Many clinicians in this managed care environment have had the unpleasant experience of receiving a telephone call from an independent, "peer review" organization. These physician and nurse groups are hired by a managed care plan to approve the "necessity" for hospitalization of a patient. Disapproval means either that the plan will not pay either provider or hospital for the stay or that both are required to submit to a lengthy, cumbersome, and usually fruitless appeals process to get the stay approved.

During such calls, the "physician reviewer" usually explains that the grounds for disapproval of the hospital admission are "standards" or "criteria," which, in their opinion, the admission failed to meet. If the provider assumes that these criteria are those contained in the Clinical Practice Recommendations of the American Diabetes Association (ADA),¹ the assumption would usually be an error. In most of these cases, the criteria are those suggested to the managed care plan by Millman and Robertson, Inc.

Millman & Robertson
This company is an organization devoted to the support of actuarial decisions regarding the allocation and utilization of resources by managed care. In fact, the company identifies itself on its letterhead as "actuarials." In short, this means that Millman & Robertson advises clients on a financial basis about how to maximize revenue and minimize cost in the management of insurance policies. In health care, the compilation of such guidelines represents the best compromise between unavoidable expenditures on health care by clinical diagnosis or situation and those that may be dispensed with at reasonable risk to the managed care plan. Risk here includes medical-legal risk and loss of good will (bad publicity). An essential aspect of these activities is that the perspectives suggested are based on the best interests of the managed care plan. Therefore, these considerations reflect the interests of the patient only to the extent that those interests cannot reasonably be avoided, either because of medical-legal limitations or considerations of financial risk.

All of these concepts and guidelines are contained in a number of compendiums published by this company. These include Inpatient and Surgical Care; Primary and Pharmaceutical Care; Pediatrics; Return-to-Work Planning; Ambulatory Care, Case Management; Home Care, Case Management; Recovery Facility Care; Workers Compensation; and The Health Care Companion for Consumers.

To read any parts of these documents, one must be a subscriber to the products of this company. Subscribers may be insurance companies, managed care plans, or other large corporations. Spokespersons for Millman & Robertson told me by telephone that providers and provider organizations may see the relevant portions of these documents when subject to review by their organization.² However, it is my experience and that of all colleagues I have surveyed that no opportunity is offered to see the guidelines when subjected to a review. In fact, when I have specifically asked to be shown the guidelines a "peer reviewer" is using, I have invariably been told that the guidelines are "proprietary."

The Guidelines for Diabetes Care
The guidelines for diabetes care are included in a number of Millman & Robertson publications. The guidelines that have affected providers the most substantially thus far are those by which the appropriateness of inpatient admissions are judged. These are in Inpatient and Surgical Care. According to Millman & Robertson, a hospital admission for uncontrolled diabetes is appropriate only when the clinical situation is life-threatening, such as coma, or the provider has no alternative means of rendering care, such as home health. Many plans have applied these guidelines to cover cases in which the blood glucose level exceeds 600 mg/dl or the patient is in coma. These guidelines are substantially different from those published by the expert committees of the ADA.² Millman & Robertson also comment on the appropriateness of the length of hospital stay for diabetes. The company is careful to point out that its recommendations for length of stay are not guidelines, but rather goals. These goals reflect the lengths of stay the managed care plan should attempt to achieve as the plan organizes outpatient resources to replace hospital stay. According to company, the length of stay for uncontrolled diabetes, when approved, should not exceed 1 day. Lest this seem extraordinary, keep in mind that the recommended length of stay for myocardial infarction is 1.5 days.

The Application of Millman & Robertson
These statements by Millman & Robertson may lend themselves to misinterpretation by providers and health organizations alike. The standards provided to managed care by this company are not clinical guidelines; they are economic goals or guidelines, as noted above. For example, it is (from the perspective of a managed care plan) a desirable goal to limit the numbers of diabetic hospital admissions and to keep the average length of stay to ~1 day. However, how the managed care plans that purchase this information choose to use it is another matter.

In the view of Millman & Robertson, Inc., managed care should use these actuarial numbers as considerations for structuring their resources and judging the results of their case management efforts. Millman & Robertson does not necessarily intend for a managed care plan to arbitrarily apply the guidelines for length of stay unless the plan has organized its outpatient resources to effectively substitute for the resources of the inpatient setting. Nor does Millman & Robertson believe that its guidelines for hospital admission should be applied by a managed care plan unless the other aspects of care have been organized. When viewed in this fashion, Millman & Robertson may be seen as a positive influence to prod all of us to focus on more efficient and reasonable application of health care resources.

Unfortunately, Millman & Robertson is also used as a basis to judge inpatient diabetes admissions by a number of large managed care plans without the recognition of the additional caveats. Every plan has the option of determining whether and how these guidelines are applied. Plans including Trigon, Wellpoint, Concentra, Intracorp, and Private Health Care Systems apply these guidelines without establishing proof that alternative resources may adequately support patient care in the outpatient setting. That causes the rejection of any inpatient stay with diabetes as a primary diagnosis if the patient has an admission serum glucose level of <600 mg/dl and is not in a diabetic coma. According to these health care plans, the patient should have been treated in a less acute environment. When denying an inpatient stay, none of these plans will release the exact criteria by which the inpatient hospital stay is being judged to the provider or hospital under judgment.

Thus, the use of Millman & Robertson by at least some large health care plans today places providers and patients in a position of assuming the vast majority of the risk for dealing with severe disruptions of glycemic control. Millman & Robertson does not prevent a provider from placing a patient in the hospital. The provider's decision is reviewed retrospectively. The application of these guidelines in repeated circumstances with individual providers would tend to discourage providers' use of the hospital for uncontrolled diabetes. Instead, providers, fearful of compiling a record in their hospitals of having many admissions disallowed, would find alternate methods of dealing with these clinical problems. Presently, there is no evidence that the alternative resources in the outpatient setting can and do support the care of such seriously ill patients. Obviously, this situation leaves providers and patients—not the managed care plan—at risk.

The Implications of Millman & Robertson
These guidelines are driving the use of hospitals by providers and patients in diabetes care far more than any pronouncement by the ADA. Their widespread use and the profound influence they have had on the patterns of hospital use for uncontrolled diabetes should cause serious consideration by outside advocacy groups for diabetic patients.

One elemental problem with the guidelines is that they are derived from the very proprietary perspective of the fiscal interests of managed care organizations. While there is nothing inherently wrong with that, there should be deep concern about the application of such guidelines to the care of patients without external input and influence by advocacy organizations such as the ADA and its expert panels.

The use of these guidelines by managed care should also be subject to some oversight. The process of using guidelines for inpatient diabetes care that were developed for the fiscal benefit of managed care to justify clinical decisions about patient care seems flawed and self-interested. The justifying assumption that there are adequate alternative resources to substitute for inpatient care does not have to be proven. Therefore, the use of "guidelines" to judge the "appropriateness" of hospital admissions for diabetes should be subject to some external review involving more parties than the managed care plan and its fiscal advisors. At the least, providers and advocacy groups for the diabetic community should be knowledgeable about Millman & Robertson and its uses.

Steven Leichter, MD, FACP, FACE, is the managing director of the Columbus Health Education & Research Foundation in Columbus, Ga., and a professor of medicine at Mercer University School of Medicine in Macon, Ga.