Exploring factors of diagnostic delay for patients with bipolar disorder: a population-based cohort study

Background: Bipolar disorder if untreated, has severe consequences: severe role impairment, higher health care costs, mortality and morbidity. Although effective treatment is available, the delay in diagnosis might be as long as 10–15 years. In this study, we aim at documenting the length of the diagnostic delay in Hungary and identifying factors associated with it. Methods: Kaplan-Meier survival analysis and Cox proportional hazards model was employed to examine factors associated with the time to diagnosis of bipolar disorder measured from the date of the first presentation to any specialist mental healthcare institution. We investigated three types of factors associated with delays to diagnosis: demographic characteristics, clinical predictors and patient pathways (temporal sequence of key clinical milestones). Administrative data were retrieved from specialist care; the population-based cohort includes 8935 patients from Hungary. Results: In the sample, diagnostic delay was 6.46 years on average. The mean age of patients at the time of the first bipolar diagnosis was 43.59 years. 11.85% of patients were diagnosed with bipolar disorder without any delay, and slightly more than one-third of the patients (35.10%) were never hospitalized with mental health problems. 88.80% of the patients contacted psychiatric care for the first time in outpatient settings, while 11% in inpatient care. Diagnostic delay was shorter, if patients were diagnosed with bipolar disorder by non-specialist mental health professionals before. In contrast, diagnoses of many psychiatric disorders received after the first contact were coupled with a delayed bipolar diagnosis. We found empirical evidence that in both outpatient and inpatient care prior diagnoses of schizophrenia, unipolar depression without psychotic symptoms, and several disorders of adult personality were associated with increased diagnostic delay. Patient pathways played an important role as well: the hazard of delayed diagnosis increased if patients consulted mental healthcare specialists in outpatient care first or they were hospitalized. Conclusions: We systematically described and analysed the diagnosis of bipolar patients in Hungary controlling for possible confounders. Our focus was more on clinical variables as opposed to factors controllable by policy-makers. To formulate policy-relevant recommendations, a more detailed analysis of care pathways and continuity is needed.