OBJECTIVES: Antidepressants are first line treatment for depression.  Effectiveness may be compromised because of discontinuation, which is commonly associated with side effects.  Using a national database of medical and pharmacy claims, we sought to identify and compare the prevalence of side effects in newly depressed adult and adolescent patients taking different classes of antidepressants.  METHODS: A new-user design was implemented using 11 years of data to identify a retrospective cohort of newly depressed subjects on antidepressant monotherapy, defined as SSRI, SNRI, TCA, MAOI, buproprion, phenylpiperazine, or tetracyclic.  Rates of side effects (per 1,000 person-months of exposure) were calculated within each antidepressant group; relative risks were calculated (SSRI as referent group).  Propensity-adjusted Cox Proportional Hazards regression was used to model the likelhood of side effects adjusted for demographic, clinical and treatment characteristics.  RESULTS:   A total of 40,017 patients had a new episode of depression and were on antidepressant monotherapy within 30 days of diagnosis [SSRI (66%), Bupropion (14%), SNRI (12%), other (8%)]. The most common side effects were headache (up to 16.8 per 1,000 person-months of therapy in adults, 17.6 per 1,000 in adolescents) and nausea (up to 7.2 per 1,000 in adults, 9.3 per 1,000 in adolescents).  Relative to adults receiving SSRIs, those receiving SNRIs had higher risk of nausea (HR=1.28, 95%CI=1.08-1.52), and of having one or more side effect of any type (HR=1.23, 95%CI=1.10-1.37).  Adults taking bupropion were less likely to have sedation (HR=0.36, 95%CI=0.16-0.79).  Adolescent receiving an SNRI were more likely to experience sedation compared to adolescents receiving an SSRI (HR=3.14, 95%CI=1.01-9.82).  CONCLUSIONS:   Side effects detected in claims must be significant enough to be reported to the provider and medically coded, so these rates are underestimates.  Nevertheless, the reported rates are nontrivial.  Future work will account for side effects in the likelihood of discontinuation and associated reduced comparative effectiveness.