Using Administrative Data to Identify Risk and Protective Factors of Adverse Youth Outcomes: Lessons Learned in the Statewide Implementation of the Communities That Care Prevention System in Colorado

Introduction: This presentation describes how 47 community coalitions across Colorado used administrative datasets to identify risk and protective factors prioritized by communities implementing the Communities That Care (CTC) prevention system. Colorado administers the Healthy Kids Colorado Survey (HKCS), which includes some of the reliable and validated scales measuring risk and protective factors that are part of the CTC Youth Survey. However, the HKCS was unable to include all the CTC measures due to survey length; additional information on risk and protection were subsequently collected through administrative datasets.

Methods: Three administrative datasets were used to identify risk and protective factors of adverse youth outcomes in Colorado: 1) U.S. Census Bureau, 2) Colorado Department of Education School-View, and 3) Uniform Crime Reports. Community decision-makers used indicators from these datasets to determine whether their community had high levels of risk for adverse outcomes. Communities compared and contrasted local-level indicators with regional and state data from administrative datasets to assist their assessment and selection of risk and protective factors in their community.

Results: Communities used indicators from these datasets to understand community economic issues (n=9), student academic achievement (n=7) and connection to school (n=9), and the community’s organization to support youth development (n=8). Education, crime, and Census data will be used in ongoing evaluations to track changes in levels of risk and protection in these Colorado communities over time. CTC communities faced challenges in using these datasets, including: obtaining data at the appropriate geographic level, having the appropriate expertise to understand the complexities in the datasets, and using data to inform community decision-making.

Conclusions: Collecting risk and protective factor information through youth self-reported attitudes and behaviors is not always feasible for communities. Communities faced with this challenge struggle to identify community-level data to measure risk and protective factors. Administrative data can be used to understand risk and protective factors of adverse outcomes and support decision making and strategic action. Utilization of these datasets can result in more practitioners using a shared risk and protective factor approach to primary prevention in our communities.