Diagnostic stability in children and adolescents with bipolar disorder, a nationwide register-based study

Abstract

Background: Diagnostic stability of bipolar disorder (BD) in children and adolescents, beyond the first contact has been investigated sparsely. The aim of this study was to investigate the diagnostic stability of BD in children and adolescents using over two decades of nationwide register-based data, and to examine factors associated with change from BD to schizophrenia (ICD-10: F20.x), schizoaffective disorder (ICD-10: F25.x) or other primary psychotic disorders (ICD-10 F23.x-24.x and F28.x-29.x). Methods: Danish register-based data for all incident BD patients diagnosed prior to age 18 years, between January 1st 1995 and December 31st 2014 (N = 519). We graphically illustrated diagnostic change at different follow-up times and studied variables associated with diagnostic change after 3-year follow-up using Poisson regression with robust standard error estimates. Results: The diagnosis of incident BD was relatively stable. The diagnosis did not change for 93% of those followed for at least 6 months, and remained unchanged for 86% and 73% of those followed at least 3 years and 10 years, respectively. In patients followed for at least 3 years after index BD (N = 478), the risk of diagnostic change was 61% higher in males versus females. The risk of diagnostic change for patients diagnosed during hospitalization was 74% higher compared to patients diagnosed at outpatient clinics/emergency rooms. The risk of diagnostic change for patients abusing substances other than alcohol and cannabis was 173% higher compared to patients not abusing such substances. The risk of diagnostic change for patients previously diagnosed with schizophrenia or related diagnosis was 257% higher compared to patients not having been diagnosed with such diagnosis previously, while the risk of diagnostic change in offspring of parents with schizophrenia or related diagnosis was 126% higher compared to patients who did not have parents diagnosed with such disorders. Conclusion: Overall, the stability of the BD diagnosis in the Danish nationwide healthcare registers was high. Factors associated with risk of diagnostic change within 3 years of the initial diagnosis were being male, diagnosis given during hospitalization, substance abuse other than alcohol and cannabis, and a prior diagnosis of schizophrenia or related diagnosis in the patient or in their parents.

DSM-5 and ICD-11 criteria for bipolar disorder: Implications for the prevalence of bipolar disorder and validity of the diagnosis – A narrative review from the ECNP bipolar disorders network

“…Subsequently the four mentioned published studies and one ongoing study ( Angst et al, 2020 ;Fredskild et al, 2019Fredskild et al, , 2020Gordon-Smith et al, 2017 ;Machado-Vieira et al, 2017 ) have investigated the prevalence of bipolar disorder according to DSM-5 versus DSM-IV or ICD-10 criteria, but further studies are greatly needed. Thus, further studies are needed within the sixth mentioned validation areas: 1) Clinical presentation according to DSM-5 versus DSM-IV/ICD-10, 2) Associations of the phenotype with paraclinical data such as brain imaging and blood-based data, 3) Delimitation from other disorders, specifically schizophrenia, schizoaffective disorder, depressive disorder, personality disorders, 4) Pathogenesis -associations to family history / genetics and environmental stressors, 5) Prognosis, specifically in relation to implications for diagnostic delay and early intervention as well as long-term diagnostic stability ( Kessing, 2005 ;Kessing et al, 2015 ;Laursen et al, 2020 ), 6) Medical treatment effects of lithium and other mood stabilizers. At some point, the official classification systems will need to incorporate information from sources other than symptoms, and this may come not only from genetic, neuroimaging, and biomarker data, but also from the increasing use of digital markers, in the frame of so-called precision psychiatry Salagre et al, 2018 ).…”