Medicare Fees and Physicians’ Medicare Service Volume

How do changes in Medicare fees affect the quantity of medical care supplied by providers?  A paper by Hadley and Reschovsky (2006) examine this question be estimating the elasticity of supply.  They find the following:

Medicare fees are positively related to both the number of beneficiaries treated (η=0.12–0.61) and service intensity (η=1.04–1.71). Physicians with apparent incentives to induce demand appear to manipulate the mix of services provided in order to increase the effective Medicare fee.


…this study’s findings argue against the perception that volume offsets render the fee schedule ineffective as a tool to limit costs. Rather, the results suggest that the volume of physicians’ services to Medicare patients is very sensitive to Medicare prices and that fee reductions will lower program costs.

The authors use data from the 2000-2001 Community Tracking Study (CTS) Physician Survey.  This survey interview physicians on their practice patterns, but is also able to tie the physician’s responses to a 5% random sample of the FFS Medicare patients they treated.  The authors measure changes in quantity as a change in relative value units (RVUs).  However, Medicare’s reimbursement in an endogenous variable.  The physicians will adjust to change quantity when prices change, but Medicare may also set prices to alter physician supply behavior.

To remedy this situation, the authors use 3 instrumental variables.  The first is the geographic practice cost index (GPCI).  The authors claim that this is endogenous because—since it is based on 1990 Census figures—likely has a true price impact and only weakly reflects cost.  The second variable is the “one-quarter” physician work GPCI .  “Actual fees in 2001 were based on only one-quarter of the variation the physician work GPCI relative to the national mean, reflecting an exogenous political decision to reduce fee variation across areas independently of the variation in the underlying cost of physician time.”  Finally, the authors use Medicaid fees as a proxy for Medicare fees.

The paper applies this methodology to arrive at the following three conclusions: 1) Medicare prices increase the supply of Medicare services, 2)  the magnitude of quantity effect depends on several market factors, and 3) physicians can manipulate the mix of services provided to increase payment per RVU.  The paper also concludes that “…the impact on access is more likely to be seen in terms of the mix and quantity of services provided to Medicare beneficiaries, rather than the inability to access physicians at all.”

Leave a Reply

Your email address will not be published. Required fields are marked *