Does adding clinical data to administrative data improve agreement among hospital quality measures?

Area of Science:

• Health Services Research
• Health Informatics
• Quality Improvement

Background:

• Hospital performance measures for patient mortality and readmission show limited agreement.
• The Veterans Health Administration (VA) is the nation's largest integrated healthcare system.

Purpose of the Study:

• To determine if combining clinical data (laboratory tests, vital signs) with administrative data improves agreement among hospital performance measures.
• To assess if this data enhancement leads to better concordance with other hospital quality indicators.

Main Methods:

• Utilized patient-level administrative and clinical data, along with hospital-level quality indicators from the VA (2007-2010).
• Examined changes in 30-day mortality and 30-day readmission rates for acute myocardial infarction, heart failure, and pneumonia after adding clinical data.
• Evaluated concordance between enhanced measures and other quality indicators.

Main Results:

• Data enhancement improved model performance and significantly altered hospital profiles for 30-day mortality.
• The impact on 30-day readmission rates was modest.
• Concordance between enhanced measures and other quality indicators (e.g., Joint Commission, VASQIP, case volume) remained poor.

Conclusions:

• Integrating laboratory tests and vital signs into performance measures did not enhance agreement across hospital quality indicators within the VA.
• Continued efforts in risk adjustment modeling are crucial, but validation is necessary before using them as reliable quality measures.