P3B301: Evaluating Patient Outcomes in a Multicenter Pediatric Primary Care Behavioral Health Integration Program

Background: Child and adolescent mental health concerns significantly increased after the beginning of COVID-19. In response, we embedded integrated health therapists (IHTs) into pediatric primary care (PC) clinics to provide short-term therapy. We created additional mental health hubs staffed to offer long-term therapy and psychiatry services for patients with more complex needs. We currently have IHTs in 7 pediatric PC sites and 3 freestanding operational hubs. With our continued expansion, we sought methods of monitoring patient progress and outcomes and in June 2022 tasked clinicians with ensuring that: 1) Patients complete measurement-based care (MBC) screening tools (PHQ-9, GAD-7 or PSC-17) at each visit and 2) Providers complete a Clinical Global Impression (CGI) at each visit. Data for each of these measures are captured in our electronic medical record (EMR) and results are stored in our Patient Registry (PR). We compared pre and post MBC and CGI measures for our patients.

Methods: Data extracted from our PR included MBC baseline and most recent visit screening values, CGI scores, patient demographics, diagnoses and whether patients completed treatment within the program. Preliminary statistical comparisons between baseline and most recent visit were compared using appropriate Wilcoxon matched pairs rank sum tests of raw scores, and chi square or Fisher’s exact tests for categorical analyses using STATA 16 (College Station, TX). Additional analyses will explore variability in scores throughout treatment and potential associations between patient screening assessments and provider assigned CGI. Analyses will also be completed by patient diagnosis and multivariable modeling may be used.

Results: To date, our integrated care program has assessed 2,167 patients who were either assigned a treatment tract within our program or triaged to services outside our system. Visit MBC and CGI metrics are consistently being obtained (Figure 1). Initial patient MBC screening and CGI-severity scores are presented in the table. Mean patient age = 12.7 + 4.1 years. Pre and post comparisons identified significant improvement in patient outcomes on each of the screening assessments: GAD7: 42.6% (n = 343) improved, 42.9% (345) no difference, 14.5% (117) worsened (p < 0.001). PHQ9: 29.9% (87) improved, 62.2% (181) no difference, 7.9% (23) worsened (p < 0.001). PSC17: 27.1% (191) improved, 63.7% (449) no difference, 9.2% (65) worsened (p < 0.001). Patient improvement was also demonstrated on provider-assessed CGI-Improvement subdomain (p < 0.001).

Conclusion: We demonstrated significant improvements in patient scores in MBC screening assessment and in provider completed CGI scores. Data illustrated screening tools and consistent measurement using validated tools can monitor patient progress. Detailed analyses of associations between MBC metrics and CGI will be presented. Subset analyses will be presented by patient diagnosis, patient completion of short-term therapy in their pediatric home, and patient progression to the hubs for additional treatment.