Using Big Data to Keep Cancer Patients Out of Emergency Departments

“Once we get the alert, we can call the patient ourselves,” says Tracey Evans, MD, who is piloting a program focused on using big data to predict medical needs of lung cancer patients. “We can schedule them for a visit to our clinic. We can recommend more frequent follow-ups or increase the steps they are taking for home care. All of this stems from big data, and the hope is it can help patients out of the emergency room.”

Evans and a team of physicians and data miners from the Perelman School of Medicine at the University of Pennsylvania are working to the power of analytics to better treat those suffering from cancer. Investigators hope to develop a formula that predicts when a patient is likely to end up in the emergencry department (ED) by flagging recent lab tests, radiology visits, or patient-reported symptoms. If physicians can better predict which patients will most likely need urgent treatment, they could take preventive measures, or send the patient to another facility rather than the ED.

Penn Medicine recently launched the Oncology Evaluation Center, which is similar to an urgent care clinic, but for cancer patients instead. “We hope to be able to help patients avoid being admitted to the hospital,” Evans said in a press release. “But even in this case, when admission is required, it’s controlled. This also keeps the patients with doctors [sic] they already know, which provides a more comfortable environment.” Although the predictions are not perfect, Evans said the tool could help physicians identify trends that might otherwise go unnoticed. “Good doctoring is pattern recognition,” Evans said. “So to have all this data here and not use it is ridiculous. It may help us find better ways of doing things that we don’t even know about.”

Unfortunately, when it comes to big data, it is not big enough in the oncology field. Much of what patients with cancer tell their physicians are kept in their notes. Although electronic medical records are making waves across the medical landscape, the notes themselves are usually written in prose, and are difficult for the computer to interpret.

“There are gaps in our data because of free text data or narrative data,” said Peter Gabriel, MD, MSE, chief oncology informatics officer in the Abramson Cancer Center. “It is very challenging for computers to process this kind of data and turn it into computable facts.” Even if the data could be processed, the results would not provide a complete picture to predict ED visits. This is because a crucial part of data is to know which patients have gone to the ED, and which symptoms prompted these visits. “But what if you went to an ER at a different hospital than where you go for your cancer treatments? How would we know? It makes it challenging to get a complete set of data,” Gabriel worries.

Currently, Penn is working to fill these gaps through a partnership with Independence Blue Cross. “By partnering with the insurer, we’re hoping to close the gap on unknown ER visits,” Gabriel said. “That should improve our predictive models for other patients.”