Using APCD Data to Identify Young People with High Mental-Health Needs

August 2020 – Together with our partners at the Vermont Blueprint for Health, Onpoint has leveraged data from Vermont’s APCD to identify the population of pediatric patients with mental-health diagnoses and explore patterns in patients' healthcare utilization and cost. Specifically, the State of Vermont was interested in identifying a high-utilizer subgroup of young people (ages 1-21 years) with mental-health diagnoses that might benefit from a mobile response unit initiative designed to improve access to mental-health services and reduce the need for emergency department visits and inpatient admissions for mental-health purposes. The aims of this study were to identify this high-utilizer group using specific healthcare utilization patterns and to provide baseline data that could be used to evaluate whether a mobile response unit initiative could improve outcomes and lower costs over time.

To begin, we identified all young people in the VT APCD who had a medical claim in 2018 with a principal mental-health diagnosis. From this group, a high-utilizer subgroup was identified as patients who had at least one acute mental-health event (e.g., an ED visit or inpatient hospitalization) or multiple Case Management and/or Home- and Community-based Services with a principal mental-health diagnosis. Using these criteria, roughly one in six young people with a mental-health diagnosis (16%) was characterized with high utilization.

The youth in the high-utilizer group differed in many notable ways from their counterparts who had a mental-health diagnosis but were not classified as high-utilizers. Compared to the non-high-utilizer group, the high-utilizer group had higher proportions of patients that were male, were 11 years or older, and were enrolled in Medicaid. Patients in the high-utilizer group were substantially more likely to suffer from intentional self-harm and suicide-related injuries, schizophrenia and other psychotic disorders, personality disorders, impulse control disorders, and alcohol-related disorders, among others. Additionally, the average and median mental-health-service expenditures for this group were approximately 10 times greater than the expenditures for the non-high-utilizer group.

Due to the high volume and cost of their necessary menta-health services as well as the nature of their diagnoses, the patients in the high-utilizer group appeared likely to benefit from a mobile response unit initiative. The data generated in this analysis can serve as a baseline against for future initiatives and analyses. Using analytic approaches that generate results such as these, our clients can use APCD data to inform the development of targeted interventions and evaluate their effectiveness, ultimately yielding data-informed programs that improve health outcomes while reducing cost.

Figure 1. Demographics & Expenditures by Utilization Group