Relationship between patient self-ratings and physician ratings of general health, depression, and anxiety
G. Wyshak and A. J. Barsky
Department of Population and International Health, Harvard School of Public Health, Boston, Mass.
OBJECTIVE: To examine the relationship between patient self-ratings and
physician ratings of general health, depression, and anxiety and patient
and physician ratings in comparison to Diagnostic Interview Schedule (DIS)
diagnoses of depression and anxiety. DESIGN: Observational study. SETTING:
A general medical outpatient clinic. PATIENTS: Clinic attendees. METHOD:
Consecutive clinic attendees on randomly selected days completed a
self-reported screening questionnaire for hypochondriasis, composed of the
Whiteley Index and the Somatic Symptom Inventory. A random sample of
patients (N = 79), 95% of whom had scores below the cutoff for
hypochondriasis (n = 75) and 5% of whom had scores at or above the cutoff
for hypochondriasis (N = 4), returned at a later date to complete a
research battery consisting of self-reported questionnaires, structured and
semistructured interviews, and a structured interview for diagnoses of
anxiety and depression based on the DIS, which used criteria from the
Diagnostic and Statistical Manual of Mental Disorders, Revised Third
Edition. OUTCOME MEASURES: Responses to questionnaires and interviews.
RESULTS: Agreement between patients and physicians was statistically
significant concerning patients' physical health, depression, and anxiety.
Based on one question, both physicians' and patients' ratings of depression
and anxiety compared favorably with DIS diagnoses. By means of receiver
operating characteristics analysis, values for the areas under the curve
and their SEs were as follows: for depression, 0.789 +/- 0.075 for patient
self-ratings and 0.825 +/- 0.054 for physician ratings; for anxiety, 0.734
+/- 0.058 for patient self-ratings and 0.667 +/- 0.065 for physician
ratings. CONCLUSION: One simple question about a patient's status with
respect to depression or anxiety is sufficient to detect these disorders
with high sensitivity and specificity, yielding values comparable to those
yielded by instruments consisting of many items. Asking patients to rate
their own levels of depression and anxiety may constitute one portion of
the family physician's diagnostic assessment for these conditions.
