Using Patient-Reported Outcomes (PRO) to Improve the Quality of Cancer Care
By Lynne Lederman, PhD
At the American Society of Clinical Oncology (ASCO) Quality Care Symposium, held February 26 to 27 in Phoenix, Arizona, the contribution of patient reported outcomes (PRO) to quality in cancer care was discussed during two breakout sessions. PRO are patient self-reported measures of symptoms, quality of life (QoL), or functional status. Although measures of how patients feel have historically not been included in quality assessment, lately there have been efforts to develop PRO performance measures, including formation of an ASCO workgroup to integrate them into ASCO quality programs.1 These efforts could be considered a response to the Institute of Medicine’s report on cancer care as a system in crisis where care is not patient-centered.2
The need for PRO measures has been demonstrated repeatedly, e.g., in a recent randomized, controlled clinical trial in patients undergoing outpatient chemotherapy for advanced solid tumors. Those who were assigned to report 12 common symptoms to their healthcare providers had a greater improvement in health-related QoL than patients assigned to symptom monitoring at the discretion of clinicians.3 The voice of the patient is particularly important in monitoring therapy-related toxicity because the frequency and severity of adverse reactions are underestimated and underreported by physicians.4,5
PRO tools that are available with applicability to oncology include PROMIS® (Patient Reported Outcome Measurement Information System),6 and the PRO version of the Common Terminology Criteria for Adverse Events (PRO-CTCAE).7 Others are discussed here.
Effect of Risk Perception on QoL
Sarah T. Hawley, PhD, MPH, Cancer Surveillance and Outcomes Research Team, University of Michigan, Ann Arbor, Michigan, presented preliminary results of a study on recurrence risk perception and quality of life after treatment of breast cancer.8 Although the risk of distant recurrence after treatment for early stage breast cancer is very low, women often misunderstand and overestimate their risk. Little is known about how women are given information about their risk or about the effect of overestimation of risk on their QoL.
Objectives of the iCanCare study (P01CA163233) were to use both numeric and descriptive methods to describe the extent of overestimation of risk in women with newly diagnosed early stage breast cancer, assess factors associated with overestimation, and look at the association of overestimation of risk with patient-reported QoL. Patients were identified from those reported to SEER (National Cancer Institute Surveillance, Epidemiology, and End Results) registries in Los Angeles County and Georgia who were treated in 2013 to 2014. The analytic dataset included 1,022 women with ductal carcinoma in situ (DCIS) or low risk invasive (LRI) stage I breast cancer.
Patients were asked to estimate their perceived risk of recurrence; the numeric estimate could be 0 to 100%, the descriptive risk was a choice of very low/low/moderate/high/very high. Overestimation was defined as ≥10% DCIS and ≥20% for LRI, or at least moderate for either disease. QoL was assessed by the PROMIS Global QoL short form; worry frequency (in the last month) about cancer coming back was classified as not at all to always.
In the total population, over a third of patients overestimate their risk of recurrence, and those with DCIS are more likely to overestimate than those with LRI. Patient factors associated with numeric overestimation include being black, and having less education. Being currently bothered by symptoms was associated with both numeric and descriptive overestimation. Worry frequency is also associated with overestimation. Patients who overestimate both numeric and descriptive risk are 10 times as likely to worry frequently as were those who correctly estimated their risk.
There are limitations to this study, including the lack of a pre-diagnosis measure of anxiety, and the small sample size from limited geographic areas. Nevertheless, women with favorable prognoses overestimate their risk of distant recurrence, and this is associated with reduced QoL. This study suggests the need for clinicians to be aware of this overestimation and its effect on QoL metrics and to ensure patients understand actual risk. There remains a need to understand the impact of overestimation on long-term patient behavior.
Dr. Hawley said, “We are exploring some of the patient-reported reasons for indicating why they think their risk is what it is. We are also interested in exploring the impact of overestimation on behaviors such as initial treatments and systemic treatment.” She said that her group is actively developing tools for patients and clinicians to help patients understand the risk of distant recurrence, and clinician-directed tools to provide guidance in how to communicate complicated risk information to patients. “We are also exploring the degree to which clinicians provide patients with resources to help them manage their worry, especially in the survivorship period, and thinking about interventions that can be directed toward that outcome. We also hope that through dissemination of our findings, more attention will be paid to ensuring that patients have an accurate understanding of their actual risk of distant recurrence.”
Using Patient-Reported Outcomes (PRO) to Improve the Quality of Cancer Care (continued)
Mastectomy Decision-Making in Early Breast Cancer
Steven J. Katz, MD, University of Michigan, Ann Arbor, Michigan, presented data about contralateral prophylactic mastectomy (CPM) decision-making from the same iCanCare study.9 CPM has been increasing in women with breast cancer who do not have an increased risk of a secondary primary cancer, and will not improve survival in this subgroup of patients. Outcomes that were included in this analytic sample of 2,469 patients were consideration of CPM, type of surgery, and patient knowledge about treatment.
Of patients receiving CPM, two-thirds believed that CPM had a survival advantage over breast-conserving surgery (BCS), which is not true; one-third believed CPM improves survival for all women. Of patients with average risk for a secondary primary cancer, 28% said their surgeon told them it would improve survival. Patients who preferred their doctor make treatment decisions for them or whose doctor recommended against CPM were much less likely to have CPM. This analysis shows many patients in this population have unrealistic expectations about survival benefits of CPM, and low knowledge about treatment tradeoffs. Deliberation about CPM after diagnosis may be a challenge for surgeons, and requires better communication and patient education if unnecessary procedures are to be avoided.
Roxanne E. Jensen, PhD, Lombardi Comprehensive Cancer Center, Georgetown University, Washington, DC, gave an overview of patient-facing systems in cancer care. Patient-facing systems should be simple and fast to use, allow seamless integration into electronic health records (EHRs), be suitable for multiple uses, e.g., clinical practice, quality improvement, and research, and be engaging for patients and providers (Figure 1).
Ease of use is important, and page layout matters, too (Figure 2). As tools move to mobile devices like phones, a vertical format may be more useful than a horizontal one.
Dr. Jensen and colleagues have reviewed PRO systems in clinical care.10 About two-thirds of these were meant for use during treatment, and a quarter were limited to specific, more common cancers, and measured endpoints specific to those diseases. Most systems were aimed at a physician audience who were most likely to have input into which questions to ask as well as access to reports. For example, providers might select questions about gastrointestinal symptoms because that is what they want to track during chemotherapy, but systems are not designed for patients to choose what symptoms they are interested in following.
Dr. Jensen said, “A big trend I saw with these reports is that about half the patients couldn’t see their scores. There were only 3 systems where caregivers had access and they tended to be those that focused on end of life care or hospice where the caregiver had a more active role with the patient. It was surprising to me that patients weren’t able to see their scores. We are beginning to ask how can we give the information to patients directly.” She said that patients also want to share their information with caregivers so they can have a better picture of how they are feeling. “I get so much enthusiasm for people seeing their reports and scores and using them, I think this will be an exciting area to be developed in the future.”
Dr. Jensen says she observes a lot of patient enthusiasm for filling out forms, but it’s discouraging if patients don’t see anything happening from the hours they spend answering questions. “This goes against what we are trying to do with these systems. This is a developing area of research. Everyone has so much technology now, things that were so hard to do from a research perspective 5 to 10 years ago are now so simple on everyone’s smart phone. We are in a unique opportunity to involve the patients.”
Using Patient-Reported Outcomes (PRO) to Improve the Quality of Cancer Care (continued)
Integrating PRO into EHRs
Collette Pitzen, RN, BSN, CPHQ, MN Community Measurement, Minneapolis, Minnesota, presented several examples of PRO-based tools developed by MN Community Measurement, and how these might be integrated into EHRs. MN Community Measurement, a non-profit organization, collects HIPAA compliant, patient-level data from 1,600 clinics in the state.
One PRO tool that is being developed addresses symptom control during chemotherapy. Ms. Pitzen notes that medical groups are currently being recruited to pilot test this tool.11 This tool is currently being developed for patients with newly diagnosed cancer at the first round of chemotherapy, with an assessment during days 5 to 15 of the first 3 cycles of chemotherapy. This appears to be the proper window for assessment of chemotherapy-related outcomes. The PRO tool is an excerpt of the NCI PRO-CTCAE. The PRO-CTCAE tool has 81 items, but since each was evaluated independently, questions can be selected for a shorter tool. The tool was designed to generate useful data. Symptoms of pain, nausea, and constipation are considered to be under control if the severity is rated as none or mild or <2 (Figure 3).
Ms. Pitzen said that in developing tools, they prefer those in the public domain so they don’t need permission for use. The intent of the use of the PRO tool is two-fold: 1) data from the tool will be collected to calculate retrospective measures of symptom control (e.g., nausea is rated as none or mild), and 2) the tool will be used to assess the patients symptoms in real time during the process of providing chemotherapy care, with the expectation that when symptoms are not in good control, the providers will intervene. However, this intervention is not a part of the data collection strategy or measurement.
Arlene E. Chung, MD, MHA, MMCi, University of North Carolina at Chapel Hill School of Medicine, North Carolina, gave a presentation on the use of capturing patient-generated health data, including PRO associated with chemotherapy, via the use of wearables and tethered portals. These data could be collected on an on-going basis from large numbers of patients using current technology, e.g., smart phones and wristbands, and eventually using newer technology such as smart garments, contact lenses with sensors, implantable or ingestible chips, or tattoos. “There will be an acceptability issue with some of these,” Dr. Chung noted.
She said that new changes in CMS reimbursement, such as monthly payments for remote monitoring of chronic conditions using digitally stored and/or transmitted data should encourage data collection. However, questions remain, such as how to manage the data and make it useful for patients and providers. Just because data can be collected 24 hours a day does not mean that someone is monitoring abnormal results in real time.
Session Chair Ethan M. Basch MD, MSc, Cancer Outcomes Research Program, University of North Carolina, Chapel Hill, North Carolina, said, “We should come away from this discussion excited and frightened.” He added that PRO are likely to be required in clinical practice in the future and that ASCO wants to lead the way in oncology. Key challenges include patient logistics, e.g., how to get patients to participate and what happens if they miss reporting; and how to integrate PRO data into electronic health records as well as retrieve data for use.
The 2017 ASCO Quality Care Symposium will further explore quality measurement and improvement in oncology care, and will include looking at the use and experience with electronic tools to measure and report the patient experience.
- Basch E, Snyder C, McNiff K. et al. Patient-reported outcome performance measures in oncology. J Oncol Practice. Published online April 22, 2014. DOI:10.1200/JOP.2014.001423.
- Institute of Medicine: Delivering high-quality cancer care: charting a new course for a system in crisis. Washington, DC. National Academies Press; 2013.
- Basch E, Deal AM, Kris MG, et al. Symptom monitoring with patient-reported outcomes during routine cancer treatment: a randomized controlled trial. J Clin Oncol. 2016;34:557-565.
- Gravis G, Marino P, Joly D, et al. Eur J Cancer. 2014;50:953-963.
- Laugsand EA, Sprangers MAG, Bjordal K et al. Health care providers underestimate symptom intensities of cancer patients: a multicenter European study. Health Qual Life Outcomes. 2010;8:104-117.
- Hawley ST, Janz N, Jagsi R, et al. Recurrence risk perception and quality of life after treatment of breast cancer. J Clin Oncol. 34, 2016 (suppl 7S; abstr 175).
- Jagsi R, Hawley ST, Griffith KA, et al. Contralateral prophylactic mastectomy decision-making in the population-based iCanCare study of early-stage breast cancer patients. J Clin Oncol. 34, 2016 (suppl 7S; abstr 177).
- Jensen RE, Snyder CF, Abernethy AP, et al. Review of electronic patient-reported outcomes systems used in cancer clinical care. J Oncol Practice. 2013;10:e215-e222.