Fear of harm, a possible phenotype of pediatric bipolar disorder: a dimensional approach to diagnosis for genotyping psychiatric syndromes.

BACKGROUND: In a prior concordance study of affected sibling pairs with a community diagnosis of pediatric bipolar disorder (PBD) a behavioral phenotype termed Fear of Harm (FOH) was found to have one of the strongest concordance coefficients (rho) between probands and siblings, and the widest contrasts between the rho-estimates for the proband/sibling vs. proband/comparison pairs [Papolos, D., Hennen, J., Cockerham, M.S, Lachman, H., 2007]. A strategy for identifying phenotypic subtypes: concordance of symptom dimensions between sibling pairs who met screening criteria for a genetic linkage study of childhood-onset bipolar disorder using the Child Bipolar Questionnaire (CBQ) was employed. J. Affect. Disord. 99, 27-36.]. We used the Child Bipolar Questionnaire (OUT) (CBQ) to further elucidate this behavioral phenotype of PBD. We hypothesized that selective factors including parent reported symptoms of mania and depression, would be distinguishing features of impairment between groups defined by 1) the magnitude of their score on a continuous measure of FOH, and 2) the high FOH group would have significantly greater levels of severity on course of illness variables. These measures included earlier age of onset of first psychiatric symptoms, first hospitalization, and frequency of psychiatric hospitalizations, as well as, degree of social impairment as determined by exposure to the juvenile justice system and school performance problems.

METHODS: The sample was comprised of children with community diagnoses of bipolar disorder or at risk for the illness based on enriched family history with multiple first degree relatives diagnosed with BPD (N=5335). Included were all subjects who had >40 positively endorsed CBQ symptom items at frequencies of very often, almost always, and always. This group was divided randomly into two groups, the exploratory group (N=2668) and the hypothesis testing (study) group (N=2666). The exploratory group was used for the development of hypotheses and the study group was used to test these hypotheses on a new set of data. All results reported here derive from the latter group. In subsequent analyses, we classified each child as having a high degree of FOH, low FOH, or no FOH. We examined a subset of the sample for differences in age of onset of first psychiatric symptoms, course of illness and measures of symptom severity. These groups were compared using the chi-square procedure for categorical data and the Analysis of Variance (ANOVA) with Scheffe pair wise tests for continuous variables. The Child Bipolar Questionnaire V.2.0, the Yale-Brown Obsessive Compulsive Scale (YBOCS) and the Overt Aggression Scale (OAS) were the principal instruments used to obtain diagnostic information for this study.

RESULTS: We found that children representative of the FOH phenotype when compared to children with PBD who lack this trait had higher indices of severity of mania and depression, as well as other indices that reflect severity and course of illness. Trait factors were derived from a factor analysis of CBQ in a large population of children diagnosed with or at risk for PBD, and used to further elucidate trait features of children with FOH. Children with the FOH traits were also more likely to be defined by six CBQ factors; Sleep/Arousal, Harm to Self and Others, Territorial Aggression, Anxiety, Self-esteem, Psychosis/Parasomnias/Sweet Cravings/Obsessions (PPSO).

LIMITATIONS: This data is derived from samples enriched with bipolar disorder cases. Further validation is needed with samples in which childhood-onset BD is rarer and diagnoses more diverse. Clinician diagnosis was not validated via research interview.

CONCLUSIONS: The FOH phenotype, as defined by a metric derived from combining items from the YBOCS/OAS, is a clinically homogeneous behavioral phenotype of PBD with early age of onset, severe manic and depressive symptoms, and significant social impairment that is strongly associated with 6 CBQ factors and can be easily identified using the CBQ. Through the examination of dimensional features of PBD in an enriched sample of large size, we were able to further refine a phenotype and identify clinical dimensions potentially linked to endophenotypic markers that may prove fruitful in differential diagnosis, treatment and etiological studies of PBD. The nature of the sets of specific symptoms that comprise the FOH factors enabled us to propose a biological model for the phenotype (OUT) that involves a complex orexigenic circuit which links hypothalamic, limbic, and other brain nuclei primarily responsible for the regulation of behavioral and proposed physiological features of the FOH phenotype.