Impact of Integrated Care on Thoughts of Self-Harm for BIPOC

Katherine Grimes, MD, MPH

Cambridge Health Alliance – Faculty
Katherine Grimes edited

Scientific Abstract

Background: Despite high need, only 1/5 U.S. children with MH/SUD needs receives treatment. Poverty, violence and other community risk factors nearly double national child mental health prevalence rates. This study investigates the impact of an innovative pediatric integrated care approach, including peer-to-peer family support, on identification and engagement of children who are at risk for self-harm and experience access barriers in usual care.

Methods: Pediatric patients from an urban safety-net system were identified by their PCPs as at-risk for childhood trauma or other mental health needs and referred to the Collaborative Practice Model (CPM) study for child mental health evaluation. Study participants received the CPM multidisciplinary team evaluation within the primary care clinic, including peer-to-peer parent support and child mental health specialty consultation. Structured assessments were obtained at baseline, including assessment of Adverse Childhood Experiences (ACEs). Standardized measures of clinical functioning were obtained at baseline, 6 and 12 months. We used multivariable regression to assess differences in ACEs and CAFAS scores at baseline by sociodemographic characteristics. We then analyzed repeated measures data (baseline, 6 mos., 12 mos.) on clinical functioning (CAFAS total score as well as subcategories) were analyzed using generalized linear mixed models with person-specific random intercepts and fixed-effects parameters for group (e.g., race/ethnicity), time, and group*time interactions.

Results: Study population included 171 children 3-18 years old. Race/Ethnicity: 76% were children of color, 60.8% of our sample identify as Latinx, 17.0% as non-Hispanic white, 16.4% as non-Hispanic Black/African American. Average ACE scores were 3+, indicating exposure to a severe burden of stressors. Mean CAFAS scores at baseline indicated that on average, children in the sample were exhibiting moderate to marked functional impairment at the time of study enrollment. In mixed effects modeling, total CAFAS scores were reduced by about 18.2 pts. at 6 months (p<.001) and 22.9 pts. at 12 months (p<.001). The 12-month reduction was significantly larger than the 6-month reduction (-4.7 pts., p=0.023). At baseline, Black youth had significantly higher scores than White youth (10.6 pts., p=0.041). All race/ethnic groups experienced significant reductions at 6 and 12 months. Reductions at 12 months were 12.2 pts. larger for Black youth compared to White youth (p=0.009). For the CAFAS subcategory of Self-Harm, the overall predicted probability of any SI/SH at baseline was about 25.0% (SE=3.1). There were significant overall reductions in likelihood of SI/SH at 6 months (-15.7 pts., p<.001; or a 63% reduction) and 12 months (-18.0 pts., p<0.001; or a 72% reduction).

Conclusions: At a time when we face a national emergency in children’s mental health, and the risk for youth suicide is climbing, particularly among populations, such as BIPOC children, who have historically been underserved by usual care, there is unprecedented urgency to find, test and implement effective interventions for youth having thoughts of self-harm. Preliminary results for the CPM model of pediatric integrated care show it to have significant promise in improving health outcomes.

Live Zoom Session – March 9th

research Areas


Katherine E. Grimes, MD, MPH, Lindsay DiBona, LICSW, Tim Button, MPH, Timothy B. Creedon, PhD

Principal Investigator

Katherine E. Grimes, MD, MPH