In the TEAM (Treatment of Early Age Mania) research study,  279  patients with a diagnosis of bipolar I disorder ranging in age from six years to 15 years were given one of three treatments: risperidone, lithium or divalproex.  The study lasted for eight weeks.  Raters of improvement were blind to the medication taken by the subjects.  Subjects given risperidone improved more often than subjects given lithium or divalproex.  The study was done at five university medical centers in 5 different cities.

The second of two papers reported by the study investigated what might statistically predict or explain patients’ response to the medications. The best predictor of response to medication was the “study site” – that is the location of the university medical center where the medication was given.  This was not a minor statistical artifact, but was the central finding of the study.  “The main finding was that study site moderated the treatment effect.…Site was also a strong predictor of response regardless of the specific treatment assignment ranging from 15.3% at utmb./utsw [Houston,Galveston] to 76.9% at Pitt [Pittsburgh]” (Vitiello, et al., p. 876).

The study did not intend that the location where the medication was given would determine the response to the medication.  The intention was to study the effect of the medications.  A reasonable expectation is that the medication will have a similar effect regardless of where it is given.  The table below is an excerpt from the paper.  It provides a concrete example of the large differences in medication responses of the research subjects based on the locations of the study. 

Response  by Study Site

Site                               Lithium                        Risperidone         Divalproex

                                           %                        %                          %

Pittsburgh                           73                         86.7                      66.7

Wash. DC                           59.1                      83.3                      54.6

St. Louis                             26.3                      68.8                      7.1

Houston, Galveston              5.3                       43.8                       4.2

Baltimore                            13.3                      55.6                      17.7

Total                                   35.6                      68.5                      24.0

Explanations for the different rates of response by location were sought. The effect of location on outcome was greater than the effect of which medication the patient took.  The patients at the different locations differed on a number of measures: sex, race, family history of bipolar disorder, severity of mania, severity of ADHD, source of referral, and level of maternal warmth.  Of all of the variables examined, the only variable that significantly predicted outcome to the medications was severity of ADHD.

The TEAM investigators speculate that variables other than the pharmacologic treatment prescribed “can have a powerful influence on outcome”(Vitiello, et al., 2012, p. 877).  Amongst those suggested are “organization of the clinical setting… therapeutic alliance…[and] other factors within the family, school or community”(Vitiello, et al., 2012,p. 877) . 

Other hypotheses, such as differences by location in the validity of the bipolar 1 disorder diagnosis, have been suggested; in a letter to the editor that followed the publication of the Vitiello, et al., study, Blader (2013) challenged the bipolar 1 diagnosis, highlighted the ADHD severity finding, and suggested for future studies that the use of video would be helpful in documenting the appearance  of bipolar-diagnosed patients in that their behavior to many seems different from the clinical descriptions of bipolar disorder found in DSM IV. 

The TEAM authors at the end of a reply to Blader’s letter to the editor, speculate that follow-up of the patients in the study would help understand whether these patients develop into classic adult bipolar 1 patients (Vitiello, et al., 2013).

The TEAM authors seemed to have overlooked that they initially vouched for the patients having bipolar I as a condition of enrolling in the study (Vitiello, et al., 2013).

A post about the research subjects in the TEAM study can be found here.  A post that describes the  presentation of the TEAM study at a scientific meeting can be found here.


Vitiello, B., et al.  Treatment Moderators and Predictors of Outcome in the Treatment of Early Age Mania (TEAM) Study.  J. Am Acad Child Adolesc Psychiatry.2012, 51: 867-878.

Blader, J.  Another Look at the Treatment of Early –Age Mania (TEAM) Trial.  J Am Acad Child Adolesc. Psychiatry. 2013, 52:205-206.

Vitiello, B., et al., Reply to Blader. .  J Am Acad Child Adolesc. Psychiatry. 2013, 52:206-207.

Copyright, Stuart L. Kaplan, M.D.  Dr. Kaplan is the author of Your Child Does Not Have Bipolar Disorder: How Bad Science and Good Public Relations Created the Disorder

You are reading

Your Child Does Not Have Bipolar Disorder