fise of the American Orthotics & Prosthetics association for his contribution to the design of the study and development of this manuscript. Funding The publication cost of this article was funded by the American Orthopedic & Prosthetic Association. Availability of data and materials The datasets generated and/or analyzed during the current study are not publicly available because they contain research-identifiable information. However, the data are available from CMS upon acceptance of a data use agreement. About this supplement This article has been published as part of Journal of NeuroEngineering and Rehabilitation Volume 15 Supplement 1, 2018: Advancements in Prosthetics Dobson et al. Journal of NeuroEngineering and Rehabilitation 2018, 15 (Suppl 1):55 Page 71 of 72