AHRQ 2.0: The Importance of Data-Informed Insights
A few weeks ago, I wrote about my vision for creating an AHRQ 2.0—transforming the Agency into a leader in health services research in a digital age. I described four pillars that are central to this effort. The first pillar is: developing a platform to create data-informed insights.
In addition to other initiatives, AHRQ is well known for two very powerful data programs. The Healthcare Cost and Utilization Project (HCUP) is the Nation’s most comprehensive source of hospital data and also includes information on inpatient care, hospital-based outpatient surgery, and emergency department visits. The Medical Expenditure Panel Survey (MEPS) is the only national source of data measuring how Americans use and pay for medical care, health insurance, and out-of-pocket spending.
Over time, feedback from the health care and research communities has affirmed the unparalleled value of HCUP and MEPS, as well as other AHRQ resources, including the National Healthcare Quality & Disparities Report. However, I believe that their impact can be increased exponentially by greatly expanding the use of these data. We can give Americans greater return on their investment in AHRQ by adding to the types of data we make available, increasing the number of users, and by diversifying the ways in which these data, tools, and research can be applied.
To meet these goals, we must leverage our assets to become a health data platform, one that collects and manages large sets of data and related information from a number of sources, and facilitates cutting-edge analytics.
Let’s start with what AHRQ already has. HCUP has collected significant data over time that can be used to identify important trends and patterns in health care delivery and understand them on both national and more regional and local levels. MEPS offers the ability to identify trends and patterns in cost and access to care, along with insights into what kinds of coverage is offered by employers and States.
Today, our data programs are powerful and are used by many researchers and analysts. But they can nevertheless do more to contribute to our knowledge of these critical health care issues. Consider for a moment all the possible determinants of health—everything that comprises a person’s wellbeing. Many of these are tracked in data programs maintained by others. HCUP and MEPS can incorporate more and richer data, becoming the foundation of a data platform that paints a more complete picture of patients and their needs.
Think of the implications! The impact on research, policymaking, and health care decision-making is potentially far-reaching. For instance, we could draw on AHRQ data as well as data from our sister agencies within the Department of Health and Human Services, and along with State and local data, we could pinpoint potential health care hotspots before they happen.
This is called predictive analytics: the rapid analysis of mined data with an emphasis on predicting future events rather than describing current ones. There have been conversations on using our data to simulate large groups of individuals and their actions to provide insights about substantive and technical issues. Further developed, this approach has great promise to improve decision-making within government. It also offers researchers and health care decision-makers tools to inform decisions outside of government.
AHRQ has been at the forefront when it comes to analyzing health care data to promote research, policymaking, and improved health. I believe that we can build a data platform that will enable researchers to conduct even more incisive research; policymakers to enact informed policies; and providers and patients to make the best health care decisions possible, all furthering AHRQ's critical mission to improve safety, quality, and access to care.
In the weeks ahead, I’ll be blogging further about other aspects of my vision for AHRQ, aspects that are vital the Agency’s future. I hope you’ll stay tuned!
Gopal Khanna is Director of the Agency for Healthcare Research and Quality.
Page originally created November 2017