Service utilization and associated direct costs for bipolar disorder in 2004: an analysis in managed care.J Affect Disord. 2007 Aug; 101(1-3):187-93.JA
BACKGROUND
Bipolar disorder is a chronic and costly condition. This analysis examines health care costs associated with bipolar disorder in 2004 and contrasts them with those for depression, a better understood mental illness.
METHODS
Health care costs associated with bipolar disorder and non-bipolar depression were determined using private payer administrative claims. Individuals having 2 claims with a primary ICD-9-CM code for bipolar disorder or depression were categorized as bipolar disorder or non-bipolar depression patients, respectively. Comparisons between patient groups were adjusted for demographic differences and comorbid diagnoses.
RESULTS
On average, bipolar patients (n=6072) used significantly more psychiatric resources per person than depression patients (n=60,643), and had more mean psychiatric hospital days, psychiatric and medical emergency room visits, and psychiatric office visits (p<.001 for all). Bipolar patients were slightly less likely to be treated with antidepressants, but substantially more likely to be treated with antipsychotics, anticonvulsants, lithium, and benzodiazepines (p<.001 for all). Mean direct per-patient costs were $10,402 for bipolar patients and $7494 for depression patients (p<.001), with the primary differences observed for psychiatric medication ($1641 vs. $507) and psychiatric hospitalization ($1187 vs. $241).
LIMITATIONS
Patients were categorized based on diagnostic codes in administrative claims data, which may not always be accurate. Results may not generalize beyond private payer populations in the US.
CONCLUSIONS
Bipolar disorder is associated with significantly greater per-patient total annual health care costs than non-bipolar depression, as well as significantly greater psychiatric costs. Bipolar disorder, a chronic condition often suboptimally treated, may represent a good target for disease-management programs.