OBJECTIVES: Many state Medicaid programs have implemented restrictive formulary policies designed to reduce prescription drug spending.  However, restrictions limiting access to antidepressants could disrupt therapy for patients with major depressive disorder (MDD).  This study’s objective is to examine the effect of formulary restrictions on state spending and MDD patient outcomes. METHODS: We used data on patient outcomes from 24 state Medicaid programs from 2001-2008.  Outcome variables included MDD-related hospitalizations and inpatient, pharmacy and total spending.  National data on employment and social program participation for Medicaid patients, including Social Security Disability Insurance (SSDI) and Supplemental Security Income (SSI) were collected from the American Community Survey (ACS) for 2008-2011.  We linked these data to information on prior authorization and step therapy targeting antidepressants use in Medicaid by state and year.  We used regression to identify associative effects of state policies on outcomes conditional on patient characteristics. RESULTS: Prior authorization was associated with a 5.7% increase in the annual, per-patient rate of MDD-related hospitalizations (p=0.048); prior authorization and step therapy combined were associated with a 16.6% increase (p=0.075).  Inpatient expenditures also increased for MDD patients exposed to formulary restrictions on antidepressants: expenditures were 5.8% higher (p=0.369) for patients exposed to prior authorization alone and 18.4% higher (p=0.014) for those exposed to both prior authorization and step therapy.  There was no statistically significant relationship between formulary restrictions and either pharmacy spending or total spending.  Prior authorization was associated with a 1.8% (p<0.001) decrease in labor force participation among Medicaid patients.  There was no significant association with SSI or SSDI receipt, but prior authorization was associated with increased receipt of other public assistance (p<0.001). CONCLUSIONS: Formulary restrictions on antidepressants in Medicaid are not associated with lower spending for patients with MDD, and may be associated with worse health outcomes and reduced labor force participation.