Outcome Prediction in Dogs Admitted Through the Emergency Room: Accuracy of Staff Prediction and Comparison with an Illness Severity Stratification System for Hospitalised Dogs—The Acute Patient Physiologic and Laboratory Evaluation (APPLE Fast) Score
Introduction
Patient prognostication can be performed algorithmically using validated scoring systems, or qualitatively by clinician judgement. Predictions of survival in people by ICU staff have previously been documented to be superior to a mortality prediction model.
The objectives of this study were to identify the accuracy with which emergency and critical care (ECC) staff and students were able to predict patient outcome within 24 hours of admission to an emergency room, to compare the accuracy of clinician prognostication with outcome prediction by APPLEfast scoring, and to identify whether experience or mood would affect accuracy.
Methods
ECC staff and students were asked to complete a survey about dogs admitted to a university teaching hospital between April 2020 and January 2021. All clinicopathological data was available for review, and the animals available for examination. Data collected included opinions on whether the patient would be discharged from hospital, as well as a mood score (from 0 very bad to 10 very well), position and experience in ECC. Where data was available, an APPLEfast score was calculated per patient. An APPLEfast score of >25 was used as a predictor of mortality.
Results
One hundred cases were assessed and 192 responses received. Seventy-five dogs (75%) were discharged and 25 dogs (25%) died or were euthanised in hospital. One hundred and forty-two (74.0%) responses predicted the correct outcome. Students, residents, faculty and nurses predicted the correct outcome in 78.1%, 83.3%, 78.2% and 62.0% of cases, respectively. APPLEfast scores were obtained in 65 cases and predicted the correct outcome in 45 cases (69.2%). Mean APPLEfast score was 24.5 (±5.9). There was no difference in outcome prediction accuracy between staff and APPLEfast scores (p=0.74).
For staff, mean experience in ECC was 4.7 years (±4.3). Mean mood score was 7.4 (±1.7). Neither experience nor mood score was associated with ability to predict outcome (p=0.56 and p=0.19, respectively).
Conclusion
Outcome prediction accuracy by staff is not significantly different to APPLEfast scoring where a cut-off of >25 is used to predict mortality.
Disclosures
No disclosures to report.