It’s a matter of regularly answering a few questions about symptoms.
A growing body of research has found that people with cancer who are routinely prompted to answer questions about their symptoms may live months longer and have a higher quality of life than people who don’t track their symptoms as closely.
“Presumably, it’s more than just the check-in,” said health sciences researcher Lisa Barbera at the University of Calgary during a phone interview in August.
“Maybe there’s something about managing the symptoms themselves that is directly beneficial,” she said. “Maybe the management of the symptoms has indirect benefits, like maybe you can stay on chemotherapy longer, or don’t have dose reduction because your side effects are being monitored better.”
Barbera recently led an analysis of data from patients treated for cancer between 2007 and 2015 at Cancer Care Ontario. During check-in, patients are directed to a touch screen where they can answer a short, 16-question survey about their symptoms. Their answers are used and addressed by providers during the visit; they’re also stored and trended over time.
Barbera and her colleagues wanted to know whether routine reporting on symptoms made a difference in outcomes.
In health systems around the world — not just in Canada — this information isn’t usually documented or analyzed. It’s also often not stored with patient’s records, she said.
The researchers compared survival rates among people who had never answered a survey to those who had answered at least one during treatment. They matched patients in each group by age, sex, cancer type and date of diagnosis, ultimately studying more than 120,000 matched pairs.
The results, published in May, were startling. Patients who answered the survey were less than half as likely to have died during the study period than people who never answered the questions.
“There is a strong argument to be made for using these measures in routine care,” said Barbera, who presented the results at the 2019 meeting of the American Society of Clinical Oncology. “But this idea, that we have to change how we do our assessments with patients, is paradigm-shifting.”
Until the past few years, the idea that a patient’s experience could be measured and then inform treatment decisions was seen as being too nebulous to be useful in treating cancer. After all, symptoms seem like a subjective experience.
“Even if we thought [patient-reported outcomes] were important, they were too soft and subjective to do anything useful with,” said Andrea Pusic, chief of plastic surgery and reconstruction at Brigham and Women’s Hospital in Boston, in August. She has used PROs to study patient satisfaction with breast reconstruction after mastectomies.
Broadly, oncology has undergone a general shift toward patient-centered care over the past two decades. In the past two years, research into the use of PROs has surged in oncology. That’s partly because researchers have developed sophisticated new tools to use that information, and partly because of two recent randomized clinical trials, both published in 2017, that suggest PROs have untapped benefits.
One, published in JAMA, was carried out between 2007 and 2011 and led by oncologist Ethan Basch. It involved more than 750 people treated for a range of advanced cancers at Memorial Sloan Kettering Cancer Center in New York City. (Basch is now at the University of North Carolina at Chapel Hill.)
During their treatment, half the patients were asked to answer an online survey every week with 12 questions about their symptoms. (The delivery method is crucial, as some studies estimate that up to half of patients don’t report symptoms to their doctors during visits.)
If a patient’s answers suggested discomfort or another new problem, a provider followed up. The other patients were assigned to a standard care, which meant regular visits with an oncologist.
Basch’s group found that people who reported their symptoms via the online program lived, on average, five months longer than people who didn’t. In the context of cancer treatments, that’s a big deal.
For a new therapy, it would be considered a rousing success: The clinical trial that led to the approval of Yervoy (ipilimumab) to treat metastatic or unresectable melanoma reported a survival benefit of four months associated with the drug. (Yervoy has an estimated cost of more than $250,000 per year.)
The other study, published in the Journal of the National Cancer Institute, involved more than 100 patients treated for advanced lung cancer and was led by oncologist Fabrice Denis at Jean Bernard Center in Le Mans, France.
Roughly half of the study participants tracked symptoms using a Web-based program; the rest did not. Denis and his collaborators reported in their preliminary analysis that patients who monitored symptoms lived, on average, seven months longer than patients who didn’t. Earlier this year, their final analysis confirmed that survival benefit.
Basch hypothesizes that PROs provide a way to detect remissions or treat side effects quickly, without waiting for an appointment or even having to call the doctor.
“People lived longer with PRO monitoring. Relapses were identified earlier. And when they were found early, and intervened on earlier, people lived longer and did better,” he said during an interview in August.
And a longer life isn’t the only benefit researchers have found since then.
“Monitoring PROs improved people’s symptoms and quality of life,” he said. “It decreased use of the emergency room.”
To Basch, the evidence is clear. These technologies pick up symptoms that clinicians will miss, not because of a lack of empathy but because of the limitations of human communication within the large-scale health system.
The challenge, he said, is implementing good PRO tools in practice. “The truth is that it’s hard to do, and can easily be done poorly,” he said
“It’s frustrating that it takes so long to get this into routine practice,” said Sandra Wong, chief of surgery at Dartmouth-Hitchcock Health in New Hampshire. “And practice totally lags behind evidence right now. We have all this great data about how it works, and we know it will take years and years before people put it into practice.”
The gold standard of medical evidence is the randomized clinical trial, like the ones carried out by Basch and Denis. But both of those studies analyzed patient data within a single institution. Achieving similar results in messier, everyday clinical practice can be difficult. Barbera, in Calgary, launched her recent retrospective study as a way to measure the impact of PROs in a real-world setting.
Barbera points to interactions between doctors and patients as a potential challenge to reaping the benefits of PROs.
“Clinicians may underestimate what’s happening, or miss what’s happening, and patients don’t want to volunteer what’s happening,” she said. “They want to be seen as well-behaved patients, good patients.”
Wong agrees. “We rely on the patient to call us when something’s wrong,” she said. “A lot of patients wait too long.”
Pusic, in Boston, notes that doctors aren’t taught how to collect and use information about symptoms from patients. “We underestimate the difficulty of putting that information in the hands of clinicians in a way that they would find useful,” she said. “It’s data they’re not used to dealing with.”
Wong said many institutions around the world have launched new programs to capture patient-symptom data with varied success. And they’re not just in cancer: A range of fields have begun to investigate the best ways to use PROs. “We’re seeing it done a million different ways,” Wong said.