HHS Report Highlights Progress on PCOR Data Infrastructure
A recent report of the U.S Department of Health & Human Services Office of the Assistant Secretary for Planning and Evaluation (ASPE) highlights progress on the collection, linkage, and analysis of data for patient-centered outcomes research.
The ASPE annual report highlights the accomplishments of 34 multi-agency projects to build data capacity and support the HHS mission to advance science and improve knowledge about the comparative effectiveness of health care interventions. In fiscal year 2021, the report focuses on projects aligned with four national health priorities—COVID-19, maternal health, health equity and social determinants of health (SDOH), and the opioid crisis.
The section on social determinants and health equity gives a flavor of the type of projects being supported.
Advancing Equity and Inclusion through Improved Collection and Use of SDOH and Other Non-Healthcare Data to Enhance Person-Centered Care. Increased recognition of the effects of SDOH and differential access to care on health outcomes has attracted corresponding attention to the data and infrastructure needed to support patient-centered outcomes research on SDOH, the report notes. The following Office of the Secretary’s Patient-Centered Outcomes Research Trust Fund (OS-PCORTF) projects address the imperative for expanded access to demographic and social risk data, environmental data, and other non-health sector data to support health equity and research.
• The Community and Clinical Data Initiative (CODI): Integrated Data for Patient-Centered Outcomes Research Project (2018) and the Community and Clinical Data Initiative (CODI 2.0) (CDC, 2020) focus on increasing access to longitudinal clinical and community data with linkage and de-duplication tools. CODI 2.0 will further develop its infrastructure to create linkages across the health and social sectors for both adult data and pediatric data. These linkages will bring in data from larger and more diverse geographic areas and data networks, and it will allow the project to expand its clinical focus from obesity into diabetes and heart disease.
• Enhancing Patient-Centered Outcomes Research: Creating a National Small-Area Social Determinants of Health Data Platform (AHRQ, 2019). This project leverages publicly available data sources to develop a national standardized database with readily linkable SDOH variables at the small-area and other geographic levels. The beta version of the SDOH database is currently available for researchers to incorporate community SDOH characteristics in health outcomes research.
• Data Set on Intellectual and Developmental Disabilities (ID/DD): Linking Data to Enhance Person-Centered Outcomes Research (ASPE, 2021). The planned dataset will link key sources of ID/DD data, including Support Intensity Scale scores, Medicaid claims, National Core Indicators survey data, National Core Indicators COVID-19 supplement survey data, and potentially, other relevant state-level data for four to six states. The dataset will fill a gap in the data available to conduct ID/DD-focused patient-centered outcomes research and generate evidence to support person-centric health decision-making and equitable policymaking.
• Data Linkage: Evaluating Preserving Privacy Methodology and Augmenting the National Hospital Care Survey with Medicaid Administrative Records (CDC/NCHS, 2020). This project is currently linking data from the NCHS NHCS with Transformed Medicaid Statistical Information System (T-MSIS) data to expand the capacity to conduct studies of a wide range of HHS priority issues, particularly among the Medicaid covered population. This will improve access to linked federal data assets, enabling patient-centered research on health outcomes for a wide range of research priorities, including the outcomes of initiatives targeting opioid use and mental health services, efficacy of treatment protocols and drugs, disparities in efficacy for vulnerable subpopulations, and the role of social programs and SDOH in health outcomes.
The report gave these examples of OS-PCORTF contributions and impact:
• Sharing Tools, Improving Research Efficiency. PCORnet and the Robert Wood Johnson Foundation have already started using CODI’s Toolbox, specifically the growthcleanr tool to clean EHR data.
• Improving Data Accessibility for a More Comprehensive Understanding of Health. The AHRQ SDOH database makes community-level SDOH data on five SDOH domains (social context, economic context, education, physical infrastructure, and healthcare context) more readily available for researchers and policymakers. As of November 2021, there were about 10,000 file downloads from AHRQ’s SDOH database web page.
• Disseminating Tools for Enhanced Data Analysis. Multiple algorithms developed from the 2018-2019 NCHS Data Linkage projects are now publicly available, including: The Opioid NLP component, which can search clinical notes text for opioid involvement; the SUD and MHI NLP component, which flags mentions of SUD and mental health issues in clinical note text; the Medical Code-based translation from SAS, which flags ICD-10-CM codes related to opioid involvement, SUD, and mental health issues in structured hospital data.
In conclusion, the report states that the OS-PCORTF projects have improved the availability of data and linked a myriad of data sources to expand their research applications. These include federal data sources that contain important information on public health trends and health care utilization, as well as clinical data sources such as EHRs, hospitals, and registries. Evidence generated from these data sources will improve collective understanding of patient health and outcomes, and the cost and quality of health care being delivered.
The projects are also prioritizing improvements in the quality and volume of data gathered directly from patients and communities to identify SDOH risk factors, increase representation of diverse populations, and continue to emphasize patient-centeredness in research and clinical care.
