Purpose: Variability exists among child and adolescent psychiatrists (CAPs) in how they diagnose juvenile Bipolar Disorder (JBD). We aimed to characterize this variability by surveying CAPs across the United States. Methods: Participants were chosen from the membership of five regions of the American Academy of Child and Adolescent Psychiatry using a systematic random sampling method. We sent approach letters and followed up by phone. Of the 100 CAPs contacted, 53 participated. The survey included questions on demographic characteristics, training, and practice settings. We asked CAPs to name 10 symptoms indicative of JBD, to rank the symptoms from 1-10, and to rate the symptoms as to how diagnostic they were. We re-categorized symptoms into Diagnostic and Statistical Manual of Mental Disorders (DSM) and non-DSM criteria and created an algorithm to determine if CAPs used DSM criteria. Results: Twenty-five (47.2%) participants were female and 28 (52.8%) were male. There were 11 (20.8%) from New England, 13 (24.5%) from Southern California, 11 (20.8%) from New York, 11 (20.8%) from St. Louis and 7 (13.2%) from Texas. On average, participants completed medical school in 1985 (73.6% attended in the U.S.), general psychiatry residency in 1991 and child residency in 1992. Participants often worked in a combination of settings, with 60.4% doing some work in private practice and 47.2% in community mental health. Most had an academic affiliation (79.2%). The most common types of Continuing Medical Education were grand rounds and conferences at local facilities (34%). Most considered themselves adequately (40%) or well (51%) trained to diagnose JBD. Symptoms most frequently cited as leading to a bipolar diagnosis were grandiosity (32), irritability (32), pressured speech (29), decreased need for sleep (23), sexual preoccupation (22) and sleep difficulty (22). Some were considered “practically diagnostic” (decreased need for sleep=4.4 on a 1-5 scale), while others received moderate ratings (irritability=3.1). Only 20 (37.7%) CAPs reported sufficient symptoms to meet DSM bipolar diagnostic criteria. This varied by region ranging from 9.1-53.8%. Experts (CAPs practicing for at least ten years) were less likely to generate DSM criteria than subexperts (21.9% versus 61.9%). Conclusion: Preliminary analyses show that less than half of CAPs use DSM criteria for diagnosing JBD. Further analyses will examine regional trends. Further studies are needed to better understand CAPs' diagnostic decision making.