US psychiatric residents' treatment of patients with bipolar disorder.

We aimed to evaluate the practice patterns of US postgraduate year (PGY) levels 3 and 4 psychiatric residents in the treatment of patients with bipolar disorder (BD) types I and II. We also aimed to determine whether confidence in prescribing mood stabilizers is associated with residents' practice patterns. The residency training directors of 182 Accreditation Council for Graduate Medical Education-accredited US psychiatric residencies were solicited for study participation via e-mail. Their chief residents were asked to forward an online survey Web link to all PGY-3 and PGY-4 residents in their program. The survey was a cross-sectional, online questionnaire, evaluating residents' treatment choices for their BD patients in their psychopharmacology clinics during the last academic year (July 1, 2009, to May 1, 2010), along with resident confidence level in using mood stabilizers. The survey Web link was distributed to 769 residents, and 177 (23.0%) responded to the survey. The percentage of residents who did not initiate treatment in the past year with mood stabilizers was 24.5% for lithium, 26.9% for lamotrigine, 36.7% for valproate, and 73.6% for carbamazepine. Confidence in initiating treatment with each of the 4 mood stabilizers was significantly associated with the number of patients treated with each mood stabilizer in the past year. Residents reported having the lowest confidence with carbamazepine followed by lamotrigine, lithium, and valproate. Experience prescribing first-line treatments for BD, such as lithium and valproate, is inadequate for many US psychiatric residents. Psychiatry residency directors should ensure that their residents have sufficient opportunity to pharmacologically treat BD patients so that these young physicians may develop confidence using an array of evidence-based treatments.