Predicting Hospital-Acquired Infections with a Scoring System
Author Information
Author(s): Chang Ying-Jui, Yeh Min-Li, Li Yu-Chuan, Hsu Chien-Yeh, Lin Chao-Cheng, Hsu Meng-Shiuan, Chiu Wen-Ta
Primary Institution: Taipei Medical University
Hypothesis
Can a scoring system based on simple parameters effectively predict hospital-acquired infections (HAI)?
Conclusion
A scoring system was developed to predict HAIs using simple parameters, which can help identify high-risk patients during hospitalization.
Supporting Evidence
- The scoring system showed excellent discrimination with an AUC of 0.964 in internal validation.
- Seven significant variables were identified for predicting HAI.
- External validation with 2,500 patients confirmed the scoring system's effectiveness.
Takeaway
Doctors created a simple way to tell if patients might get sick from infections while in the hospital, using just a few important signs.
Methodology
The study used Logistic Regression and Artificial Neural Networks to analyze data from 476 HAI patients and 1,376 non-HAI patients.
Potential Biases
The study did not evaluate human and environmental factors that could contribute to HAIs.
Limitations
The study only included patients aged 16 to 80 and pooled data from ICUs and non-ICU wards, which may overestimate infection probabilities.
Participant Demographics
The mean age of participants was 55.29 years, with 48.2% being female.
Statistical Information
P-Value
<0.0001
Confidence Interval
95% CI: 1.6–2.29
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website