Diagnostic Stability in Bipolar Disorder: A Follow-up Study in 130,000 Patient-Years

ABSTRACT

Objective: Diagnostic stability is the degree to which a diagnosis remains unchanged during time. Our main objective was to evaluate the diagnostic stability of bipolar disorder (BD) in psychiatric outpatient consultations and determine the socio-demographic variables influencing its stability.

Methods: The Cumulative Register of Cases of the Community of Madrid provided data on all outpatient visits conducted at Madrid’s Community Mental Healthcare Centers between 1980–2009. Diagnoses were made according to ICD-9/ICD-10 criteria. Two indices were measured: temporal consistency (maintenance of the diagnosis over time) and diagnostic constancy (presence of BD diagnosis in at least 75% of visits). κ coefficient measured the agreement between diagnoses in the first and last evaluations (prospective and retrospective consistency).

Results: 14,557 patients were diagnosed with BD for at least 1 evaluation and had at least 10 visits and 1 year of follow-up. At first evaluation, 3,988 patients were diagnosed with BD (prospective consistency 50.8%), and at last evaluation 5,396 patients were diagnosed with BD (retrospective consistency 37.5%). A total of 2,026 patients were diagnosed with BD at their first and last evaluations (prospective consistency 18.3%).

Conclusions: This longitudinal study conducted in community mental health centers reflects common diagnostic practices in outpatient settings over a 30-year period (130,000 patient-years). Delay of > 10 years was found to achieve diagnostic stability. Frequent diagnostic shifts were found in relation to BD, the most common being with other affective disorders. Anxiety was also a common misdiagnosis. Greater stability was associated with having been diagnosed after hospitalization, having an age at onset > 25 years, and having an age at diagnosis < 24 years.