Predicting Medication Response in Obsessive Compulsive Disorder

In this study, the investigators hope to study a number of variables the investigators believe may help us predict why some people respond better to some medications than others. Participants will be randomly assigned to receive one of two typical medications for OCD, clomipramine or escitalopram. Individuals who would like to participate but who have previously tried one or both of these medications may instead take a newer drug, duloxetine, and undergo the identical procedures. The factors the investigators will be studying include demographics (i.e. age, gender, age of onset of OCD), genetic markers (such as variants in genes involved in breaking down drugs in the liver (cytochrome P450 system), and genes involved in several brain chemical systems, such as serotonin), the dimensions of OCD symptoms (i.e. checking, washing, and hoarding) and cortical inhibition. Cortical inhibition will be measured transcranial magnetic stimulation and is being studied because deficits in this process may be important in the development of OCD. The investigators hypothesize that certain pretreatment clinical characteristics will correlate with poor treatment response including earlier age of onset, longer duration of illness, increased YBOCS severity and presence of significant hoarding symptoms. The investigators expect that increasing degree of deficit in CI pre-treatment will predict poor treatment response, but that increase in CI from pre- to post-treatment will correlate with a positive treatment response. Differences in genetic marker status for cytochrome P450 genes will correlate with tolerability and/or response, as well as differences in genetic marker status in SLC1A1, GRIN2B, 5HT1B and 5HT2A will correlate with response.