Integrating Patient-Reported Outcomes into Clinical Practice in Oncology

Meg Barbor, MPH

May 2017, Vol 7, No 5 - Quality Care


Orlando, FL—The term “patient-reported outcome” has become ubiquitous in cancer care. Patient-reported outcomes (PROs) are still in the early stages of being integrated into quality assessment programs and routine clinical practice, but engaging patients through PROs can be an invaluable tool for assessing and improving the quality of symptom management, said Ethan M. Basch, MD, MSc, FASCO, Director, Cancer Outcomes Research Program, UNC Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, at the 2017 ASCO Quality Care Symposium.

The working definition of a PRO is “a direct self-report of a symptom, physical functioning, or overall ‘quality of life’ by a patient (or a caregiver),” said Dr Basch. PROs have traditionally been used in drug development but a growing interest in integrating them into clinical practice is now apparent, he added.

PRO-Based Performance Measures

Demonstrating this burgeoning interest is the increased use of PRO performance measures, or performance measures based on PRO data that were aggregated for an accountable healthcare entity (eg, percentage of patients in an accountable care organization whose depression score improved, as measured by the Patient Health Questionnaire-9, a standardized tool to assess depression). A defined pathway with well-established methods can now aid oncology providers in the development of their own PRO performance measures, said Dr Basch.

This pathway begins with the identification of a problem that is best evaluated by a PRO. “If one is interested in understanding the experience of patients with pain, it is patients who are in the best position to self-report that information,” said Dr Basch. Next, providers should identify or develop PRO measures that are meaningful, valid, reliable in the setting of interest, and available for use.

In addition, it is important to ensure that the measure of interest is actionable or changeable based on practice, he stressed. “Although a measure may be important to understand, if it is not actionable, it might not be amenable to assessment in a quality program,” said Dr Basch.

Finally, providers should test the feasibility of data collection, because PRO performance measures differ from the conventional metrics that are used in performance evaluation. “This information is collected directly from patients, so engaging them is a key component of rolling out these kinds of programs,” he added.

Using PROs for Symptom Management

“Symptom management is a cornerstone of cancer care, and an essential function of our jobs,” said Dr Basch. “There are clinical practice guideline recommendations and pathways with best practices for managing numerous symptoms, and there are clear expectations regarding how we should do that,” he added.

Poor symptom management is poor quality care, and to understand symptom management in the context of quality care, providers must go to the source: the patient, Dr Basch emphasized.

In 2014, the American Society of Clinical Oncology (ASCO) convened a PRO Committee that developed outcome and process measure specifications for PRO performance measures in oncology, which may be used to assess future quality within ASCO’s Quality Oncology Practice Initiative.

For example, the outcome measure defined as the percentage of patients who receive moderately or highly emetogenic chemotherapy and moderate or worse nausea 1 week after treatment (±3 days) came directly from the ASCO antiemetics guideline, “but it calls upon patient self-reporting to understand guideline effectiveness in practice,” explained Dr Basch.

Other outcome measure specifications developed by the PRO Committee include the percentage of patients with metastatic cancer and moderate or worse pain, and the percentage of patients receiving systemic cancer treatment who have moderate or worse constipation.

An example of a PRO-based process measure specification is the percentage of patients receiving systemic cancer treatment who provide self-reported symptoms 1 week after chemotherapy (±3 days). These measure specifications are being tested in several contexts, including an ongoing implementation pilot at multiple practices in Minnesota, and a Patient-Centered Outcomes Research Institute–funded ASCO/University of North Carolina project that involves interviews with stakeholders to elicit their impressions of the PRO performance measures in oncology, followed by implementation at 5 sites to test feasibility and collect data for adjustment variables.

PROs are still in the early stages of development, but they are of increasing interest to administrators as a means to avoid unnecessary emergency department and hospital visits as a result of poorly controlled symptoms.

“Good symptom management can decrease utilization and cost, and methods exist to implement PRO programs, so consider this for your practice,” said Dr Basch.