This is the seventh article in a series of stories related to cancer communications. You can read more articles in the series here.
Straight from the Source: Electronic Patient-Reported Outcomes
Patients know what they feel. Whether they are reporting pain caused by cancer or side effects of treatment, a patient's own account of symptoms and other experiences provides a unique perspective. In recognition of this fact, researchers and health care providers increasingly are trying to capture information by patient self-reporting. A 2006 consensus conference at NCI, for example, concluded that a patient's own account of subjective symptoms, such as pain or fatigue, should be routinely collected in clinical research.
For years, researchers have been trying to design electronic systems for collecting patient-reported outcomes (ePRO) that can help improve clinical care as well as clinical research. The hope has been that allowing patients to report distressing symptoms to their health care providers more frequently by computer or phone will enhance communication and decision making between patients and providers.
"If we can use this type of system to stay more acutely attuned to what patients are experiencing, we believe we can both improve patient outcomes and reduce medical care costs by minimizing the occurrence and impact of adverse events during treatment," explained Dr. Bryce Reeve, an associate professor of health policy and management at the University of North Carolina at Chapel Hill.
Preliminary trials examining whether ePRO systems could improve quality of life for patients showed little benefit from electronic monitoring, recounted Dr. Amy Abernethy, associate professor of medicine at Duke University.
"People tried back in the '90s to test some rudimentary electronic quality-of-life monitoring tools, but the problem was that most of them were designed by researchers alone and didn't fit into the way clinical care works," she explained. "But patients have always reported their symptoms and side effects in clinical care, it just wasn't electronic. Now these two pieces are coming together, and we're starting to see electronic patient-reported monitoring that fits into clinical care."
More Talking, Faster Healing?
Two recent randomized trials that tested ePRO systems showed promising improvements in patient care and clinicians' willingness to use the systems. The studies, published in the Journal of Clinical Oncology, also underscore some of the challenges associated with developing and implementing these systems.
In one trial, researchers at the Dana-Farber Cancer Institute and their colleagues tested a system called the Electronic Self-Report Assessment–Cancer (ESRA-C) in 660 patients who were beginning chemotherapy or radiation therapy. The participants completed the ESRA-C questionnaire twice: once before starting treatment and once 4 to 6 weeks later, during treatment.
Although all of the patients completed the questionnaires, the researchers randomly assigned the patients to two groups: half of the patients had their questionnaire answers seen by their health care team, and the other half had their questionnaire answers set aside without their providers seeing the answers. The researchers audio recorded the patients' clinic visits after completion of the second questionnaire to see if receipt of an ESRA-C report increased the discussion of symptoms and treatment side effects.
Overall, doctors who received an ESRA-C report were 30 percent more likely to discuss symptoms that were above a set threshold of severity, including those that related to sexuality and social functioning, which are often awkward for patients and physicians to broach. Patients and doctors still tended to discuss common side effects often, regardless of whether they were of concern to individual patients.
In the second trial, researchers from the University of Texas M. D. Anderson Comprehensive Cancer Center recruited 100 patients to test an automated telephone system for reporting distressing side effects. All patients had undergone thoracotomy for lung cancer or lung metastases, a surgical procedure that can cause many severe symptoms during recovery.
All patients received automatic calls from the system about twice a week for 4 weeks. Before the study began, patients were randomly assigned to either an intervention group or a control group. In the intervention group, if any symptom recorded during the calls exceeded a preset threshold for severity, the system generated an e-mail alert to the patient's medical team. For the control group, symptoms were recorded but no alerts were generated. During the study, severe symptoms declined an average of 19 percent in the intervention group compared with an average of 8 percent in the control group. Moreover, symptoms declined more rapidly in the intervention group.
On a more worrisome note, 16 percent of alerts in the intervention group went unanswered by the health care team. "We've always been concerned that a lot of patient issues are not getting heard," said Dr. Abernethy. At least with ePRO systems, she continued, the clinic knows how many patient requests do not receive a timely response.
Integration in the Real World
"Patients are willing, able, and enthusiastic to provide this information. The real challenge is going to be the integration of this information into clinical work flow," commented Dr. Ethan Basch, an oncologist and patient-reported outcomes researcher at Memorial Sloan-Kettering Cancer Center. "The clinic is very busy, and it can be complicated to introduce new information, particularly real-time reporting of patient data that might be showing up 24/7 if patients are reporting from home."
One selling point for these systems may be their ability to redirect information before it even reaches the oncologist; for example, to send pain management issues straight to an oncology nurse or to have signs of depression trigger a call from the clinic's mental health staff. Such automated distribution could lighten the health provider's clinical burden instead of increasing it, explained Dr. Abernethy.
"Over the past 5 years, the interest in ePRO has grown explosively, and there's a broader movement toward patient-centered care models, where the perspective of the patient is being integrated into care at every level," added Dr. Basch. "With the upsurge in interest from clinicians, it's become easier to integrate this information into the clinic."
Patients need to understand that these tools are meant to enhance, not replace, standard interactions between patients and their care providers, explained Dr. Reeve. This applies particularly to emergency situations. "If patients are experiencing highly distressing symptoms, they should call their doctor or nurse directly or call 911," he stressed, "not log into an ePRO system to report them."
As these systems become more widely used, larger multicenter clinical trials will likely be needed to determine whether this additional communication about the patient experience does indeed translate into better health outcomes for patients, Dr. Reeve explained. Many hospitals are developing customized ePRO systems, which may complicate research studies that seek to combine data across clinics.
However, Dr. Basch does not see this challenge as insurmountable. "We are already seeing big efforts to standardize health care terminology. A number of people around the country and internationally are really interested in this subject, and we're all aware of each other's work," he explained. "We're always thinking about how we can link our data and questionnaires." (See the box below.)
Standards for ePRO data will be important as these systems become more common and undergo more rigorous testing to ensure that they are helping patients, according to Dr. Abernethy.
"While we know that research-designed ePRO systems without attention to the needs of the clinic are going to fail, clinically designed systems without adequate attention to the needs of research would also be inadequate for the future of this field," she concluded.