Researchers have developed a new tool that could help predict the likelihood of developing bipolar disorder in children whose parents also have the disorder.
The tool, while available online, comes with a note of caution, however, and might not be suitable for clinical management.
Danella Hafeman, MD, PhD, assistant professor of psychiatry and attending physician in the Child and Adolescent Bipolar Services Outpatient Clinic at the Western Psychiatric Institute and Clinic at the University of Pittsburgh Medical Center, in Pennsylvania, worked on the study team and says children of parents with bipolar disorder are at higher risk to develop the disorder but only a small number will. The team’s risk calculator predicts with good accuracy who will develop bipolar disorder within five years versus who will not —which she says can be helpful for clinical and research purposes.
“We developed this risk calculator based on a previously published meta-analysis of ‘prodromal’ characteristics of bipolar disorder. While this research is still preliminary, this can provide a tool for clinicians to better evaluate prognosis and perhaps identify those who would benefit from increased monitoring and/or psychotherapy,” Hafeman says.
The risk-based calculator was designed to assess how likely a child of a parent with bipolar disorder was of developing the disorder themselves. Researchers evaluated more than 400 children aged 6 to 17 years who were participating in the Pittsburgh Bipolar Offspring Study, an ongoing community-based investigation of children of parents with bipolar I or II. Recruitment occurred from 2001 to 2007, but the nine-year follow-up portion of the investigation is ongoing.
The study identified a total of 412 at-risk children and 54 developed pediatric bipolar spectrum disorder (BPSD) during the follow-up period—18 developed bipolar disorder I or II. The research team found that individuals with more symptoms including anxiety, mood lability, depressive and manic symptoms, lower general psychosocial functioning, and whose parent experienced onset of a mood disorder at a younger age had the greatest risk of developing new onset BPSD.
Of all the children studied in the cohort across 1058 follow-up visits, 10% were “converting,” a term used to describe the period preceding the onset of BPSD, and 6.3% converted within 5 years. The mean age of participants at their study visit was 12 years, and the mean age at new onset of BPSD was 14 years.
The research team also found that symptoms could precede the onset of bipolar disorder by 2 to 10 years.
“Eventually, we hope this risk calculator can be used to identify ultra-high risk children and adolescents earlier, so that evidence-based early interventions can be used to perhaps delay or even prevent the onset of bipolar disorder,” Hafeman says. “These early interventions might include the use of particular psychotherapies that are being developed for this population. The risk calculator might also inform psychopharmacologic intervention, though more research is necessary to better understand how risk score might impact response to different classes of medication.”
While the risk calculator is available online and can be used at any point, Hafeman says she recommends that it only be used by clinicians trained in the assessment of mood disorders.
“The tool in its current form should only be used by clinicians with training to ascertain mood symptoms using brief instruments. This risk calculator was developed in a community-based cohort of individuals at familial risk for bipolar disorder, so it is immediately applicable to anyone with this family history, regardless of the clinical presentation,” Hafeman says. “In our cohort, parents underwent extensive clinical interviews to make a clear diagnosis of bipolar disorder; thus this risk calculator is also only valid in individuals whose parent meets full criteria for this diagnosis.”
For a pediatrician, the risk score might provide information regarding which patients should be referred to a specialist, and also the frequency of monitoring, Hafeman notes.
In an invited commentary published alongside the report, Esther Mesman, PhD, of University Medical Center in Utrecht, and Manon H.J. Hillegers, MD, of Erasmus University Medical Center in Rotterdam, both in the Netherlands, discussed several limitations of the tool.
“Clinicians may argue that the indicated anxiety and mood risk symptoms are evaluated routinely on a case-by-case basis and thus think that the added value of the risk calculator is limited,” the commentary notes.
The commentary suggests that because the cohort was selected without consideration of symptoms in children, future studies may seek to validate the results in a clinical setting. The authors also note that the tool might be better used to consider other mood disorders, as well.
“It is pertinent to keep in mind that, although offspring of parents affected with bipolar disorder have an elevated risk for BPSD, the risk for developing unipolar mood and other disorders is much higher (>50%),” the commentary authors write. “Therefore, while screening for BPSD is important, the focus on strategies for early identification of other psychopathology, especially depression, is essential.”
For clinicians, it is also necessary to realize that this risk calculator may overestimate or underestimate the risk in clinical, community, or minority populations, according to the commentary.
The commentary authors agree that the calculator may be more beneficial in offering information for families rather than in guiding clinical decision making. The commentary also offers a warning about the use of the tool, similar to what was noted in the study.
“Clinicians may argue that the indicated anxiety and mood risk symptoms are evaluated routinely on a case-by-case basis and thus think that the added value of the risk calculator is limited,” the commentary notes. “Regarding the online availability of the risk calculator, some caution is needed. We believe that the use of the risk calculator should be guided by an experienced clinician, and agreement about the presence of symptoms is needed before entering data in the tool. For instance, self-rated hypomania symptoms can be typical in adolescence; therefore, clinical interpretation within the context of functioning and other psychopathology is crucial.”