The quality improvement ratio: a method to identify changes in screening rates

Area of Science:

• Healthcare Quality Improvement
• Health Services Research
• Biostatistics

Background:

• Managed care organizations utilize Health Plan Employer Data and Information Set (HEDIS) measurements for report cards and improvement projects.
• Conventional two-group analysis compares measurements between periods but has limitations including overlapping members and confounding factors.
• These limitations hinder accurate identification of changes due to interventions.

Purpose of the Study:

• To introduce and validate the Quality Improvement Ratio (QuIR) as a superior method for analyzing changes in health metrics over time.
• To address the statistical invalidity and confounding issues present in conventional analysis methods.
• To provide a focused approach for evaluating the effectiveness of healthcare interventions.

Main Methods:

• The Quality Improvement Ratio (QuIR) is calculated using only members who change status between two measurement periods.
• For screening interventions, QuIR is the ratio of members newly screened to those becoming unscreened.
• Statistical significance of QuIR is determined using McNemar's test for paired binary data.

Main Results:

• The QuIR method is not affected by members who do not change status, unlike conventional methods.
• It avoids confounding from population changes and overlapping members inherent in two-group analysis.
• McNemar's test provides a statistically sound basis for evaluating QuIR.

Conclusions:

• The Quality Improvement Ratio (QuIR) provides a statistically valid and focused approach to measuring intervention effectiveness.
• It directly assesses changes attributable to an intervention, unlike conventional methods that compare overall rates.
• QuIR enhances the accuracy of quality improvement initiatives in healthcare settings.