More work needed in depression outcomes

Health state utilities are one approach that works

Writing recently on his research into depression-related outcomes, Dennis A. Revicki, PhD, concluded that future clinical trials should consider including measures of health state utilities.1

The only problem is that there just hasn’t been much done in this area, he says. Revicki, vice president for MEDTAP International in Bethesda, MD, has pioneered some work applying standard gamble techniques to derive utilities for health states in patients with major depression disorder.

His small study was with two outpatient primary care groups of patients, 40 patients from a family practice clinic in Toronto, Ontario, and 30 from a community-based primary care practice in San Diego. These were patients who had completed at least eight weeks of antidepressant treatment and were currently receiving treatment or had completed an antidepressant treatment regimen within the last two months.

Standard gamble methods were used to generate utilities for 11 depression-related hypothetical health states and for the patient’s current health state. Health status was measured using the Medical Outcomes Study Short Form 36 and the Medical Outcomes Study cognitive function scale. Clinicians rated depression severity using the 17-item Hamilton Depression Rating Scale.

Only a few patients (3%) could not complete the standard gamble interview, and 25% rated severe depression as the equivalent to or worse than death, a not unexpected finding. Revicki’s work correlated well with earlier work in assessing utilities for depression related states2 that had found that the utility for moderate, untreated depression was 0.32. His own study found a utility of 0.30 for the same stage.

Revicki’s work was a follow-up of a drug study comparing cost-efficiency and clinical effectiveness between older generation tricyclic antidepressants (TCAs) and newer serotonin selective reuptake inhibitors (SSRIs). Basically, the SSRIs have fewer side effects and are equally effective, but managed care companies balk at paying the higher prices for the newer drugs.

Revicki maintains that when patient preferences (as measured in compliance with treatment regimens) and quality-of-life measures are factored in, the SSRIs emerge as the clearer choice. But because so little work has been done on patient utilities for depressive disorders, these patients get short shrift when MCOs make formulary decisions for anti-depressants.

"Given the comparable clinical efficacy between the newer antidepressants and the TCAs, treatment decisions may need to rest on patient preferences and the effect of treatment on patient functioning and well-being, which incorporate the impact of both changes in depression severity and side effects," he said in his latest paper on the subject1.

When interviewed for Clinical Outcomes Measurement, Revicki cautions that his standard gamble approach may not be the "magic bullet" that solves this problem. It’s complicated, some patients cannot do it, and it requires an investment in training of personnel up front. "I think it would be possible to adapt [this approach], but you’d really have to do your homework and work on the implementation of it," he advises.

"The real application for this is for evaluating cost-effectiveness of various treatments, various interventions, such as packages of physicians visits and medications for psychotherapy or psychological counseling," he maintains.

Meanwhile, he worries that not enough is being done in outcomes research for depression. As has been the case with other chronic disorders, such as asthma, diabetes, and congestive heart failure, the managed care companies need to address the treatment of depression in terms of cost-effectively managing the disease, Revicki argues.

"It’s time to really look at the patient outcomes side, to try to collect data and analyze data that will help them identify the patterns of care that produce the best outcomes," he says.

While managed care companies have certainly embraced this approach for disease management outside the behavior health field, few inroads have been made for behavioral disorders. With more work like Revicki’s, perhaps that will change.

[For more information, contact Dennis A. Revicki, PhD, Center for Health Outcomes Research, MEDTAP International, 7101 Wisconsin Ave., Suite 600, Besthesda, MD 20814. Telephone: (301) 654-9729; e-mail:]


1. Journal of Affective Disorders. 1998, 48:25-36.

2. Bennett JJ, Torrance GW, Boyle MH, et al. 1995. McSad mental health state utilities: Results of a survey in major, uni-polar depression. Quality Life Res. 4:397.