Study Challenges Prevailing Beliefs About Cost-Effectiveness of Integrated Treatment

It is more cost-effective for psychiatrists to provide medication and psychotherapy to depressed patients than it is to split treatment between medical doctors and other mental health care providers, according to a study begun in 1995 (Goldman et al., 1998a).

The study results, published in the April 1998 issue of Psychiatric Services, found that integrated treatment-in which psychiatrists provided both medication and psychotherapy to a group of 191 depressed patients-was less expensive than split treatment provided to a group of 1,326 depressed patients. The study, supported by United Behavioral Health, the American Psychiatric Association's Office of Research and Duke University Medical Center, did not look at the clinical effectiveness of either treatment approach.

Researchers attributed cost savings in the integrated care group to longer breaks between sessions, which resulted in a lower total number of outpatient sessions, and intensive services used over the course of the 18-month study. Study results also indicated a tendency by integrated care patients to wait longer between sessions and to remain in treatment over a longer period of time. Researchers have theorized that patients in integrated treatment might start medication sooner than their split-treatment peers. If patients in the integrated-treatment group received psychotherapy and medication during the same session, desired clinical gains could be realized with fewer visits, especially if psychiatrists spaced sessions in accordance with clinically based expectations concerning the time necessary for medications to reach therapeutic levels.

Patient data was gathered for this study from claims records kept by U.S. Behavioral Health, a national specialty managed behavioral health care company (renamed United Behavioral Health [UBH] in January 1997). Patients in the study were between the ages of 18 and 65 and had been diagnosed with a depressive disorder by their treating UBH clinician in the study's initial treatment session. Those who carried an additional diagnosis of bipolar disorder, schizophrenia or other psychotic disorder were excluded from the study, as depression in these patients was seen as a phase of that illness rather than a co-existing condition. Also excluded were patients who received treatment from a psychotherapist and psychiatrist but did not do so simultaneously.

As the study was not randomized, researchers were unable to measure the potential influence of a number of factors, ranging from patient preference to the existence or absence of psychosocial support.

Psychiatrists received $47.50 for a 20- to 25-minute medication monitoring session and $95 for a 50-minute session. In comparison, nonmedical providers (clinical psychologists, social workers and master's level clinicians) averaged $68 for a 50-minute session. Despite higher hourly rates for psychiatrists, researchers found the adjusted mean cost of outpatient services was $868 in the integrated treatment group, compared to $1,465 in the split treatment cohort. Moreover, even though the adjusted mean cost of intensive services (i.e., inpatient, residential and day treatment) was substantially higher in the integrated treatment cohort ($447 compared to $200), the adjusted mean cost for all services remained significantly lower in the integrated treatment group ($1,336 versus $1,854).

Although the study did not examine which type of treatment was more clinically effective, researchers believe these study results could be an important first step toward increasing industry-wide receptivity to the concept of integrated care, particularly if the findings are validated by subsequent research.

"We always took a different stance and believed that you should try to facilitate psychiatrists doing psychotherapy, particularly when medication was involved and particularly in the treatment of affective disorders such as depression or anxiety," said William Goldman, M.D., study co-author and senior vice president for behavioral health sciences at UBH. "We said, 'Look, most of us have been patients and why would we want to see two people when one person could take care of our needs?'"

Goldman said he and others in the mental health field remain frustrated by the continued-and in his opinion, empirically untested-influence of a "prevailing assumption in public and private organized care systems" (Goldman et al., 1998a) that split treatment is more cost-effective. "This assumption was contradicted in our analyses of more than 1,500 providers seeing patients with a diagnosis of depression in a carveout managed behavioral health care setting," according to the recently published study.

Health economist Richard Frank, Ph.D., said he would like to see these results replicated in a randomized experiment or through more sophisticated observational analyses. Nonetheless, he believes the study represents a "first generation empirical foundation" as well as a new accountability on the part of mental health care delivery systems to their clients, in addition to their shareholders. "It's nice that managed care companies are putting out data and raising questions like, 'How shall we design our systems and manage care rather than just dollars?'" said Frank, a professor of health economics in the department of health care policy at Harvard Medical School.

"With fee-for-service, no one paid attention to what we were getting for the money we were spending in terms of clinical benefits," Frank said. "We didn't measure outcomes or quality, we measured costs and let things go the way they were going. There was no external pressure particularly to worry about that issue...for insurers or providers. Mental health was handled separately and was subject to more cost-sharing than general medicine. You didn't have anyone thinking about both the clinical choice and the budget...managed care has brought those two together."

Goldman agrees and believes that carveouts, companies that specialize in mental health and substance abuse treatment, stand to play an especially important role in the discussion of these issues, since they are frequently contracted by both public and private employers and/or are subcontracted by HMOs.

For example, Goldman said, UBH has an ongoing working relationship with a variety of research institutions to share up-to-date outcome and utilization information essential to the industry's ability to project an accurate impression of its costs and services. Anachronistic data from insurance claims are neither accurate nor relevant to discussions about managed care costs. Using such data has led to dramatically inflated estimates of actual service costs and some dicey moments in federal and state parity debates (Goldman et al., 1998b).

According to Goldman, many organizations have had to make do with outdated information for a variety of reasons, such as computer systems that are not tailored to the collection and analysis of clinical data. Another problem is the incongruous patchwork of information many companies inherited during mergers and acquisitions of other managed care entities.

"We were saying it's time to dig into this data and bring the academic and research community into this because they're not making use of these data," Goldman said. "What little research was coming out was still using 1980s data from the insurance compan[ies] and it was glaringly out of touch and producing irrelevant findings because they were trying to relate a managed care world with unmanaged data."

While good outcome studies examining the comparative economic merits of integrated versus split care are rare, there is no shortage of opinions about the clinical benefits and drawbacks of integrated care.

"No one has anything to gain" from split treatment, particularly if collaborating professionals lack a solid working relationship, said Jeremy Lazarus, M.D., consultant and former chair of both the Managed Care and Ethics Committees of the American Psychiatric Association. "The patients first have nothing to gain from having to go to multiple professionals for treatment. The professionals have nothing to gain if it complicates their lives and increases the need for collaboration for which they are usually not reimbursed." Add to that substantial legal liability for psychiatrists, who may not be able to claim complete independence from collaborating therapists as a defense in malpractice or ethics charges situations (Lazarus et al., 1997).

Finally, Lazarus said, "the managed care organization has nothing to gain in terms of saving money or being able to tout better outcomes" when it comes to split treatment. "It costs them more in terms of having to case-manage patients who can easily be seen by just a psychiatrist with as good or better outcomes. Perhaps nonpsychiatrist mental health professionals might lose from not being at the front end of triage, but there are really plenty of patients to go around if we really open up access and we capitalize on the skills of all professionals."

Lazarus believes that most psychiatrists would prefer to do integrated treatment, but said managed care practices continue to pose a strong disincentive. "We have a situation where the managed care marketplace is the driver-not good scientific or professional decision-making," he said.

However, even if such issues were resolved, the shortage of psychiatrists, especially in rural pockets of the United States, may compel those in practice to concentrate exclusively on psychopharmacological treatment, according to Betsy Owens, CSW, a social worker practicing in Albany, N.Y. "From a public policy view, there just aren't the number of psychiatrists out there to meet the needs of patients who need both [medication and psychotherapy]," said Owens, who is past chair of the private practice steering committee for the National Association of Social Workers. Owens said that even in psychiatrically flush urban areas, prospective patients can expect to wait weeks or even months to get an appointment with a psychiatrist.

Like UBH, Magellan Health Services Inc., does have psychiatrists that offer integrated treatment, said Clarissa Marques, Ph.D., executive vice president for clinical and quality management. Marques said that integrated care can benefit those patients for whom complex medication regimens and/or issues with medication compliance are central treatment issues. Similarly, patients with Axis II diagnoses might benefit from the consistency of a single treater.

Still, there is no substitute for careful matching of patient needs to clinician expertise, said Marques, who added that integrated care would not work in an established interdisciplinary practice where practitioners complement each other's strengths. Said Marques, "One of the obstacles or limitations of psychiatrists providing integrated care is the assumption that all providers can provide services equally well...A lot of psychiatrists frankly are not very well trained to do psychotherapy. While managed care may be contributing to that, I don't think it's unique to the managed care environment."

References:

References

1.

Goldman W, McCulloch J, Cuffel B et al. (1998a), Outpatient utilization patterns of integrated and split psychotherapy and pharmacotherapy for depression. Psychiatr Serv 49(4):477-482.

2.

Goldman W, Sturm R, McCulloch J (1998b), How to Make Parity Work: New Alliances. Research Center on Managed Care for Psychiatric Disorders, Working Paper No. 132, pp 1-20.

3.

Lazarus J, Macbeth J, Wheeler N (1997), Divided treatment in the managed care arena: legal and ethical risks. Psychiatric Practice and Managed Care 3(2):3-10.