Evidence-Based Assessment of Pediatric Bipolar Disorder

Youngstrom, E.A. & Duax, J. (2005). Evidence-Based Assessment of Pediatric Bipolar Disorder, Part I: Base Rate and Family History. Journal of American Academic Child and Adolescent Psychiatry, 44, 712-716.

Bipolar disorder is rare in children before puberty, there is controversy about how to diagnose it, and there are few published clinical trials to guide treatment. Additionally, there is evidence that use of stimulants or antidepressants might worsen the course of illness; the compounds most likely to be effective also have the potential for serious side effects and therefore should not be prescribed unless one is confident in the diagnosis and the potential for benefit. On the flip side, there are strong concerns that untreated bipolar disorder will follow a progressive and deteorating course. Therefore, diagnosis of pediatric bipolar disorder (PBD) is both controversial and high-stakes.

This article discusses how one might implement recommendations of Evidenced-Based Practice (EBP) to gather additional information and integrate it in order to obtain greater confidence regarding further testing or treatment. A recent meta-analysis indicates that children with a first-degree relative with bipolar have a 5-fold increase in risk, and children with a second-degree relative with bipolar have a 2.5-fold risk increase. Another recent meta-analysis reveals that no other risk factors besides family history have been sufficiently documented to justify integration into clinical decision-making. Therefore, combining the child's familial information with current diagnostic base rates (which can be obtained from publications reporting rates from similar demographies, clinical settings, and interviewing techniques), and using Bayesian methods can yield probabilities that the specific individual has the diagnosis (or alternately, the frequency with which people showing that test result at that particular clinic would have the condition). If the probability yielded is below the test/no-test threshold, then no further testing may be required. If the number is above the treatment threshold, then treatment may be warranted. (Note: Often the treatment threshold is determined by the clinician in consultation with the family, weighing information about the costs and benefits of treatment.) If the number falls in between these thresholds, then further assessment may be prudent.