Researchers at the University of Waterloo in Canada found aspects in nursing notes of healthcare providers may help predict whether intensive care unit patients will survive.
To predict the 30-day survival of ICU patients, hospitals use severity of illness scores, including lab results, vital signs and other characteristics collected within 24 hours of admission.
"The physiological information collected in those first 24 hours of a patient's ICU stay is really good at predicting 30-day mortality," said study author Joel Dubin, PhD. "But maybe we shouldn't just focus on the objective components of a patient's health status. It turns out that there is some added predictive value to including nursing notes as opposed to excluding them."
The researchers used a large publicly available ICU database of patient data from 2001 to 2012. The dataset used in the analysis included details on more than 27,000 patients and the nursing notes for these patients. The study was published in PLOS One.
The researchers applied a sentiment analysis algorithm to extract adjectives in the text of the notes and determine whether they were positive, neutral or negative statements. The researchers then used a regression model to examine the relationship between the measured sentiment and 30-day mortality.
The sentiment analysis provided a noticeable improvement for predicting 30-day mortality in the regression model for this patient group. The researchers also found a clear difference between patients with the most positive messages who had the highest survival rates, and the patients with the most negative messages who had the lowest survival rates.
"Mortality is not the only outcome that nursing notes could potentially predict," Dr. Dubin said. "They might also be used to predict readmission or recovery from infection while in the ICU."