Big data model improves prediction of key hospital outcome

February 18, 2016 by Ziba Kashef
Big data model improves prediction of key hospital outcome
Credit: Shutterstock/Yale University

More than half of hospital deaths in the United States are related to severe infections, or sepsis. Yale researchers developed a prediction model, drawing on "big data" about local patients and using machine-learning methods, that proved better at identifying at-risk patients than existing clinical practices.

Currently, emergency physicians can use simple calculators or point- scoring systems known as clinical decision rules to determine which hospitalized might die from sepsis. However, these methods often fail to identify patients most at risk because they are based on limited information, derived from models that are unable to capture the complexity of data, and developed using different patient populations.

The new model developed by researchers at Yale School of Medicine and the University of Washington uses a large amount of data collected in of local patients. Known as random forest modeling, the method draws on and learns from patient data to make predictions. Their big-data method outperformed existing models and has the potential to correctly classify an additional 200-300 patients per 5,000 with .

"By using the machine-learning technique and incorporating a large amount of variables—over 500—we created a model that has the potential to better predict sepsis mortality in patients admitted to the hospital," said Dr. R. Andrew Taylor, assistant professor of and first author on the study. Having proved the concept, Taylor and his colleagues hope to test the model in real-time with patients in New Haven while also promoting the big-data approach. "Our goal is to make patients' data work for them and create learning health systems where predictive models are developed and ultimately applied to improve patient care," he said.

The study published on this month in Academic Emergency Medicine.

Explore further: Electronic healthcare records data reveal factors linked to emergency department revisits

More information: R. Andrew Taylor et al. Prediction of In-hospital Mortality in Emergency Department Patients with Sepsis: A Local Big Data Driven, Machine Learning Approach, Academic Emergency Medicine (2015). DOI: 10.1111/acem.12876

Related Stories

Electronic healthcare records data reveal factors linked to emergency department revisits

February 12, 2016
A new study has identified distinctions in patient diagnoses and different patterns of Emergency Department usage between individuals who are more or less likely to return to the ED for care within a 72-hour period. These ...

Advances made against the deadly infection complication, sepsis

October 21, 2015
Sepsis is an inflammatory response to infection that's known to develop in hospital settings and can turn deadly when it's not discovered early on. In a new study, a hospital surveillance program focusing on reducing the ...

New depression model outperforms psychiatrists

January 21, 2016
A simple questionnaire filled out by depressed patients can help identify whether a particular medication is likely to help them, according to a new study by Yale University researchers published Jan. 20 in the journal Lancet ...

Sepsis and shock response team in the ED reduces mortality

October 19, 2015
Researchers from Mayo Clinic in Jacksonville, Florida, formed a multidisciplinary sepsis and shock response team (SSRT) to help alert emergency department providers when these disorders are suspected. An automated electronic ...

Sepsis tied to higher post-discharge mortality risk

February 9, 2016
(HealthDay)—Sepsis survivors have a substantially increased risk of all-cause mortality and major adverse cardiovascular events after discharge, according to a study published online Jan. 25 in the American Journal of Respiratory ...

High volume of severe sepsis patients may result in better outcomes

January 17, 2014
A recent study led by Boston University School of Medicine (BUSM) shows that "practice may make perfect" when it comes to caring for patients with severe sepsis. The study showed that patients admitted to academic medical ...

Recommended for you

High-fat diet in pregnancy can cause mental health problems in offspring

July 21, 2017
A high-fat diet not only creates health problems for expectant mothers, but new research in an animal model suggests it alters the development of the brain and endocrine system of their offspring and has a long-term impact ...

To combat teen smoking, health experts recommend R ratings for movies that depict tobacco use

July 21, 2017
Public health experts have an unusual suggestion for reducing teen smoking: Give just about any movie that depicts tobacco use an automatic R rating.

Why sugary drinks and protein-rich meals don't go well together

July 20, 2017
Having a sugar-sweetened drink with a high-protein meal may negatively affect energy balance, alter food preferences and cause the body to store more fat, according to a study published in the open access journal BMC Nutrition.

Opioids and obesity, not 'despair deaths,' raising mortality rates for white Americans

July 20, 2017
Drug-related deaths among middle-aged white men increased more than 25-fold between 1980 and 2014, with the bulk of that spike occurring since the mid-1990s when addictive prescription opioids became broadly available, according ...

Aging Americans enjoy longer life, better health when avoiding three risky behaviors

July 20, 2017
We've heard it before from our doctors and other health experts: Keep your weight down, don't smoke and cut back on the alcohol if you want to live longer.

Parents have critical role in preventing teen drinking

July 20, 2017
Fewer teenagers are drinking alcohol but more needs to be done to curb the drinking habits of Australian school students, based on the findings of the latest study by Adelaide researchers.

0 comments

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.