Supporting and Enhancing the AHRQ Quality Indicators Program

There are dozens of QIs within the AHRQ QI Program, which are currently bundled across five modules: Patient Safety QIs, Inpatient QIs, Pediatric QIs, Prevention QIs, and Emergency Department Prevention QIs. Examples of QIs include hypertension admission (a Prevention QI) and Heart Failure Mortality rate (an Inpatient QI). These QIs can serve to monitor a hospital or health system’s clinical performance and pinpoint quality-of-care issues, driving health care improvement. The QIs must be refined regularly to remain consistent with clinical guidelines, meet coding standards, and ensure continued applicability to medical concepts given changes in clinical practice and technology. 

On this project, Mathematica refines the methodologies and technical specifications for current AHRQ QIs and helps develop novel QIs. We research, test, and evaluate the scientific acceptability of the QIs to ensure they are valid, reliable, feasible, and usable measures of health care quality. To reduce indicator bias, we conduct risk adjustment to account for differences in patient populations across hospitals or geographics that affect outcome rates and are unrelated to the quality of care. This work requires securely accessing and managing a variety of data sources, such as Healthcare Cost and Utilization Project State Inpatient Databases and State Emergency Department Databases, U.S. Census Bureau data, Centers for Disease Control and Prevention data, and Medicare and Medicaid claims accessed via Mathematica’s Health Data Innovation Lab. 

For each QI we develop and update, we produce technical documentation, research code, and QI software updates so that health system partners can use the QIs to inform performance monitoring and decision making. Hospitals, health systems, insurers, and researchers can use their own hospital administrative data to calculate QI rates for their hospital or geographic area using the SAS code or the AHRQ QI software, WinIQ, provided by AHRQ and updated by Mathematica. We also support the annual public release of new and updated QIs and provide technical assistance to end users. 

Finally, Mathematica contributes to research and innovative approaches to quality measurement, including supporting the development of new QI modules and exploring the use of new data sources for QI development. For example, Mathematica is working to expand the QI program to cover measures related to severe maternal morbidity and diagnostic safety. Mathematica is supporting AHRQ to refine the severe maternal morbidity definitions and launch new modules to calculate this type of morbidity and diagnostic safety topics.