A diagnosis of depression can be complicated by concomitant illnesses or medications that mimic the symptoms of depression. Williams and colleagues researched the most effective methods of diagnosing depression and determining which symptoms are related to depression and which are related to other causes.

Major depression is defined as depressed mood or anhedonia lasting at least two weeks plus a set number of psychologic or physical symptoms. Dysthymia has fewer symptoms than major depression and is a chronic illness lasting at least two years. A condition known as depression not otherwise specified includes conditions that do not have enough symptoms (fewer than five) or are of insufficient duration to be considered major depression.

The U.S. Preventive Services Task Force recommends a case-finding approach to diagnosing depression. Patients are routinely asked to complete a depression questionnaire at routine points of contact with their physician. Another approach is to evaluate patients when their symptoms and signs raise the index of suspicion of depression. Patients who have chronic medical illness, chronic pain syndrome, stress, and unexplained clinical syndromes are up to 3.5 times more likely to have depression. The diagnosis of depression, regardless of which approach is used, requires a thorough clinical interview. The accompanying table offers suggestions for assessing major depression.

The authors' MEDLINE search for studies evaluating instruments for depression assessment yielded 11 questionnaires, six of which are depression-specific instruments. The latter consists of the Beck Depression Inventory (BDI), the Center for Epidemiologic Studies Depression Screen (CES-D), the Depression Scale (DEPS), the Geriatric Depression Scale (GDS), the Zung Self-Assessment Depression Scale (SDS), and the Single Question (SQ) test. The remaining questionnaires, the nondepression-specific tests, are the Duke Anxiety and Depression Scale (DADS), the Hopkins Symptom Check List (HSCL), the Primary Care Evaluation of Mental Disorders (PRIME-MD), the PRIME-MD Patient Health Questionnaire (PHQ), and the Symptom Driven Diagnostic System–Primary Care (SDDS–PC).

The BDI, CES-D, and SDS are used to rate depression severity and monitor response to therapy. The GDS has been tested in patients older than 60 years. Selection of a specific instrument will take into account patient characteristics, response format, and brevity, as well as the need to monitor treatment and screen for other psychiatric problems. For example, the SQ test may be appropriate if the patient needs to be screened during a routine visit or if events have occurred that would increase the likelihood of depression.

Overall, taking into account all studies utilizing these instruments, a depressed patient is 3.3 times more likely to have a positive screen than a patient who is not depressed. Two methods are recommended to determine whether depressive symptoms are due to depression or to other medical conditions. The inclusive approach “counts” all depressive symptoms as due to depression. The etiologic approach (or that taken by the Diagnostic and Statistical Manual of Mental Disorders, 4th ed.) considers whether the symptoms have been accounted for by a general medical condition; if not, they are counted toward a diagnosis of depression. Finally, it was found that physicians who recognized depression had been more likely to inquire about the patient's feelings, psychologic issues and mood, and the presence of anhedonia.

The authors conclude that after the diagnosis of depression is made, it is important to determine whether the patient accepts the diagnosis, has treatment preferences, and has ever used antidepressants (and which ones), and how many prior episodes the patient has had.