OBJECTIVE: To examine factors associated with diagnosis of depression at a primary care visit for reasons unrelated to depression. METHODS: We used the 1998-2000 National Ambulatory Medical Care Survey for office-based physician visits. We included visits to their primary care physician for patients with no indication of prior episodes of depression. We excluded patients under 15 years old and visits where the major reason for the visit was for mental health or social problems. We created a multivariate logistic regression model to examine which factors were associated with a depression diagnosis. Study variables included: age, race, sex, geographic region, urban setting, payment source, time spent with physician, new patient, HMO status, capitated visit, and major reason for the visit. RESULTS: There were 18,612 patients meeting study criteria, of whom 11,365 (61%) were female, 2,037 (11%) black, 1,940 (10%) were under age 25 and 5,629 (30%) were at least 65 years old. A total of 234(1.5%) patients received a depression diagnosis. Multivariate analysis showed that younger (age 15-24; OR = 0.486, p = 0.0177) and older patients (age 65 +; OR = 0.517, p <0.0001) were less likely to receive a depression diagnosis. Factors associated with increased likelihood of depression diagnosis: female (OR = 1.81, p <0.0001), self-pay (OR = 1.64, p = 0.329), and major reason for visit a chronic problem, both routine (OR = 2.24, p <0.0001) and flareup (OR = 1.58, p = 0.0349). There were non-significant trends towards reduced rate of diagnosis in blacks (OR = 0.645, p = 0.0811) and visits related to surgery/injury(OR = 0.343, p = .0620), and towards higher rates in the West(OR = 1.38, p = 0.0847). There was no association between diagnosis of depression and urban setting, new patient, capitated visit, HMO enrollment. Association of time spent with the physician and depression diagnosis was marginal, though statistically significant. CONCLUSIONS: When seeing their primary care physician for reasons unrelated to mental health or social problems, patients who were age 25-64, female, self-pay or visiting for a chronic illness were substantially more likely to be diagnosed with depression.