Just a few years ago, few had heard of the term “patient-reported outcome” or “PRO.” Now, PROs seem to come up frequently in discussions about almost every aspect of oncology—in clinical care delivery, clinical trials, quality assessment, and comparative effectiveness research (Fig. 1 shows uses and benefits of PROs in each of these contexts).
PROs are reflections of how patients feel and function—for example, symptoms related to disease or toxicities, physical functioning, or quality of life. They are measured via questionnaires that are rigorously developed and tested to assure that the questions are clear, that they are measuring what we think they are measuring, that they are reliable, and that scores change as we might expect.
PROs in Routine Cancer Care
Interest is burgeoning to integrate systematic collection of symptoms from patients during routine cancer clinical care. This is generally done using electronic “ePRO” software that enables patients to self-report via the Internet, automated telephone systems, or downloadable applications. Patients can report during visits via iPads or self-service computer kiosks, and between visits using their own devices. Reports showing longitudinal symptom trajectories in tables or graphs can be viewed by clinicians at visits, and email alerts can be triggered in real-time to nurses anytime a severe or worsening symptom is self-reported.
Most of the major electronic health record (EHR) vendors can support this function to some extent. Although most systems are still rudimentary, they are advancing with each software update. Most commonly, ePROs are collected through the EHR’s patient portal, which can push email invitations to patients asking them to complete online questionnaires. These systems require patients to be signed up for the portal, generally include unattractive questionnaires, and don’t provide intuitive ways for clinicians to view the PRO information in the medical record. However, local customizations are possible to improve these functions, or a third-party ePRO system can be licensed to avoid these limitations. It is expected that all of these EHR functions will rapidly improve.
Multiple studies have been conducted (see references below) that tested whether it is feasible to integrate PROs into routine cancer care (it is), and whether outcomes are improved as a result (they are).
Feasibility: In general, when patients receiving active cancer treatment or following cancer surgery are offered an electronic system to self-report symptoms, 80% to 85% will regularly provide this information, even over long periods of time, and even when patients are elderly, ill, or have end-stage disease. Different approaches have been used, including the Internet, automated telephone systems, or downloadable applications. For example, the province of Ontario, Canada, has installed self-service kiosks across their cancer centers for symptom self-reporting. It is essential with these kinds of programs to communicate well with patients and explain that this is an important part of care, with reminders and calls to patients who don’t report on schedule. Clinician buy-in is also key, and implementation plans should be developed with input from physicians, nurses, and administrators.
Benefits: In multiple prospective studies, integrating ePROs into routine cancer care has improved patient–clinician communication, patient satisfaction, symptom management and control, and quality of life. Nurses find symptom email alerts to be useful and actionable. Recent data have found reductions in emergency room visits, longer tolerability of chemotherapy, and improved survival (Fig. 2). This evidence of benefits has led to growing interest in this approach.
Barriers: The main barriers to widespread adoption of PROs in routine oncology care are largely logistical and include technology configuration, integration into existing workflow, and engagement of patients and clinicians. Buy-in of local leadership and clinicians is vital to successful implementation.
PROs in Clinical Trials
Drug development trials are perhaps the most well-traveled context in which PROs have been used in oncology. Many of the PRO questionnaires in use today were developed for use in clinical trials. The U.S. Food and Drug Administration (FDA) has provided guidance on methods for developing and using PROs in product development and has indicated that benefits in PROs can be the basis for full drug approval and that PRO endpoints can be an essential adjunct in trials with progression-free survival primary outcomes or non-inferiority designs.
Recently, the FDA has encouraged the use of PROs to assess three complementary but distinct areas in cancer clinical trials specifically: disease-related symptoms (e.g., pain related to metastases), physical functioning (e.g., the ability to conduct activities of daily life), and symptomatic adverse events (e.g., measured by the National Cancer Institute’s new PRO version of the Common Terminology Criteria for Adverse Events). The June 6 Education Session “Integrating Patient-Reported Outcomes Into Cancer Clinical Trials and Regulatory Review” will highlight these efforts (visit the ASCO iPlanner on the Attendee Resource Center [am.asco.org/arc] for session time and location information).
Understanding patients’ symptom and physical functioning experience is vital to appreciating the properties of cancer treatments, and PRO questionnaires are the gold standard for assessing these areas. For example, in early-phase trials, patient symptoms and physical functioning can reflect tolerability and inform decisions about dosing. In phase III trials, PROs can be useful to understand the impact of therapy on disease-related symptoms, such as pain or fatigue.
Ideally, the PRO component of a clinical trial will be planned early. This starts with identifying outcomes that are meaningful in a given context or population via literature reviews or qualitative (patient interview) work, then choosing or developing a rigorous questionnaire that maps to those outcomes. Investigators should ideally partner with an expert in PRO methods to assist with the design. Increasingly, there are efforts to standardize approaches to PROs in various contexts side by side with other endpoints. For example, the Prostate Cancer Working Group has recommended standards for outcomes in clinical trials, including biomarkers, imaging, and PROs.
PROs in Quality Assessment
An area of rapid development is the use of PROs to measure quality of care delivery. Other specialties are ahead of oncology. In orthopedics, for example, pain, mobility, and physical functioning are increasingly assessed via PROs postoperatively to assess performance at the practice or provider level. Most PRO-based quality metrics assess control of symptoms or physical functioning, which differentiates them from patient satisfaction or experiential questionnaires that assess processes such as timely returning of telephone calls by clinicians.
ASCO has a PRO committee that is developing and testing PRO measures, which, in the future, may be used to assess quality within ASCO’s Quality Oncology Practice Initiative (QOPI®). ASCO is following standards for PRO-based quality measures established by the National Quality Forum and is drawing from ongoing outside initiatives. For example, PRO measures to assess quality of oncology care are currently being piloted across the state of Minnesota, and ASCO is a partner in these efforts.
An example of a PRO-based quality measure is the proportion of patients in a practice with metastatic disease receiving systemic therapy who experience moderate or worse pain. Similar measures might be used for other symptoms, such as nausea or constipation. Key considerations when developing PRO-based quality measures include meaningfulness to patients and responsiveness to changes in clinician behavior (“actionability”).
PROs in Quality Assessment
Studies designed to understand treatment benefits and harms in real-world contexts often aim to understand the patient experience via PROs. Registries and pragmatic trials increasingly integrate PROs, and PROs have been strongly encouraged by the major U.S. funder of comparative effectiveness research (CER), the Patient-Centered Outcomes Research Institute. For example, arthralgias with aromatase inhibitors (and resulting drug discontinuation) and urinary/sexual symptoms following prostatectomy have been characterized through prospective registry studies.
As shown in Figure 1 (see page 31C), there is synergy between the various roles of PROs: If PROs are routinely collected in practice, they can be used for clinical care of individual patients and also aggregated across patients to assess quality and conduct CER. A particular concern when collecting PROs in real-world settings outside of clinical trials is missing data from those unwilling or unable to self-report. Therefore, efforts to engage patients and use of backup data-collection strategies are essential to assure data completeness (e.g., calling people who miss reports, which can boost response rates by 10% to 15%).
PROs uniquely convey the patient experience with symptoms and physical activity and are key to understanding the impact of cancer treatments. Collection of PROs during routine cancer care can inform and improve symptom management. In clinical research, PROs can characterize the relative benefits and harms of therapies. Finally, in quality assessment, PROs can reflect patient-meaningful impact of care delivery. There is rapidly growing interest to broaden the use of PROs across these oncology contexts, and we are likely to see PROs used in practice and research increasingly in the coming years.