The University Record, October 28, 1998
By Pete Barkey
Health System Public Relations
Health System researchers say a new study illustrates the need for fresh approaches to defining and diagnosing depression by primary care doctors. The study False Positives, False Negatives and the Validity of the Diagnosis of Major Depression in Primary Care is published in the current edition of the Archives of Family Medicine.
The research team, led by Michael Klinkman, associate professor of family medicine, studied several hundred patients in southeast Michigan who were making visits to their family doctor. What emerged from the many findings is the idea that depression is a fluid disease that needs to be tracked over time to correctly diagnose and treat it in the primary care setting.
We have not yet found the proper way to classify depressive disorders in primary care, Klinkman says. The boundaries between major depression and minor depression that have been so clearly defined in the psychiatric setting become blurred in primary care patients, where symptoms change over time in response to many influences. Distress and depression might easily be detected during one office visit and totally hidden during another.
Klinkman says the study had three major findings:
Three different groups of patients emerged in primary care practices:
1. A group that was clearly not depressed.
2. A group that was clearly depressed and impaired.
3. An in-between group. Some of these patients met screening criteria for depression and others did notbut all looked indistinguishable in their clinical characteristics. These were distressed patients with a prior history of mental health problems.
Physicians detected depression in the in-between group based on their knowledge of the patients prior mental health history.
Primary care doctors detected depression in their patients more frequently when they knew them well.
Researchers used a two-stage sampling strategy to select patients. In stage one, 1,928 adult patients were given standard diagnostic tests for major depressive disorder (MDD) while they were waiting for a scheduled visit with their family doctor. Completed questionnaires were received from 1,580 of these patients.
Immediately after the office visit, physicians completed a rating form that measured perception of patient depression, general health and stress, as well as other measures of the physicians treatment history with the patient.
For the second stage, a weighted subsample of the 1,580 patients was selected to undergo further testing by a trained social worker or masters level clinical psychologist. From this subsample, a group of 372 patients was used in the study.
Previous research at the U-M found that family physicians successfully diagnosed most severely depressed patients, but often missed those with more minimal impairment. Klinkman and his team closely examined those patients who were missedthe false-positives (patients identified by their physician as depressed who did not meet diagnostic criteria for major depression) and false-negatives (patients who met criteria for major depression who were not identified by their physician). They found that the two groups were indistinguishable in their clinical characteristics. The study showed the main indicator for how a physician diagnosed and treated depression in these patients was prior knowledge of clinical history.
In comparison to the standard structured clinical interview, family doctors using the aforementioned clinical history cues misdiagnosed the majority of depressed patientsover-diagnosing (false-positives) and under-diagnosing (false-negatives) at equal rates. Klinkman says this is because of rigid diagnostic criteria that define major depression as a symptom count during a two-week period.
One thing that seems clear from our study results, Klinkman says, is that depression in primary care is more accurately defined by symptoms and impairment over time. A patient currently being treated for depression would not be diagnosed as a case using standard criteria. This makes little sense.
Klinkman and his investigators propose a new conceptual model of diagnosing and treating depression in the primary care settingone that views major depressive disorder as a chronic condition that ebbs and flows over time.
By looking at the disease over a longer period of time, the true positives would still be identified, while many false-positives in this study would be more accurately seen as patients with MDD undergoing successful treatment. Klinkman says the remainder of the false-positive and false-negatives should be diagnosed as a distressed group who were possibly depressed, but had minimal impairment and warranted further evaluation to more accurately diagnose.
The U-M research team consisted of Klinkman, James Coyne, Susan Gallo and Thomas Schwenk. The study was funded by a grant from the National Institute of Mental Health.