1010 — Do VA Antidepressant Augmentation Practices Reflect the Research Evidence?

Author List:
Valenstein M (Ann Arbor VA COE/SMITREC/ U of Michigan)
McCarthy JF (Ann Arbor COE/SMITREC/ U of Michigan)
Austin KL (SMITREC)
Greden JF (U of Michigan)
Blow FC (SMITREC/ U of Michigan)

Objectives:
Depression is a common, potentially treatable disorder; however, 25-30% of depressed patients do not respond adequately to antidepressant monotherapy. Treatment guidelines recommend antidepressant augmentation for these patients, with the evidence-base for specific augmentation strategies varying substantially. Lithium augmentation has the highest level of support. However, little is known about augmentation practice patterns in real world settings. In this study, we examine antidepressant augmentation within the VA during FY 2002.

Methods:
Using data from the VA National Registry for Depression, we identified 237,819 patients with a depression diagnosis in mental health settings in FY2002 who did not have a schizophrenia, dementia, or bipolar I diagnosis and who filled an antidepressant prescription. Augmentation was defined as having overlapping supplies of an antidepressant and a specified augmenting agent (e.g., lithium, anticonvulsants other than gabapentin, antipsychotic medications, an additional antidepressant agent meeting dosing criteria, or “other agents” including T3, stimulants or buspirone). Multivariate logistic regression analyses were used to examine predictors of receiving “any” augmentation and of receiving specific augmentation strategies.

Results:
52,088 (22%) of patients received augmentation. Only 0.5% received lithium augmentation. By comparison, 4% received an anticonvulsant, 7% received an antipsychotic agent, and 11% received a second antidepressant agent. Nine percent received “other” augmenting agents. Patients who were younger (p< 0.0001) and white (p <.0001) were more likely to receive augmentation, as were those with a previous psychiatric hospitalization, concurrent PTSD, or other anxiety disorders. Patients with concurrent substance use disorders were less likely to receive augmentation. Among those receiving augmentation, African Americans were more likely to receive antipsychotics, while whites were more likely to receive lithium.

Implications:
Augmentation is common in clinical settings,suggesting providers appropriately move to augmenting agents when patients fail to respond to antidepressant monotherapy. However, providers may not be carefully considering the evidence base when selecting specific augmentation strategies. Lithium has the highest level of research support, but is used the least. The frequency and type of augmentation varies by race/ethnicity and age.

Impacts:
Efforts may be needed to promote the use of augmentation strategies with the best research evidence across subgroups of depressed patients.