OBJECTIVES To determine the prevalence of self-reported depression symptoms among diabetic individuals enrolled in Gold Choice, a Medicaid managed care organization specifically for people with mental health and/or substance abuse diagnoses; and to assess the sensitivity and specificity of individuals' self-report with encounter data. METHODS The 9-item depression scale of the Patient Health Questionnaire (PHQ-9) was mailed to 454 Gold Choice members in Western New York diagnosed with diabetes; and 249 completed PHQ-9 forms were returned (55% response rate). The PHQ-9 forms were compared to primary care encounter data to determine whether the respondents had been diagnosed with depression. Descriptive and inferential statistical analysis was undertaken. RESULTS The majority (56%) of individuals in the sample screened positive for depression (PHQ-9 > or = 10), and half (49%) did not have evidence of a diagnosis in their encounter data. The percentage of those diagnosed with depression rose with increasing PHQ-9 severity levels, with 63% of individuals with the most severe depression (PHQ-9 > or = 20) having a diagnosis. This trend was statistically significant, confirmed by independent sample t-tests and chi-square tests. The sensitivity of the PHQ-9 was moderate (66%), as was the specificity (52%). CONCLUSIONS The results of this study suggest that depressive disorders may be under-recognized and under-treated amongst individuals with diabetes in the primary care setting. Half (51%) of those with PHQ-9 scores > or = 10 had depression diagnoses, suggesting poor compliance rates and/or a need for therapy reassessment.