Comparing APACHE II Scores for Predicting Hospital Mortality
Author Information
Author(s): Ho Kwok M, Dobb Geoffrey J, Knuiman Matthew, Finn Judith, Lee Kok Y, Webb Steven AR
Primary Institution: Royal Perth Hospital
Hypothesis
Does using admission data for APACHE II scores predict hospital mortality as effectively as using the worst 24-hour data?
Conclusion
Using admission physiological variables to calculate the APACHE II score is as effective as using the worst 24-hour variables for predicting hospital mortality.
Supporting Evidence
- The mean admission APACHE II score was 12.7, while the worst 24-hour score was 15.4.
- The actual hospital mortality was 16.3%, closely correlating with the predicted mortality from both models.
- The overall discrimination ability of the admission APACHE II model was 83.8%, which is considered good.
Takeaway
Doctors can use the first set of health data from when a patient arrives at the ICU to predict how likely they are to survive, just like using the worst data from the first day.
Methodology
This was a retrospective cohort study using prospectively collected data from a tertiary ICU.
Potential Biases
The admission APACHE II model retains weaknesses of the worst 24-hour model, including potential errors from imprecise diagnoses.
Limitations
The study was conducted in a single center, which may limit the generalizability of the results.
Participant Demographics
The cohort included adult patients admitted to a medical-surgical ICU, with a mean age of 53.5 years and a male-to-female ratio of approximately 62% to 38%.
Statistical Information
P-Value
0.0001
Confidence Interval
95% CI = 82.9–84.7
Statistical Significance
p=1.00
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website