• Politics
  • Diversity, equity and inclusion
  • Financial Decision Making
  • Telehealth
  • Patient Experience
  • Leadership
  • Point of Care Tools
  • Product Solutions
  • Management
  • Technology
  • Healthcare Transformation
  • Data + Technology
  • Safer Hospitals
  • Business
  • Providers in Practice
  • Mergers and Acquisitions
  • AI & Data Analytics
  • Cybersecurity
  • Interoperability & EHRs
  • Medical Devices
  • Pop Health Tech
  • Precision Medicine
  • Virtual Care
  • Health equity

New Model Predicts Which Lung Cancer Patients Will Go to ER


How data could spur cost savings in treating a disease that's expected to cost $14.7B per year.

university of pennsylvania,lung cancer,emergency department,big data,predictive model,natural language processing

Infection, pain management, and other symptoms have a way of pushing lung cancer patients to the emergency department. In fact, according to researchers, the disease causes more ED trips than any other sort of cancer.

But a new predictive model from the Perelman School of Medicine at the University of Pennsylvania promises to better understand which patients will require emergency medicine and when they might come.

“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 early intervention by a clinician,” Jennifer Vogel, MD, the study’s lead author and a radiation oncology resident at Penn, said.

So far, the model anticipated 68 of 207 emergency department visits by lung cancer patients, or 33 percent in total, during a 2-week pilot period, according to Penn.

What’s more, the program identified 131 high-risk patients, of whom 10 percent came to the ER. Just 1.5 percent of the 678 low-risk patients did the same, signaling early success in how the model categorized lung cancer patients, according to Penn.

Vogel and her team created the tool with data from 2,500 patients. Then they validated it with a second set.

Specifically, the model probed electronic medical records for telling comorbidities such as hypertension, liver disease, and cardiac arrhythmia. It then took note of systems like nausea, vomiting, and weight loss and lab results like abnormal platelet count, creatinine, and white blood cell count, according to Penn.

“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,” Abigail T. Berman, MD, MSCE, who heads the Penn Center for Precision Medicine. “It also gives physicians real-time alerts when a patient is deemed to be at high risk.”

While the value of such a tool related to any disease is clear, researchers underlined just how important this predictive model is when it comes to lung cancer.

About 40 percent of lung cancer patients head to the ER at some point, according to Penn. Of that group, 60 percent end up being admitted to the hospital. One report suggests that lung cancer makes up 33 percent of all cancer ED visits.

By 2020, the total cost of lung cancer care in the US could rise to $14.73 billion, according to the National Cancer Institute.

Researchers believe their predictive model could ultimately reduce that number—and burdens on patients and providers.

“Our hope is that triage nurses and physicians will be able to use this information to intervene before an ED visit is necessary,” Berman noted. That could include scheduling preemptive outpatient visits or better managing troublesome symptoms, according to Penn.

But for now, the researchers plan to analyze and sort physicians’ actions and the reason for each ED trip in the pilot phase. Then they intend to improve the model by adding natural language processing elements.

Related Videos
Image: Ron Southwick, Chief Healthcare Executive
George Van Antwerp, MBA
Edmondo Robinson, MD
Craig Newman
Related Content
© 2024 MJH Life Sciences

All rights reserved.