Consider a man—we'll call him Mr. Jones—newly diagnosed with advanced lung cancer. In his first month, he starts chemotherapy. Within days, his nausea is overwhelming. He calls his clinic and gets the lunch hour voicemail: "If this is an emergency, call 911." He tries his anti-nausea medications but can't keep them down. So he drives himself to the emergency department. From there, Mr. Jones is admitted to the hospital, where he remains for 3 days so his doctors can stabilize the effects of his nausea and dehydration—a debilitating conclusion to a life-altering month.

We have seen this happen all too often. We can do better.

Acute hospitalizations are the single largest driver of differences in how doctors care for patients with cancer and account for nearly half of spending on cancer care. Many hospitalizations—like Mr. Jones'—are  preventable. And once discharged, a patient is often admitted again: more than one in four patients having cancer treatment are readmitted with 30 days.

The Center for Medicare and Medicaid Services (CMS) recently proposed a new measure that seeks to reduce hospitalizations for cancer patients getting chemotherapy. In short, if a patient receiving chemotherapy is hospitalized for one of 10 conditions (including nausea and fever) within 30 days of treatment, the hospital's Medicare reimbursement could suffer. However, no one knows the best approaches to prevent unnecessary hospitalizations.

We published a study today that takes an exhaustive look at ways to reduce unnecessary hospitalizations for patients with cancer.  We examined guidelines, demonstration projects, and dozens of studies. Based on our review, we propose five strategies that represent state-of-the-art cancer care.

  • First, we should identify patients at high risk of preventable hospitalizations to ensure that the patients with the highest need get the most support. Big data approaches may be able to determine which patients are at the highest risk for complications from treatment, prompting early intervention.  For example, by using predictive analytics to risk-stratify patients with heart failure, one group reduced hospital readmissions by nearly 20%.
  • Second, we should enhance access and care coordination so that patients can get the care they need when they need it. Technology has great potential in this area. Telemonitoring and other 'connected health' approaches may be helpful to monitor patients closely when they are at home, catching potential problems earlier. For example, a group of researchers found that the simple act of prompting patients getting chemotherapy to report their symptoms to their doctors electronically between visits helped patients not only feel much better and require fewer hospitalizations, but also live longer.
  • Third, we should standardize clinical pathways so that patients get the best treatment for their symptoms every time. One group learned that using standardized treatment protocols for common symptoms within the practice could reduce preventable hospitalizations by half.
  • Fourth, we should develop urgent cancer care tactics, which will allow us to become better at addressing the urgent needs of patients that arise outside of the standard office visit. One medical group adjusted its clinic schedule to create dedicated daily appointment slots for urgent visits. This simple intervention cut hospitalizations for symptom management by nearly a third.
  • And lastly, we should employ early palliative care, which has been shown to improve both quality and duration of life. One study found that patients with end-stage lung cancer who received early palliative care had less depression, reported higher quality of life, and lived significantly longer, despite receiving less aggressive end-of-life care.

While these may seem simple and straightforward, health care systems have only recently begun to appreciate how important even small changes can be. Imagine Mr. Jones' care if he had access to text message tools that would have allowed him to report symptoms sooner—and allowed his care team to know he was at higher risk of needing emergency care. With this proactive approach, he could have had access to stronger medications to manage his nausea sooner, dictated by an evidence-based care pathway. And if he still needed emergency care? He could have gone to his oncologist's office if the schedule had spots reserved for such visits—or even an urgent care center with particular expertise in treating cancer patients—rather than the general emergency department.

In short, the cancer care system could have anticipated Mr. Jones' needs more effectively. This isn't just about reducing unnecessary hospitalizations—it's about listening to our patients, trying to foresee their needs, and making what is inevitably a difficult period a little more tolerable.

Nathan Handley is a medical oncology fellow at Penn Medicine and student at The Wharton School of The University of Pennsylvania. Lynn Schuchter is Chief of the Division of Hematology-Oncology at The University of Pennsylvania. Justin Bekelman is a radiation oncologist and leads a research program in cancer care innovation at the Abramson Cancer Center at the University of Pennsylvania.