Primary care providers encounter a large proportion of the population with depression. Yet, many primary care patients with depression remain undiagnosed and untreated. This study aims to examine depression screening patterns and the role of screening in depression diagnosis and treatment in the outpatient primary care setting. Logistic regression with propensity score weighting was used to estimate the odds of depression diagnosis and treatment under the counterfactual scenario in which patients who visited providers with lower depression screening rates had visited providers with higher screening rates instead. Results indicate that physicians appear to use depression screening selectively based on patients’ presenting symptoms. Visits by patients with depressive symptom complaints were associated with higher odds of depression screening than other visits. When visits were weighted to have similar demographic and clinical characteristics, visits to providers with higher screening rates had higher odds of diagnosis, compared to visits to providers with lower screening rates. Higher screening rates were associated with higher odds of depression diagnosis and treatment, and even modest increases in screening rates could meaningfully increase population-level rates of depression identification and treatment in primary care. Future research is needed to identify barriers to depression care and implement systematic interventions to improve services and patient outcomes.