Monitoring Mortality Rates After Heart Surgery
Author Information
Author(s): Tom Marshall, Mohammed A Mohammed
Primary Institution: University of Birmingham
Hypothesis
Does case-mix adjustment affect the identification of special cause variation in hospital mortality rates after coronary artery bypass grafting?
Conclusion
There is moderate-good agreement between signals of special cause variation between observed and risk-adjusted mortality.
Supporting Evidence
- Cohen's Kappa was 0.54 indicating moderate agreement between crude and risk-adjusted control charts.
- Sensitivity of observed mortality as a test for special cause variation was 0.40.
- Specificity of observed mortality was 0.98, indicating few false positives.
Takeaway
This study looked at how hospitals track death rates after heart surgery and found that both raw and adjusted rates can show important trends.
Methodology
Cross-sectional and longitudinal control charts were used to analyze mortality rates from 41 hospitals over ten years.
Potential Biases
There are risks of distortion of data and processes due to the use of outcome data for quality judgments.
Limitations
The study is limited to New York State's reporting of mortality rates after coronary artery bypass grafting.
Participant Demographics
Data from 41 hospitals in New York State from 1994 to 2003.
Statistical Information
P-Value
0.54
Confidence Interval
0.28 to 0.78
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website