Impact of Medicare payment reductions on access to surgical services

Health Services Research, Dec, 1995 by Janet B. Mitchell, Jerry Cromwell

For nearly 25 years Medicare has reimbursed physicians using the "customary, prevailing, and reasonable" (CPR) methodology, which calculates payment rates from current and historical charge patterns in local geographic areas. Beginning in January 1992, however, a fee schedule replaced this methodology, with the fees themselves based in part on the Resource-Based Relative Value Scale (RBRVS) developed by Hsiao, Braun, Dunn et al. (1988). Simulated effects of the new payment system on physicians' Medicare practice revenues have shown dramatic changes for many specialties, with particularly large declines for surgeons (Department of Health and Human Services 1991; Levy, Borowitz, Jencks, et al. 1990; Physician Payment Review Commission 1989).

Knowing how surgeons will respond to these payment reductions is of critical importance to policymakers. If surgeons respond by providing even more procedures, then policymakers will find Medicare program outlays increasing more rapidly than expected. Convinced of this response, the Health Care Financing Administration (HCFA) made its controversial "volume offset" adjustment of 50 percent (Department of Health and Human Services 1991). (HCFA assumed that physicians would provide sufficient additional services to recoup 50 percent of the revenues lost through fee reductions.) It is possible, however, that surgeons may respond to payment reductions by providing fewer procedures to their Medicare patients; if so, then policymakers must worry about access to care. Of particular concern are those Medicare beneficiaries who may be more vulnerable because of limited ability to pay. Faced with reduced surgical fees, surgeons may cut back on their poorest patients first, namely, those who are unable to pay any balance bill amount or may not even be able to pay the 20 percent coinsurance. These would include Medicare beneficiaries who are jointly eligible for Medicaid, and those Medicare beneficiaries without any private supplemental (Medigap) coverage. Medicare beneficiaries who are black or very old (85 years or more), or who live in rural areas, are significantly less likely to own private supplemental policies (Garfinkel, Bonito, and McLeroy 1987; Long, Settle, and Link 1982). These patients may already be experiencing unequal access to surgical services. Several studies, for example, have documented the lower utilization of cardiac procedures by blacks (Boutwell and Mitchell 1993; Ford, Cooper, Castaner et al. 1989; Wenneker and Epstein 1989).

The three-year phase-in of the Medicare Fee Schedule, combined with the long lag times needed to construct utilization trends from claims data, means that policymakers will not know for some time how surgeons have responded to these payment changes and whether or not access may have been compromised for some patients. Several years before adopting the Medicare Fee Schedule, however, Congress had already begun to reduce payments for selected surgical procedures. As part of the Omnibus Budget Reconciliation Act (OBRA) of 1986, Congress made an across-the-board reduction in maximum allowable payments for cataract surgery, effective January 1, 1987. Then in OBRA-87, Congress introduced a more sophisticated approach that reduced payments disproportionately more in high-fee areas, and they applied the reductions to a much larger group of surgical procedures: total hip replacement, total knee replacement, knee arthroscopy, bronchoscopy, permanent pacemaker replacement, coronary artery bypass graft (CABG) surgery, upper gastrointestinal (GI) endoscopy, transurethral prostatectomy (TURP), suprapubic prostatectomy, dilatation and curettage of the uterus (D&C), carpal tunnel release, and (again) cataract surgery. These reductions went into effect April 1, 1988. Knowing how surgeons have responded to these payment reductions will help alert policymakers to likely effects under the Medicare Fee Schedule.

Most previous studies of physician responses to fee changes have examined fee freezes or relatively small payment changes (Christensen 1992; Gabel and Rice 1985; Mitchell, Wedig, and Cromwell 1989; Rice and McCall 1984). Under OBRA-87, however, payment reductions could be as high as 17 percent, with cumulative reductions even higher for cataract surgery. Furthermore, the OBRA-87 formula resulted in a wide range in the size of the payment reductions across areas (from zero to the maximum allowed of 17 percent), providing us with a "natural" quasi-experimental design. In order to control for the underlying trends in surgical diffusion, we analyzed five years worth of data, rather than simply comparing use rates pre- and post-OBRA-87. We examined population-based utilization rates at the small-area level to determine whether surgical use changed in response to the fee reductions. Because aggregate use rates may mask changes in use for subgroups of the Medicare population, we also examined utilization rates for potentially vulnerable Medicare beneficiaries, such as blacks, to determine whether access to surgery had been impaired.