Utility of Survey and Administrative Data for Understanding Health Impacts of Policy Changes Related to Substance Use

Existing administrative and survey data are critical for understanding the impacts of exigent policies on health outcomes related to opioid, cannabis (“marijuana”), and other substance use disorder (SUD). However, SUD-related data in such datasets, including health insurance claims data, have been operationalized differently by regulatory agencies and data providers, creating inconsistencies in data definitions and availability across sources and over time. These phenomena can make it difficult to evaluate health impacts of new policies and interventions such as prescription drug monitoring, naloxone distribution programs, parity laws for provision of SUD treatment, and cannabis legalization.

We interview key informants and review the salient literature to determine the state of the data available for evaluating SUD related health outcomes. We use Massachusetts as a case study for availability of relevant state level data and the potential for linkages within state data sources. We collect comprehensive information on survey and administrative data, including source, cost, inclusion of SUD information, time periods available, sample size, and other key features of the data. We compare and contrast availability of key measures, and note time periods and changes in data collection over time that may impact research utility. We focus on items related to rates of substance use; rates of SUD; and impacts of SUD on the healthcare system and health outcomes. We then evaluate the use of these datasets and sample sizes relevant for statistical power for specific vulnerable populations, including pregnant and parenting women, children, women, racial and ethnic minorities, and LGBTQ individuals. We find that survey and administrative datasets vary substantially in their utility for evaluation and research. While these data can be mined to examine a diversity of policy relevant questions, there are substantial data limitations, including prohibitive regulatory and monetary costs to obtain the data and limited availability by state. Additional considerations include variance in systematic implementation across years and states, as well as fidelity of implementation of data collection mechanisms by state agencies. Potential linkages at the individual level among state level datasets across a large number of state data sources (e.g., substance use treatment, health insurance claims, ambulance trips, criminal justice encounters) to create a state level public health data warehouse can advance our in-depth understanding of health impacts of policy, depending on data specifics and data availability.

Our findings provide a comprehensive resource for using survey and administrative data to evaluate the health impacts of SUD-related policies and interventions. Understanding their strengths and limitations is important for ongoing research related to policy changes. The construction of state level public health data warehouses or record linkage projects connecting individual level information in state data sources are valuable for analyzing the impacts of policy changes if certain conditions are met.