Quantile Regression Predicting Annual Physical Therapy Expendituresa
↵a The reference categories, which are reflected in the intercept term, are: age 65–74 y, female, non–end-stage renal disease, not originally entitled to Medicare by disability, non-Medicaid, osteoarthritis is the primary reason for therapy, and mobility scale score of 71–100 (highest mobility). The quantile regression predicts the 70th percentile of the per beneficiary annual physical therapy expenditure distribution. The primary reasons for therapy groups are mutually exclusive and exhaustive; each sample member is assigned to one and only one group. Expenditures are Medicare allowed charges. This table analyzes a convenience sample (N=4,210) of community-residing Medicare beneficiaries who presented for physical therapy at a provider participating in data collection during a specified 12-month period. *=significant at 10% level, **=significant at 5% level, ***=significant at 1% level. Source: authors' analysis of 2011 Medicare claims and enrollment and Developing Outpatient Payment Alternatives (DOTPA) data.