Skip to contents

Data from a prospective cohort study of triage scoring for an emergency department (ED). The study examined whether the use of patient level measurements would improve an existing triage score. These data were used as a test set (called validation in the manuscript) to examine the performance of the model built using the training (primary) cohort. Some variable names have been changed for readability and for consistency with the primary dataset, but the data on 18 variables for the 6,383 participants are otherwise unchanged. Some variables in the primary dataset do not appear in these data.

Usage

danish.ed.validation

Format

A tibble with 6383 rows and 18 variables:

mort30

numeric, 1 if patient died within 30 days of admission, 0 otherwise

triage

factor, triage score given at arrival to ED. Values blue, green, yellow, orange, red, from lowest to highest priority for treatment. The value blue normally denotes severity not warranting admission to the ED. Participants coded blue are in these data but not in the primary data.

age

numeric, age in years, rounded to lower integer

sex

factor, female, male

albumin

numeric, serum albumin, in g/L

creatinine

numeric, serum creatinine, in umol/L

hemaglobin

numeric, serum hemaglobin, in mmol/L

potassium

numeric, serum potassium, in mmol/L

leuk.count

blood leukocyte count, in 10E9/L

sodium

numeric, serum sodium, in mmol/L

c.react.protein

numeric, serum C-reactive protein

oxygen.sat

numeric, peripheral arterial oxygen saturation, %

resp.rate

numeric, respiratory rate per minute

heart.rate

numeric, heart rate, beats/min

systolic.bp

numeric, systolic blood pressure, in mmHg

readmit.hosp

factor, readmitted to hospital within 30 days, with values yes, no

days.in.hosp

numeric, number of days admitted to hospital

icu.status

factor, patient admitted to ICU, with values yes, no

References

Kristensen, Michael, et al. "Routine blood tests are associated with short term mortality and can improve emergency department triage: a cohort study of> 12,000 patients." Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine 25 (2017): 1-8. https://sjtrem.biomedcentral.com/articles/10.1186/s13049-017-0458-x?report=reader