|
|
| |||||||||||||||||||||
Several studies, documenting higher mortality due to more serious sickness at admission,2 increased numbers of deaths after discharge,3 and reduced access to outpatient care for indigent and minority patients4 under the prospective payment system, challenge Iglehart's conclusions. At a more basic level, however, nearly all the studies assessing the effect of the prospective payment system use data from 1984 to 1986, a period before the full implementation of the system. Enacted in 1983, the prospective payment system was phased in over a four-year period, during which reimbursements were based on both a hospital-specific and a federal portion of charges. The weight of the hospital portion was decreased by 25 percent annually, and the federal portion increased proportionately, until the system was fully in place in 1987. Not only were the original reimbursements based on reported costs for the fiscal year 1981-1982, a year in which high inflationary expectations produced unprecedented growth in the costs of medical services, but they were also derived from current expenditures under the existing system of cost-based reimbursement, a system long known for bloated charges. Thus, as Iglehart notes, the early years of prospective payment were profitable for most hospitals, and healthy margins, the result of high reimbursements and cost shifting, allowed hospitals to treat uninsured patients and maintain high quality of care for Medicare patients.
As the increases in reimbursements under the prospective payment system have lagged behind the recommendations of the Prospective Payment Assessment Commission, however, operating margins have fallen. For the fiscal year 1989-1990 alone, the commission estimated that the aggregate margin under prospective payment would be equal to zero, with a majority of hospitals having negative margins5. A reduction in hospitals' ability to shift costs, largely resulting from the trend toward managed care, has further diminished operating margins. As the financial reins continue to be tightened, the likelihood of compromised access to care or quality of care will increase. Indeed, the Center for Health Policy of the American Medical Association reported that the number of new Medicare beneficiaries accepted by physicians fell by 3.4 percent in 19906. Similarly, special attention should be given to several proxies for reduced access and quality, including increases in the patient loads of public hospitals, in the rate of transfers from private to public hospitals, in delayed interventions, and in preadmission screenings. Although several factors may be responsible for these trends, lower margins undoubtedly exacerbate them. Thus, Iglehart's conclusions may be premature, since the crippled financial health of many hospitals may cause greater problems of access to care and quality of care for both Medicare beneficiaries and uninsured patients.
Scott A. Kupor, A.B.
Aki Yoshikawa, Ph.D.
Stanford University
Stanford, CA 94305
References
Iglehart writes, "To protect beneficiaries against larger increases in their out-of-pocket costs, the HCFA [Health Care Financing Administration] has encouraged doctors to become participating physicians and thereby accept assignment (defined as 80 percent of the fee-schedule amount, less any unmet deductible) as payment in full." According to the provider manual issued to me by Equicor, Medicare carrier for Tennessee, "Accepting assignment means accepting the Medicare approved amount as payment in full."
In the ordinary meaning of the language, Iglehart's statement quoted above seems to indicate to the reader that a participating physician, by accepting assignment, agrees to accept 80 percent of the fee-schedule amount as payment in full. The term "payment in full" would seem to suggest to even the most cynical reader that no further payment is due the provider. Iglehart does not mention the compelling point that not only is the provider permitted to bill the patient for the 20 percent of the fee-schedule amount, but according to the terms of assignment the provider is required by law to attempt to collect the balance of the allowed charge not paid to the provider by Medicare. Insistence on the copayment is presumed to limit overutilization by both the patient and the provider.
In the same paragraph, the author writes, "No additional charges, known as balance-billing charges, may be levied by participating physicians directly on beneficiaries." In the absence of any explanation of the responsibility of the beneficiary to pay the provider the 20 percent not paid to the provider by Medicare, the use of the term "balance billing" would seem to me likely to suggest to many readers not familiar with the jargon of Medicare that the provider is forbidden to bill for the 20 percent "balance" unpaid by Medicare. Of course, the meaning of "balance billing" here refers to billing for charges in excess of 100 percent of the fee-schedule amount approved by Medicare.
Robert G. Horn, M.D.
1916 Patterson
Nashville, TN 37203
Table 2 of his article suggests that the administrative cost of Medicare as a percentage of spending is 2.1 percent. That is the actual HCFA administrative cost. After Medicare dollars leave the HCFA, they are then administered by each state's fiscal intermediary. In Florida, that intermediary (Blue Cross and Blue Shield) spends nearly $8 billion to distribute $17 billion annually. This sum is nearly 32 percent of additional administrative costs after HCFA's share. Medicare has the highest overhead and the worst administration of any insurance program in the United States.
A directory of participating physicians, which might enhance their practice, has never been published during my 15 years of participation.
Computerization should lower the administrative costs of a participating physician. In our three-man practice, 3 1/2 of the 10 employee positions are dedicated to the collection of Medicare monies owed for our services as pulmonary physicians and intensivists.
Medicare reviews are not contemporaneous. My current annual review is for hospitalizations during 1985.
Stephen M. Kreitzer, M.D.
2919 Swann Ave.
Tampa, FL 33609
There is no evidence that HMOs are administratively more efficient than traditional indemnity insurers. Another reliable published source lists HMO overhead costs as 11.7 percent of premiums3. Although even this estimate of the overhead is below the 14 percent average of other private insurers, the figures are not directly comparable. HMOs pay no premium taxes, which add about 2.6 percent to the overhead of indemnity plans4. (The reported figures for the overhead costs of both HMOs and indemnity insurers should be adjusted upward by 2 percent to reflect interest earned on money collected as premiums before it is paid out as benefits.) Moreover, HMOs have few individual (non-group) members, whose coverage is particularly expensive to market and administer. Reported figures for overhead in staff-model and group-model HMOs are difficult to interpret, since apportioning administrative costs between program (i.e., insurance) administration and service (i.e., hospital and clinic) administration is necessarily imprecise. Audited figures for the overhead costs of HMOs participating in the Medicare Competition Demonstration Project were strikingly high -- 19 percent of premiums5.
The potential for administrative waste in managed care is well illustrated by Prudential's plan in New Jersey. The care of 110,000 members is managed by 18 nurse reviewers, 5 physician reviewers, 8 provider recruiters, 15 sales and 27 service representatives, and about 100 clerks6 -- an administrative work force comparable to the number of physicians needed for a population of this size.
Canada's single-payer national health program runs with an overhead of less than 1 percent and realizes substantial additional savings by minimizing paperwork for hospitals and doctors. Those searching for similar administrative savings through managed care, or its reincarnation as managed competition, will be sorely disappointed.
David U. Himmelstein, M.D.
Steffie Woolhandler, M.D., M.P.H.
Cambridge Hospital
Cambridge, MA 02139
References
The author states "HCFA has encouraged doctors to become participating physicians and thereby accept assignment (defined as 80 percent of the fee-schedule amount, less any unmet deductible) as payment in full." This is true only if assignment is defined as acceptance of the fee-schedule amount, 80 percent of which is paid by Medicare (minus the unmet deductible); the balance (20 percent), also known as copayment, and the annual deductible must be sought from the patient1,2. In fact, a physician's waiving of the copayment or deductible (except under special and unusual circumstances) is grounds for prosecution by the inspector general2.
Table 3 of the article contains a typographic error. The 10th entry, "Electrocardiography (12 leads) (93307)," should have read, "Echocardiography (93307)"3.
Robert Lebow, M.D.
Harrington Memorial Hospital
Southbridge, MA 01550
References
To the Editor: Drs. Himmelstein and Woolhandler are correct that Canada's scheme is less expensive to administer than the plans of America's array of third parties. Perhaps the debate over systemic reform will determine how much Americans are prepared to pay for this pluralism. Whether HMOs are more or less administratively efficient than indemnity insurers has never been carefully researched. As Himmelstein and Woolhandler point out, "apportioning administrative costs" in HMOs "is necessarily imprecise." The difference between the HMO overheads estimated by Thorpe,1 which range from 2.5 percent to 7 percent, and that testified to by the president of the largest HMO trade association -- 9.4 percent -- seems largely definitional. The size of a plan is equally important. The administrative expenses of the Kaiser Permanente Medical Care Program, the largest HMO (6.7 million members), totaled 6 percent in 1992, but they ranged from a high of 10.6 percent in the smallest of its 12 regions (Kansas City [Mo. and Kans.], 41,545 members) to a low of 4.9 percent in its second largest (Southern California, 2.3 million members).
Drs. Horn and Lebow are correct in taking me to task for misstating the effect of Medicare's assignment policy on physicians' fees. However, Dr. Kreitzer's assertion regarding the program's administrative costs is incorrect. According to the 1992 report of Medicare's trustees (three Cabinet secretaries and two public members), "The ratio of administrative expenses to Part A [hospital] benefit payments has generally fallen within the range of one to three percent," including "estimates of workloads and approved budgets for intermediaries and the HCFA"2. Regarding Part B (physicians' services and other outpatient medical expenses), the trustees said, "The ratio of administrative expenses to benefit payments has been under 5 percent in recent years and is projected to decline in future years"3. Blue Cross and Blue Shield of Florida, Medicare's administrative agent, said its administrative costs for Part B from October 1992 through February 1993 were 2.6 percent, and that its costs for Part A were 0.4 percent (McCabe P [Blue Cross and Blue Shield of Florida]: personal communication). The HCFA does, in fact, publish the Medicare Participating Physicians/Suppliers Directory every year, according to the agency.
The interpretation by Kupor and Yoshikawa of the cited studies of the prospective payment system differs greatly from that of the HCFA or the Prospective Payment Assessment Commission, which monitors Part A of Medicare for Congress. The commission said in a 1991 report4:
Until recently, the effects of PPS [prospective payment system] on quality of care had not been fully evaluated. A Rand study sponsored by the HCFA examined pre-, post-PPS outcomes of care and health status at discharge for nearly 17,000 Medicare beneficiaries. The study showed that the overall trend in improved quality of care was not interrupted by the implementation of PPS, as measured by 30-day and six-month postadmission mortality. In addition, the incidence of poor care, as measured through physician review of medical records, declined from 25 percent prior to PPS to 12 percent post-PPS. However, after the implementation of PPS, a larger proportion of Medicare beneficiaries were discharged home in an unstable condition -- 15 percent, compared with 10 percent before PPS. . . . Patients in unstable condition at discharge were one and one-half times more likely to die within 90 days of discharge than patients discharged in stable condition. However, extending the length of stay for these patients would not necessarily have reduced their mortality rate.
Berry's paper on the effects of prospective payment programs5 is actually a study of state rate-setting programs in the 1970s, well before the prospective payment system was even enacted.
John K. Iglehart
References
| |||||||||||||||||||||
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
