The main barrier to the routine use of EBUS and EUS in staging lung cancer is the variable procedure reimbursement seen in different countries as well as inconsistent introduction into national or local guidelines.
Modifications of the claims codes could address this issue to some extent, but such modifications are not easy to implement and uptake may be variable if reimbursement is not directly tied to the new codes.
The outcome (use of a generic or not) was modelled by logistic regression, adjusted for patients' characteristics (gender, age, treatment complexity, substitution groups) and with several variables describing reimbursement incentives (deductible, co-payments) and the generics' market (prices, packaging, co-branded original, number of available generics, etc).
First, we estimated the models excluding the Medicare reimbursement variable to assess whether the coefficients of the other key variables were sensitive to its inclusion.
With respect to the relative Medicare reimbursement variable, the Hausman-Wu test suggests that it is likely endogenous in the case of cataracts, implying that relative Medicare reimbursement for surgery may be higher in areas where physicians perform more surgery because those physicians may be able to manipulate reimbursement rates to their advantage.
Second, we estimated the models limiting the sample to beneficiaries in counties that had at least ten surgical episodes in the prior time period, because the value of the reimbursement variable might be unstable if based on only a few surgical episodes in the countye.
Different health care systems, variable practice and reimbursement models have been implicated.
For continuous variables such as reimbursement amounts or reimbursement rates, we can use OLS regressions.
Of the protocol-related variables, higher subject reimbursement and study site were statistically significant predictors of participation at a subsequent visit.
We constructed a variable measuring relative Medicare reimbursement for physicians' services as the ratio of average physician reimbursements for all surgical episodes to average physician reimbursements for all non-surgical episodes in the county in the two prior years (2004 and 2005).
