Widespread misdiagnosis of depression complicates outcomes measurement

A few simple questions may keep patients from slipping through the cracks

Although depression costs Americans about $43 billion annually in lost productivity, decreased quality of life, and high mortality, evidence suggests that far too often, the condition goes undiagnosed and untreated by managed care organizations. This state of affairs complicates the efforts of disease managers to measure depression outcomes adequately, experts say.

"Poor diagnosis of depression may be the single biggest problem in the chain of clinical processes that need to take place to treat the patient," says John P. Docherty, MD, professor and vice chairman of the department of psychiatry at Cornell University Medical College in Ithaca, NY, and deputy medical director of New York Hospital’s Westchester division. Indeed, recent studies have shown that among primary care physicians, accuracy rates in diagnosing depression fall well below 50%.

"One of the biggest problems for primary care providers is that patients with depression don’t come in saying they’re sad," says Donna DePaul-Kelly, interim executive director of the Chicago-based National Depressive and Manic Depressive Association. "They say they’re tired, or they’re having trouble sleeping. They don’t have their appetites anymore. They’re saying things that, to a primary care physician, don’t add up to depression."

Many primary care physicians are simply ot well trained in diagnosing psychological maladies or in providing adequate treatment, says Herbert C. Schulberg, PhD, a professor with the University of Pittsburgh School of Medicine’s department of psychiatry. In a study conducted to assess the quality of care that primary care physicians provided to depressive patients, Schulberg found that patients were not seen often enough and that medication was not prescribed for adequate lengths of time. "In fact, even though we told the primary care physicians that their patients had a major depression, we could not identify for about 30% of them any depression-specific treatment over the next eight months," Schulberg says.

Payers discourage treatment

Meanwhile, patients treated by psychiatrists fared much better. "The findings were quite dramatic in that the two standard interventions — medication and psychotherapy — when delivered in a structured, protocol-type way, produced far superior outcomes," says Schulberg. "The moral of it is that if you treat patients with guideline standards, they will do a heck of a lot better."

One problem is that many payers set up disincentives for primary care providers to recognize and treat depression, says Robert M. A. Hirschfeld, MD, professor and chairman of the department of psychiatry and behavioral sciences at the University of Texas Medical Branch at Galveston. "Very often, physicians will not get paid for treatment of psychiatric disorders or will be compensated at a significantly lower rate," says Hirschfeld. "Those kinds of issues all conspire to reduce the recognition and treatment of depression."

The problem is compounded for patients undergoing milder episodes of depression, says Docherty. "We’re beginning to see that people who have milder depressions also have a lot of disability, a vulnerability of developing more severe depression, and generally poor long-term outcomes," he says.

Depressive patients themselves may share some of the responsibility, says Hirschfeld. Because of the feelings of hopelessness that depression engenders, many people fail to seek help. Others fail to recognize that depression is a treatable medical illness. "We have to ask ourselves, in light of the scientific breakthroughs and advances of the last couple of decades, why is this still such a big problem?" asks Hirschfeld. "Medicine has a lot of the answers, but people still aren’t getting them, and it’s really unfortunate." (See related story on ways to rectify the undertreatment of depression, p. 45.)

Largely in response to the difficulties primary care physicians have in establishing a diagnosis of depression, researchers at the Center for Outcomes Research and Effectiveness (CORE) at the University of Arkansas for Medical Sciences in Little Rock convened a panel of health experts to develop a depression outcomes module. The module is designed to "measure the types of care depressed patients receive, the outcomes of that care, and the patient characteristics that influence either the type or the outcome of that care." Similar in structure to outcomes modules for other chronic conditions, the CORE module is meant to be used for general health populations and not necessarily for psychiatric clinics, says Suzanne McCarthy, applied projects facilitator at CORE. (See excerpt from the Depression Outcomes Module, pp. 43 and 44.)

Used by managed care organizations, large employers, and consumer groups, the module includes instruments for patient screening, patient baseline assessment, clinician baseline assessment, patient follow-up assessment and medical record review. "Our mission is to improve health care in America, and the best way to do that is to generate a technology that gives providers a way of systematically looking at what are the best ways to deal with issues that were traditionally thought of as being fairly complex and intense," says McCarthy.

Initial use of the module by MCOs has yielded some surprising results. For example, one organization became puzzled by the poor outcomes of many of its patients with asthma and diabetes. When providers began using the depression screener, they found that about 30% of these patients suffered from clinical depression. "When they initiated care of the depression, the issues in and around management of the other chronic illness and the patient’s response to the illness changed dramatically," says McCarthy.

Screening tools

Initial screening tools for depression need not be elaborate, notes Docherty. Indeed, a few simple questions can serve to identify at-risk populations. "For example, ‘Have you lost pleasure in things lately? Have you lost interest? Have you been feeling depressed? If the answer is yes to any of those, you have a much higher rate of being depressed," says Docherty. On the basis of these initial responses, the physician could then administer a confirmatory module, such as CORE’s, to confirm the diagnosis.

Make sure, however, that any outcomes measurement system you use is appropriate for your patient population. "Your population has to be able to handle the instrument," Docherty says. "The reading level must be appropriate, and it must be properly translated in a culturally congruent way so that the patients are really able to use them."

Make sure as well that the system is designed to achieve the findings you’re interested in, Docherty recommends. For example, a single measure in which a patient rates his or her level of depression on a scale from one to 100 points may be adequate to your needs, as opposed to an intensive Hamilton scale, says Docherty. "It depends on what you really need the data for and how you’re going to analyze it. The economy of data collection can be very important."