Journal of NeuroEngineering and Rehabilitation

month. This chart indicates that the cost of the pros- thetic was slowly amortized over time; by the end of Month 15, the cumulative Medicare episode payment for the study group was similar to that of the com- parison group, indicating that the cost of the pros- thetic was fully amortized. Discussion The literature indicates that the receipt of orthotic and prosthetic services could increase a patient ’ s mo- bility, ultimately reducing their health care utilization and increasing their quality of life. Based on this pos- sibility, this study investigated the economic impact and value of lower extremity orthoses, spinal orth- oses, and lower extremity prostheses. Propensity score matching techniques allowed for the comparison of clinically and demographically similar patients who received these services to those who did not, and thus for a determination of the economic impact of these services on the Medicare population. Because this study is based on Medicare claims data, it ex- cludes some other sources of economic value and outcomes, such as the ability for patients with prostheses to return to work or become more inde- pendent from social services. These are sources of economic impact from the societal and consumer ’ s perspective, although they are not generally relevant to the largely nonworking Medicare population and were outside the scope of the current analysis. Results indicated that over an 18-month period, patients who received lower extremity orthotics or spinal orthotics had reduced Medicare payments. Savings were in the range of $2000 for both types of orthotic services, or approximately 8% of total Medi- care health costs in the follow-up period. Beneficiaries who received lower extremity prostheses had similar total episode payments over 15 months, despite the higher cost of the prosthetic device, due to lower expenditures in other care settings. Within the lower extremity orthotics analysis, these results demonstrated lower payments to physicians, outpatient hospitals, and for Part D drugs. This may suggest overall lower morbidity or comorbidity in patients who receive the orthotic service. In addition, higher utilization of post-acute care may be an important reason why acute care hospital admissions Table 5 Spending and Utilization for 18-Month Spinal Orthotic Episode (2007 – 2010 and 2011 – 2014) 2007 – 2010 analysis 2011 – 2014 analysis update n = 6247 Matched pairs n = 34,575 Matched pairs Care setting Study Comparison Difference Study Comparison Difference Physician $7907 $7439 $468* $6291 $6570 -$279* DME $2605 $1288 $1317* $722 $621 $101* Acute Care Hospital / Other inpatient $11,373 $11,830 -$457 $5913 $6294 -$381* Long Term Care Hospital $517 $837 -$320** $190 $269 -$79* Inpatient Rehabilitation Facility (IRF) $990 $1188 -$198** $433 $341 $92* Outpatient $3786 $4120 -$334 $2734 $3294 -$559* Skilled Nursing Facility $2188 $3175 -$987* $1234 $1281 -$47* Home Health $2802 $2388 $414* $1100 $901 $199* Hospice $431 $426 $5** $234 $534 -$300* Total Part D Drug Spending – – – $4709 $5550 -$840* Total $32,598 $32,691 -$93 $23,560 $25,655 -$2094* Average number of therapy visits 14.95 12.91 2.04 6.14 2.06 4.08* Average number of fractures and falls 2.05 1.56 0.50* 0.32 0.32 0.00 Average number of inpatient admissions – – – 0.40 0.68 − 0.28* Length of Stay for inpatient admissions (days) – – – 1.84 3.53 − 1.69* Average number of emergency room admissions 1.35 1.32 0.03 0.81 1.03 − 0.23* Average number of IRF Admissions – – – 0.02 0.03 − 0.01* Length of Stay for IRF Admissions (days) 0.62 0.68 − 0.06 0.24 0.32 − 0.07* 12-Month Mortality Rate – – – 0.00 0.01 − 0.01* * Difference is significant at α = 0.05 ** The difference in spending between the study and comparison groups for IRF, LTCH, Other Inpatient and Hospice settings combined was significant at α = 0.05 Source: Dobson | DaVanzo analysis of custom cohort Standard Analytic Files (2007 – 2010 and 2011 – 2014) for Medicare beneficiaries who received O&P services from January 1, 2008 through June 30, 2009 or January 1, 2012 through June 30, 2013 (and matched comparisons), according to custom cohort database definition Dobson et al. Journal of NeuroEngineering and Rehabilitation 2018, 15 (Suppl 1):55 Page 68 of 72