In this issue of the Journal, Koenig and Kuchibhatla, Druss et al., and Rost et al. address the matters of expenditures, enrollment, and the use of general health services for those suffering major depressive illnesses in fee-for-service and managed care systems.

These papers paint a grim picture for individuals suffering from major depression. Rost et al. find that only a small percentage of the health care dollar spent on patients with major depression goes for depression treatment and the percentage is lower in primary care than in psychiatry, although, as Koenig and Kuchibhatla document, overall medical costs are higher for depressed individuals. While Rost et al. find only a "minuscule" portion of the health care dollar is spent on treating depression, what is actually spent in fee-for-service systems may be exaggerated, because billed charges rather than reimbursements were tabulated. For example, third-party insurers often deeply discount physician charges, paying only a percentage of the fee up to a fixed maximum dollar figure. In the study by Rost et al., insurers refused to provide actual reimbursement figures for over 20% of the study population. The $151 psychiatric charge for a largely Medicaid/Medicare population also appears to be questionable.

This is cause for great concern because under managed care, the actual dollar amounts spent on all mental illness treatment have gone down. The SAMHSA Statistics Sourcebook (R155711), quoting a Foster Higgins 1992 survey, reported that 8% of the health care dollar was being expended on mental illness and substance use coverage, down from the 11% in 1990 reported by DuPont et al. (R155712). With managed care companies taking 40%–70% (R155713) of the health care dollar for overhead, profit, and huge CEO and other executive salaries, the actual amounts spent on direct care for the mentally ill are probably more in the range of 2%–4% of the health care dollar, and those with major depression get only a part of this.

This misallocation of resources is not "market based," as is regularly asserted in both the medical and lay literature, in the sense of a "free market," because it largely excludes the choice of people who need care and those who provide it (R155714). Further, in "market-driven" situations there are corrections and modifications in the face of complaints and clear problems, and these have been largely absent from the responses of the corporate/insurance industry/managed care organization (MCO) combine currently imprisoning "the market." The economist Albert O. Hirschman argues that the ability of patients to take their business elsewhere in the market may not be enough to empower them when all MCOs and insurers act similarly. He believes patients need an effective way to voice their complaints in order to bring about appropriate responses to their needs (R155715).

Senator Alfonse D'Amato and Congressman Charlie Norwood introduced "The Patient Access to Responsible Care Act." Norwood is quoted as saying, "Far too often the bottom line in medicine is the profits of the company and not the health of the patient." D'Amato, in support of this bill, attacked the "greed" of CEOs and declared, "There is no free market in health care today" (R155716).

If there were a free market, there surely would be more investment in treating those with major depression. If managed care had generated true cost savings, it would have allowed for expansion of care and coverage. However, the ranks of the uninsured have grown. Rather than a reduction in the total cost of health care, there has been cost shifting (R155717). In a free market, public demand and/or corporate enlightenment regarding mental illness equity would lead to reductions in suffering (R155718), mortality (R155719), absenteeism, and overall health care costs.

While not addressing outcomes, the studies by Koenig and Kuchibhatla and by Druss et al. support the finding by Rogers et al. (R1557110) that outcomes for depressed patients were worse in prepaid systems. Certainly, failure of diagnosis and dumping (R15570111) cannot improve outcomes. Further, outcomes cannot be enhanced by managed care's eliminating the cross-subsidies essential for research leading to advances in patient care (R15570112).

The three studies raise a number of frightening specters our profession must address if patients suffering from one of the worst illnesses are to get the high-quality, compassionate care they require. Although the value of gatekeepers has been questioned within the managed care industry, they remain predominant in managed care systems (R15570112). Mental health carve-outs may have a vested economic interest in gatekeeper systems that underdiagnose major depression (Koenig and Kuchibhatla, Druss et al.). A further concern, with the growing awareness that individuals with major depression have high rates of medical service utilization (Koenig and Kuchibhatla, Druss et al., Rost et al.) is that HMOs will assume that high utilizers are depressed and dump them. Or, rather than dump depressed individuals directly, the HMO may give them partial but inadequate care while "informing" the patients that they are getting optimal care, abandoning them to years of unnecessary disability and pain.

Finally, in their search for truth, the authors of all three papers employed complex statistical analyses. These have become a necessary part of our science. The authors have a responsibility to make their presentation of statistical data comprehensible to clinicians who are not academics or statisticians. Recently, research and statistical techniques themselves and their validity have come under careful scrutiny (R15570113). With regard to these "gold standard approaches" Bailer has correctly stated, "We never know as much as we think we know" (R15570114).

Address reprint requests to Dr. Eist, 5705 Rossmore Dr., Bethesda, MD 20814.