What question did this study set out to answer?

To compare hospital postoperative mortality risk-adjustment models derived from administrative data versus chart-based clinical data and evaluate outlier detection following coronary artery bypass surgery.

January 17, 2005Medical Care

Mortality After Cardiac Bypass Surgery

View Full Paper

Ask AI

Bookmark

Why the study?

Does adding clinical variables to administrative data improve the prediction of postoperative mortality and hospital outlier status in veterans undergoing CABS?

Population

15,288 veterans who underwent isolated coronary artery bypass surgery in 43 VA hospitals between October 1…

Comparison

Risk-adjustment model using administrative data… vs Risk-adjustment models using purely…

Design

Cohort

Follow-up

postoperative

Authors

JGJane M. GeraciBoston University MJMichael L. JohnsonUniversity of Houston HGHoward S. GordonIllinois College

Discussion

Member takes

Implication

May improve hybrid risk models for CABG mortality and outliers in veterans; leaves open generalizability and need for prospective validation.

Key Points

To compare hospital postoperative mortality risk-adjustment models derived from administrative data versus chart-based clinical data and evaluate outlier detection following coronary artery bypass surgery.
Analyzed 15,288 veterans undergoing isolated coronary artery bypass surgery across 43 Veterans Affairs hospitals between October 1, 1993, and March 30, 1996.
Constructed multivariable logistic regression models predicting postoperative mortality using administrative Patient Treatment File codes and clinical Continuous Improvement in Cardiac Surgery Program data.
Calculated hospital observed-to-expected mortality ratios (90% confidence intervals) to evaluate outlier status before and after adding selected clinical variables to the administrative model.
The administrative model achieved a c-index of 0.698, which was significantly lower than the chart-based clinical model discrimination of c = 0.761.
Adding two clinical variables from the chart-based program into the administrative model improved discrimination to c = 0.741.
The clinically enhanced administrative model successfully identified 5 of the 6 high-mortality outlier hospitals flagged by the dedicated clinical model.

Structured PICO

Does adding clinical variables to administrative data improve the prediction of postoperative mortality and hospital outlier status in veterans undergoing CABS?

Population

15,288 veterans who underwent isolated coronary artery bypass surgery (CABS) in 43 VA hospitals between October 1, 1993, and March 30, 1996

Intervention

Risk-adjustment model using administrative data supplemented with selected clinical variables

Comparator

Risk-adjustment models using purely administrative data (PTF) or purely clinical data (CICSP)

Outcome

Postoperative mortality and hospital outlier status (observed-to-expected mortality ratio)hard clinical

Adding a small number of clinical variables to administrative data significantly improves the prediction of postoperative mortality and identification of hospital quality outliers after coronary artery bypass surgery.

Cite This Study

Geraci et al. (2005) studied this question.

synapsesocial.com/papers/6a813e88b06901c4fb22d60c https://doi.org/10.1097/00005650-200502000-00008

View Full Paper

Ask AI

Bookmark

Implication

May improve hybrid risk models for CABG mortality and outliers in veterans; leaves open generalizability and need for prospective validation.

Mortality After Cardiac Bypass Surgery

Discussion

Member takes

Implication

Key Points

Structured PICO

Cite This Study

Mortality After Cardiac Bypass Surgery

Discussion

Member takes

Implication

Key Points

Structured PICO

Cite This Study

Also Consider

Also Consider