Characterizing Management of Depression in a Primary Care Setting

Introduction: Depression is an important cause of disability in the US and is largely managed by primary care providers (PCPs). Understanding how PCPs manage depression is an important step toward identifying opportunities for improvement.  
Methods:  Electronic health records (EHRs) of patients >18 years with ≥1 PCP contact in the Sutter Health network during 2013-2017 were evaluated. The study population (N=1,618,037) included four mutually exclusive patient cohorts: (1) no depression; (2) clinical evidence of depression; (3) non–major depressive disorder (MDD) diagnosis (defined as having ICD-9/ICD-10 codes for any depression type other than MDD); and (4) MDD diagnosis. Results presented herein compare non-MDD (n=7760 [0.5%]) and MDD (n=22,970 [1.4%]) cohorts.

Results: Screening for depression using PHQ-9 (scores ≥5) ranged from 16.1% in patients with non-MDD depression to 26.3% in MDD patients. Patients with MDD had more antidepressant use (70.7% vs 42.6% for non-MDD; P < 0.05). MDD was also associated with higher percentage of patients with at least one comorbidity (Charlson Comorbidity Index score ≥1, 23.4% vs 19.2%; P < 0.05), higher number of prescriptions (76.8% vs 67.0%; P < 0.05), visits to mental health providers (7.4% vs 4.0%; P < 0.05), urgent care visits (16.8% vs 15.0%), emergency department visits (8.8% vs 5.7%), and inpatient admissions (3.7% vs 1.9%) compared with non-MDD depression.

Conclusion: Use of PHQ-9 as a screening tool appears consistent with previously reported data (Akincigil and Matthews, 2017) but needs further investigation because of potential underestimation of depression screening rates.  Overall health resource utilization was greater in patients with MDD than in those with non-MDD depression.