Although posttraumatic stress disorder (PTSD) is relatively common in community epidemiologic surveys (5-6% for men, 10-12% for women), and psychiatric patients with PTSD are known to have poor functioning and high levels of psychiatric comorbidity, there are no studies that address PTSD prevalence, functioning, and burden in primary care settings. This article reports on (1) the prevalence of PTSD using Diagnostic and Statistical Manual of Mental Disorders-Fourth Edition diagnostic criteria in Veterans Affairs (VA) primary care settings, (2) associated sociodemographic characteristics and comorbidities, (3) functional status related to PTSD, (4) the extent to which PTSD was recognized by providers and (5) health services use patterns (including specialty mental health) of PTSD patients. Patients were randomly selected from those who had an outpatient visit in FY 1999 at one of four VA hospitals; 888 patients consented (74.1% of 1198 contacted); 746 patients (84.0% of consenting patients; 62.3% of contacted patients) were reached for telephone diagnostic interviews. Diagnostic interviews with the Clinician Administered PTSD Scale yielded estimates of current PTSD prevalence of 11.5%. At statistically significant levels, PTSD was positively associated with a variety of comorbid psychiatric disorders, war zone service, age <65 years, not working, less formal education and decreased functioning. Of patients diagnosed with PTSD by study procedures, 12-month medical record review indicated that providers identified only 46.5% and only 47.7% had used mental health specialty services. PTSD-positive [PTSD(+)] patients who used mental health care in the past 12 months were more apt to be identified as having PTSD than nonmental health service users (78.0% vs. 17.8%). Although PTSD(+) patients had more medical record diagnoses than PTSD-negative [PTSD(-)] patients (6.28 vs. 4.95), their use of primary care, urgent care and inpatient care was not different from PTSD(-) patients.