Misdiagnosing Bipolar I as MDD Increases Health Care Utilization, Costs

Prompt and accurate diagnosis of bipolar I disorder (BD) will likely significantly reduce health care resource utilization (HRU) and costs, according to study findings published in Journal of Affective Disorders.

Researchers sourced data for this study from the IBM® MarketScan® Research Databases, Multi-State Medicaid Database, and the Medicare Supplemental and Coordination of Benefits Database. Patients (N=1,044,219) with 1 or more diagnosis of BD and/or major depressive disorder (MDD) between 2014 and 2019 were included. Using a weighted approach, HRU and costs were evaluated among the patients who were only diagnosed with BP (n=16,072; 57.5% women; mean age, 37.7 [SD, 13.9] years) and compared with those who were first misdiagnosed with MDD and subsequently diagnosed with BD (n=14,729; 64.1% women; mean age, 35.3 [SD, 13.7] years).

Prior to weighting the BD-only and MDD misdiagnosed groups, 38.4% and 43.1% were diagnosed by a primary care provider, and 42.6% and 56.0% were receiving antidepressants, respectively.

After weighting, both the BD-only and MDD misdiagnosed groups had 0.2 hospitalizations in the previous 12 months, 1.0 emergency department (ED) visits, and 13.0 outpatient visits. Of those, 0.1 hospitalizations, 0.2 ED visits, and 4.0 outpatient visits were due to mental health, respectively.

The total health care costs in the previous 12 months were between $11,761 and $11,864, of which between $3635 and $3640 comprised mental health costs and between $618 and $619 comprised mental health pharmacy costs.

During the entire follow-up period, MDD misdiagnosis was associated with more all-cause hospitalizations (rate ratio [RR], 1.94; 95% CI, 1.86-2.05; P <.001), ED visits (RR, 1.33; 95% CI, 1.27-1.39; P <.001), outpatient visits (RR, 1.38; 95% CI, 1.35-1.41; P <.001), physician office visits (RR, 1.39; 95% CI 1.36-1.42; P <.001), home health care visits (RR, 1.10; 95% CI, 0.90-1.31; P =.289) and other outpatient visits (RR, 1.35; 95% CI, 1.26-1.45; P <.001) compared with correct primary diagnosis.

Patients diagnosed with MDD experienced higher rates of mental health-specific hospitalizations (RR, 2.19; 95% CI, 2.08-2.32; P <.001), ED visits (RR, 1.77; 95% CI, 1.68-1.87; P <.001), outpatient visits (RR, 1.77; 95%, CI, 1.71-1.82; P <.001), physician office visits (RR, 1.77; 95% CI, 1.72-1.82; P <.001), home health care visits (RR, 1.75; 95% CI, 1.19-2.41; P =.008), and other outpatient visits (RR, 1.75; 95% CI, 1.53-1.98; P <.001) compared with the BD-only group.

Misdiagnosis was associated with increased costs for all-cause hospitalizations (mean difference [MD], $3253; 95% CI, 2748-3750; P <.001), ED visits (MD, $729; 95% CI, 590-901; P <.001), outpatient visits (MD, $2177; 95% CI, 1729-2583; P <.001), physician office visits (MD, $2010; 95% CI, 1600-2384; P <.001), home health care visits (MD, $32; 95% CI, -31 to 115; P =.040), and other outpatient visits (MD, $135; 95% CI, 61-197; P <.001) compared with correct diagnosis during the entire follow-up period.

Mental health-specific costs were increased for hospitalizations (MD, $3254; 95% CI, 2878-3654; P <.001), ED visits (MD, $477; 95% CI, 417-539; P <.001), outpatient visits (MD, $2105; 95% CI, 1838-2368; P <.001), physician office visits (MD, $1993; 95% CI, 1757-2237; P <.001), home health care visits (MD, $27; 95% CI, 3-52; P =.040), and other outpatient visits (MD, $85; 95% CI, 33-135; P <.001) among the misdiagnosis group compared with the correct diagnosis group.

Similar trends were observed in the first 12-month follow-up.

These results may be limited by a lack of generalizability for patients who do not have health care insurance coverage.

Study authors concluded, “The misdiagnosis of BD as MDD is associated with significantly higher healthcare utilization and costs compared with correctly diagnosed BD. The results of this study suggest the prompt and accurate diagnosis of BD has the potential to significantly reduce healthcare utilization and costs.”

Disclosure: Multiple authors declared affiliations with industry. Please refer to the original article for a full list of disclosures.