Effectiveness of Outpatient C-L Models for Bipolar Disorder in Primary Care: A Secondary Analysis

Abstract: Background:

Individuals with bipolar disorder commonly present in primary care. Collaborative care and direct-care referral models can more closely match clinical services to need. It is unknown which model is more effective for individuals with bipolar disorder.

Methods:

We conducted a randomized comparative effectiveness trial (ClinicalTrials.gov NCT02738944, approved by the University of Washington IRB) with 1004 patients screening positive for bipolar disorder and/or PTSD in twelve Federally Qualified Health Centers in three states. A subset of 179 patients with psychiatrist-diagnosed bipolar disorder were treated for 12-months and included in the analysis. We hypothesized that in secondary analysis, outcomes would be better for patients with bipolar disorder treated in the telehealth referral arm (direct specialist care), compared to the collaborative care (CC) arm.

Participant characteristics and outcomes were assessed at enrollment before randomization and 12-months by masked survey. The primary outcome was change in Veterans RAND 12-Item Health Survey Mental Health Component Summary (MCS) from baseline to 12-months. Secondary outcomes were change in Symptom Checklist for Depression (SCL-20), and proportion in euthymic mood state classified by Internal State Scale (ISS). T-tests and multiple regression were used to test differences between study arms in changes in clinical outcomes. Medication data was extracted from clinical registries.

Results:

Participant mean age was 41 years old, and 74% were women. Medication patterns were similar in both arms. Treatment with second generation antipsychotic (SGA) (26% baseline to 55% at 12-month) or mood stabilizer medication (28% to 37%) increased, while antidepressant treatment without concurrent mood stabilizer or SGA decreased from 45% to 25%, with no differences by arm. Overall, MCS scores improved significantly (MΔ=9.74; SD=16.00) and did not differ by arm (MΔ=9.74; referral-MΔ=10.48, CC- MΔ=9.28; t=-0.41, p=.69). Depressive symptoms assessed by SCL-20 scores improved significantly in both arms and did not differ by arm (referral-MΔ=-0.75, CC- MΔ=-0.75; t=0.02, p=.98). Individuals in euthymic mood state (classified by the ISS) increased from 11% to 25% and was similar in both groups.

Discussion:

Collaborative care and direct referral care were both associated with significant improvements in mental health quality of life and clinical status in individuals with bipolar disorder in primary care. Strengths of collaborative care (CC) including team-based care, increased opportunity for clinical contact, and flexible intensity of psychiatrist-involvement, may explain why CC was similarly effective to direct care from psychiatrists and psychologists.

Conclusion:

Both care models were associated with significant improvements in mental health quality of life and clinical status. Clinics could implement the more feasible option or consider stepped-care strategies in which patients who do not improve with primary care or collaborative care are referred to specialty care.

References:

1. Fortney JC, et al. JAMA Psychiatry. 2021;78:1189-1199.

2. Cerimele JM, et al. Gen Hosp Psychiatry. 2020;65:28-32