Increase in Bipolar Diagnosis in Youth Prompts Debates and Calls for Research.
By Arline Kaplan|December 1, 2007
Psychiatric Times. Vol. 24 No. 14
A recent analysis that found a 40-fold increase in the diagnosis of bipolar disorder in youth has highlighted diagnostic dilemmas and prompted calls for more research studies.1
Mark Olfson, MD, MPH, professor of clinical psychiatry at the New York State Psychiatric Institute and Columbia University, along with NIMH researcher Gonzalo Laje, MD, and their colleagues analyzed data from the National Ambulatory Medical Care Survey, an annual survey of office-based physicians in which physicians or their staff record details about representative patient visits during a 1-week period.
The researchers found that in 1994 and 1995, the annual number of office visits resulting in a diagnosis of bipolar disorder among people aged 19 years or younger was approximately 25 per 100,000 youth. By 2002-2003, the number had jumped to 1003 visits per 100,000 youth. At the same time, the annual number of adult visits that resulted in a diagnosis of bipolar disorder increased almost 2-fold, from 905 to 1679 visits per 100,000 adults.
Another recent report, by Blader and Carlson,2 analyzed data from the National Hospital Discharge survey and found a 5.6-fold increase between 1996 and 2004 in the number of children and adolescents aged 5 to 13 years, and a 4-fold increase in patients aged 14 to 19 years who were discharged with a diagnosis of bipolar disorder. Bipolar disorder-related discharges among children, for example, rose from 1.3 per 10,000 US children in 1996 to 7.3 per 10,000 US children by 2004.
The precise community prevalence of bipolar disorder in young people is not known, Olfson told Psychiatric Times.
"According to the National Comorbidity Survey Replication (2001-2003), however, the overall 12-month prevalence is 0.6% for bipolar I disorder and 0.8% for bipolar II disorder. The median age at onset is 18.2 years for bipolar I and 20.3 years for bipolar II, with approximately 25% of cases of each disorder beginning by age 12 years.3 The median time from onset to first treatment for individuals with bipolar disorder is 6 years, he said.4"On the basis of these numbers, one would expect substantially less than the 1% rate of annual youth bipolar visits in office-based medical practices as reported in our recent study."
The office visit study had some limitations, according to its authors. Diagnoses were based on the judgment of the treating physician rather than on an independent objective assessment, so it was not possible to determine what proportion of young people actually had the disorder. Neither did the study explain why the diagnosis had been applied to so many more children and adolescents than in years past.
Overdiagnosis—or correction of underdiagnosis?
"My sense, though," Olfson said, "is that some of the increase may be accounted for by the introduction of the bipolar II disorder diagnosis in the DSM-IV (1994) and a broadening of the conceptualization of bipolar spectrum disorders. However, the magnitude and rapidity of the increase in treatment of youth bipolar disorder, the predominance of boys [66.5%] in this study treated for bipolar disorder, their young age [mean, 12.8 years], and the substantial co-treatment for attention-deficit/hyperactivity disorder [ADHD, 32.2%]—all suggest the possibility of overdiagnosis. At the same time, a substantial increase in academic attention to bipolar disorder in young people may have also helped psychiatrists and other mental health professionals to correct historical under-recognition."
In their inpatient analysis, Blader and Carlson2 said that the higher rates of inpatient admissions among youth that are associated with bipolar disorder may reflect a shift in community diagnostic practices to include a broader spectrum of clinical symptoms, with bipolar disorder being diagnosed in children who display "impulsive, volatile, aggressive dyscontrol" children in whom other diagnoses, such as ADHD, may have been used in the past. Another reason, they said, may be "upcoding" to putatively more severe conditions for reimbursement or for administrative reasons.
A broader definition of bipolar
Whatever the reasons, Jon McClellan, MD, associate professor of psychiatry at the University of Washington and coauthor of "Practice Parameter for the Assessment and Treatment of Children and Adolescents with Bipolar Disorder" confirmed that the "[bipolar] diagnosis is being used much more often than it had been previously.
"Whether it is an overdiagnosis or an accurate diagnosis depends on how you want to define the disorder, and that is really what has changed. The definition has become much broader than it was 10 years ago," he said.
Another confounding factor, according to McClellan, is that some mental health professionals say they are using DSM-IV-TR criteria for bipolar disorder, but when one reads their studies, they have each applied the criteria in their own way.
In addition, differences in symptom presentation between some kids and adults often makes diagnosis more challenging, McClellan told Psychiatric Times.
"There is some agreement that the patterns of illness and symptoms described in children, particularly among those under 12 years of age, do not look like the same classic disorder that has been historically defined in adults," McClellan said. "But whether or not it is the same disorder remains to be seen."
"For some youth," Olfson said, "the course of manic symptoms tends to be more persistent than episodic, as it typically is in adult bipolar disorder. In addition, some youth appear to present with irritability, impulsive aggression, and oppositionality, with fewer or no euphoric symptoms. Some symptoms of mania, such as grandiosity and an increase in goal-directed activity, are less commonly reported in younger children in whom bipolar disorder is diagnosed. Hopefully, longitudinal research that follows identified youth into adulthood will help to clarify how manic symptom patterns migrate and change over time."
Co-occurring disorders also can make diagnosis more difficult, wrote NIMH director Thomas Insel, MD, in a recent Director's Update about bipolar disorder in children.6
In most studies, as many as 60% of children who received a diagnosis of bipolar disorder also have ADHD, Insel said, "raising questions about whether the current diagnostic criteria are specific enough to distinguish symptoms of bipolar disorder from symptoms of other related illnesses in children."
Manuel Mota-Castillo, MD, assistant clinical professor of psychiatry at St Mathews University in the Cayman Islands and Florida and author of a book on ADHD,7 argues that too often the labels of ADHD, oppositional defiant disorder (ODD), or conduct disorder (CD) are misdiagnoses and conceal other conditions, frequently a mood disorder.
"I have also seen children with anxiety, psychosis, and [posttraumatic stress disorder] given ODD/CD labels," he said.
Mota-Castillo urged fellow clinicians to take an in-depth look at what is happening with the patient and to specifically check family histories for bipolar disorder or schizophrenia. It is beyond any discussion that bipolar disorder is highly heritable, so why is a diagnosis of ADHD being given to children of parents with bipolar disorder or to children of parents with schizophrenia? he asked.
Another issue for Mota-Castillo is the reluctance of psychiatrists to diagnose bipolar disorder in young children. "It is not a capital sin to give a child the right diagnosis," he told Psychiatric Times. "We need to do away with the concept that young children cannot have bipolar disorder," he added, noting that several years ago he published a case series of children aged 2 to 5 years who received a diagnosis of juvenile mania and who were successfully treated with valproic acid (Depakote).8
Because of the diagnostic issues surrounding pediatric bipolar disorder, clinical psychiatrists as well as academicians and bipolar organizations are urging more research.
"The fact that the controversy over the diagnosis still exists is a good thing, as it will almost certainly drive more research into the cause, course, and treatment of children with these debilitating disorders," said James Hudziak, MD, chair of the Professional Advisory Board for the Child and Adolescent Bipolar Foundation, in a recent press statement.9
McClellan noted that research might help settle the major issue of whether there should be separate criteria established for diagnosing bipolar disorder in children.
"I think you need evidence that there actually is a discrete disorder before you start creating more new criteria," he said. "Ultimately, we need other kinds of markers like biological markers and anatomical markers to really make this work."
According to Insel, the NIMH is committed to the development of biological tests that can help validate the diagnosis of bipolar disorder in children.
"Recent research advances showed that electroencephalograms10 and magnetic resonance imaging studies of the brain11 can reveal differences between bipolar disorder and related behavioral syndromes which cause some of the same symptoms in children as bipolar disorder," he wrote in the Director's Update.
Research with MRI, according to Olfson, suggests that bipolar disorder likely involves several different brain structures, including prefrontal and subcortical regions.
"More specifically, pediatric bipolar disorder is associated with abnormalities in circuits encompassing the amygdala, striatum, and ventral prefrontal cortex," he said. "Although imaging technologies continue to improve, much more work needs to be done before neuroimaging can be used to diagnose bipolar disorder reliably in clinical practice. One of the important factors complicating identification of biomarkers is that bipolar disorder is still a fairly heterogeneous disorder. Genetic studies, for example, have proposed several candidate genes for bipolar disorder."
Research is being conducted in Europe as well that may help distinguish clinical phenotypes. Masi and colleagues12 from Italy just published results of a naturalistic study in which they addressed the clinical implications of DSM-IV subtypes for bipolar disorder in 217 referred patients, ranging in age from 8 to 18 years. Sixty-six of the patients (30.4%) were younger than 12 years. The sample included both inpatients and outpatients, and the participants were followed for 6 to 40 months (mean of 17 ± 6 months). According to the subtyping, 78 patients (35.9%) presented with a type I bipolar disorder; 97 (44.7%), type II; and 42 (19.4%), bipolar disorder not otherwise specified (NOS).
Patients with type I presented more frequently with psychotic symptoms and an elated rather than irritable mood, while those with type II had depression as the intake episode in more than half of the cases. Those with bipolar disorder NOS presented with an earlier onset of the disorder, a chronic rather than episodic course, and an irritable rather than an elated mood. Patterns of comorbidity differentiated the 3 groups, with type II having the highest rate of anxiety comorbidity, and those with bipolar disorder NOS having more frequent comorbidity with ADHD and ODD.
While research continues on the biomarkers, experts were asked what tools and techniques they currently use to help them diagnose bipolar disorder in children and adolescents.
More than just interviewing parents and looking at reports, Mota-Castillo said that he watches closely to see how the child behaves, such as whether the child is respectful or extremely dis- respectful, and he interviews the child. He also looks carefully at the patient's history.
"I am not going to diagnose as bipolar somebody who did very well until fourth grade and had all As and all of a sudden is failing in school. I will think of other problems, such as divorce or substance abuse," he said.
As aids in assessment, Mota-Castillo uses the Incomplete Sentence instrument for children and the Bipolar Spectrum Diagnostic Scale,13 which he has adapted for use in patients aged 6 to 12 years as well as teenagers, and for use with Spanish-speaking youth. The Bipolar Spectrum Diagnostic Scale, a descriptive story that captures subtle features of bipolar illness, was developed by Ronald Pies, MD, professor of psychiatry at SUNY Upstate Medical Center, in Syracuse.
For research studies, McClellan said he uses structured interviews.
"For example, we are currently doing a lithium study with kids [Collaborative Lithium Trials], where we are using the K-SADS [Kiddie Schedule for Affective Disorders and Schizophrenia]; in a prior study, I used the SCID [Structured Clinical Interview for DSM-IV]," he said. "I don't think there is any magic with any of them because I can use the same tool as someone else who believes the syndrome looks different, and we will come out with a different outcome."
At present, Olfson said, clinical psychiatrists who treat youth should consider, to the extent possible, incorporating into their assessments screening tools or short bipolar assessment scales, such as the Child Mania Rating Scale; the parent version of the Mood Disorder Questionnaire; or the Conners' Abbreviated Parent Questionnaire, which is used for assessing ADHD in children and adolescents.
Beyond screening tools, Olfson said, the evaluation of a child for bipolar disorder should involve parent reports and include an assessment of all aspects of the child's life.
"Before a diagnosis of bipolar disorder is made," he said, "psychiatrists should carefully consider not only the symptom pattern over time, but the effects of the symptoms on the child's social interactions outside of the family, and his or her relationships with peers, teachers, and other individuals that are important in the child's life."