Team develops model to predict ER visits in lung cancer patients

September 27, 2017, Perelman School of Medicine at the University of Pennsylvania
The predictive model was designed by researchers at the Perelman School of Medicine at the University of Pennsylvania with the eventual goal of developing a tool for early intervention that will help patients avoid ED visits. Credit: Penn Medicine

A pilot program that uses big data to predict which lung cancer patients will require a trip to an emergency department (ED) successfully anticipated a third of all ED visits over a two week trial period, and was further able to identify which patients were at high risk and low risk of requiring such care. The predictive model was designed by researchers at the Perelman School of Medicine at the University of Pennsylvania with the eventual goal of developing a tool for early intervention that will help patients avoid ED visits. They will present their data as an oral abstract at the American Society of Therapeutic Radiation Oncology (ASTRO) 2017 Annual Meeting in San Diego (Abstract #2022).

Lung cancer is the most common diagnosis among cancer patients who visit emergency departments, most frequently because of infection, pain management, or other symptoms related to their disease. Roughly 40 percent of will visit the ED during the course of their treatment, and 60 percent of those visits result in hospital admission. In addition, reports have shown lung cancer dwarfs other cancer types in terms of ED visits among cancer patients, making up 33 percent of all such visits according to one recent study. These visits come with a cost for patients - financially and psychologically - as well as for the healthcare system itself. The cost of lung care overall in America is expected to increase to $14.73 billion by 2020, according to the National Cancer Institute.

"The need to be able to anticipate these visits is crucial, but there are very few studies that assess risk factors in a way that allows for by a clinician," said the study's lead author Jennifer Vogel, MD, a resident in Radiation Oncology at Penn.

The model developed by Penn uses patient information pulled from . It identified key comorbidities like hypertension, liver disease, and cardiac arrhythmia. It also flagged specific symptoms like nausea, vomiting, and weight loss, as well as the values of lab results, such as abnormal platelet count, creatinine, and white blood cell count.

"Our model pulls all of this together and weighs each factor to determine a personalized risk for each patient at any given point in time," said senior author Abigail T. Berman, MD, MSCE, an assistant professor of Radiation Oncology at Penn and the associate director of the Penn Center for Precision Medicine. "It also gives physicians real-time alerts when a patient is deemed to be at ."

After developing the model with data from 2,500 patients and validating it with a second set, the researchers put it to use during a two-week . During that time, the model was able to anticipate 68 of the 207 ED visits (33 percent) required by patients. The predictions also showed promise in categorizing patients into risk levels. Of the 131 patients identified as "high-risk", 13 (10 percent) presented to the ED. For the 678 patients grouped as "low-risk", only 10 (1.5 percent) required an ED visit. This demonstrates that the model successfully differentiates between high and low risk patients, as designated as high risk were 6.6 times more likely to visit the ED compared to those designated as low risk.

"Our hope is that triage nurses and physicians will be able to use this information to intervene before an ED visit is necessary," Berman said. These interventions can include reaching out to the patient to preemptively schedule an outpatient visit, taking action to better manage the symptoms that would lead to the ED visit, or other proactive measures. Researchers say the next step is to categorize the reasons for each ED visit and the actions physicians took during the pilot phase. They also plan to incorporate natural language processing elements into the model in order to improve its predictive value.

Explore further: Cancer patients may have undiagnosed depression

Related Stories

Cancer patients may have undiagnosed depression

September 25, 2017
(HealthDay)—Depression is common, though often overlooked, in people with cancer, a new study suggests.

Many cancer patients' Emergency Department visits appear preventable

May 30, 2017
As many as 53 percent of cancer patients' Emergency Department visits that do not require admission could be avoided with better symptom management and greater availability of outpatient care tailored to their needs, according ...

Health system sees success with e-visits via patient portal

June 8, 2017
(HealthDay)—Patient portals can successfully offer access to physicians without office visits, according to a report published online May 30 by the American Medical Association.

Single home visit significantly improves adherence, reduces exacerbations in patients with severe asthma or COPD

October 18, 2016
A single home visit to patients with severe asthma or COPD may significantly improve patient adherence with office visits and inhaler use and may reduce severe exacerbations requiring emergency department visits.

Older lung cancer patients face significant treatment burden

January 6, 2017
Depending on the type of treatment older lung cancer patients receive, they can spend an average of one in three days interacting with the healthcare system in the first 60 days after surgery or radiation therapy, according ...

Among all cancers, lung cancer appears to put patients at greatest suicide risk

May 23, 2017
A lung cancer diagnosis appears to put patients at the greatest risk of suicide when compared to the most common types of non-skin cancers, according to new research presented at the ATS 2017 International Conference.

Recommended for you

A code for reprogramming immune sentinels

December 10, 2018
For the first time, a research team at Lund University in Sweden has successfully reprogrammed mouse and human skin cells into immune cells called dendritic cells. The process is quick and effective, representing a pioneering ...

Potential seen for tailoring treatment for acute myeloid leukemia

December 8, 2018
Advances in rapid screening of leukemia cells for drug susceptibility and resistance are bringing scientists closer to patient-tailored treatment for acute myeloid leukemia (AML).

Study may offer doctors a more effective way to treat neuroblastoma

December 7, 2018
A very large team of researchers, mostly from multiple institutions across Germany, has found what might be a better way to treat patients with neuroblastoma, a type of cancer. In their paper published in the journal Science, ...

Major breakthrough in quest for cancer vaccine

December 6, 2018
The idea of a cancer vaccine is something researchers have been working on for over 50 years, but until recently they were never able to prove exactly how such a vaccine would work.

'Chemo brain' caused by malfunction in three types of brain cells, study finds

December 6, 2018
More than half of cancer survivors suffer from cognitive impairment from chemotherapy that lingers for months or years after the cancer is gone. In a new study explaining the cellular mechanisms behind this condition, scientists ...

Scientists develop new technology for profiling unique genetic makeup of myeloma tumor cells

December 6, 2018
Cancer arises when cells lose control. Deciphering the "blueprint" of cancer cells—outlining how cancer cells hijack specific pathways for uncontrolled proliferation—will lead to more efficient ways to fight it. Joint ...

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.