Abstract: Health systems are evolving to make the patient the central focus of health care delivery. This has resulted in increased use and interest in patient-reported outcomes (PROs) to include the patient’s perspective into research and in clinical practice. However, the wide variability in PRO measurement practices presents a dilemma to all who want to embrace this evolution. In this article, the authors share their experiences with implementing PRO collection and discuss the key considerations for success, including appropriate measure selection, achieving buy-in from health care providers, and utilizing PRO data in clinical practice.
As the field of medicine evolves, emphasis on providing patient-centered care has been increasing. In its 2001 report, the Institute of Medicine made the argument that patient-centered care—that is, respecting and responding to individual patient preferences, needs, and values, and ensuring that patient values guide all clinical decisions—is essential to the delivery of high quality care.1 Since then, many have highlighted the importance of patient-centeredness for achieving care quality and value.2-5
The greater focus on patient-centered care has been accompanied by increasing interest in using patient-reported outcomes (PROs) in the care delivery process. PROs represent the patient’s report of a health condition and its treatment without input or bias by any third party.6 Demographics, medications, and past medical and family history, although reported by the patient, do not qualify as PROs because they are simply facts and do not capture a health-related outcome. PRO data may include such items as disease and/or treatment symptoms, function, and health-related quality of life (QOL).
PROs come directly from the patient themselves and thus represent an important voice in health care decision-making. The collection of PROs places the patient in the center of the care delivery process. The use of PROs has been shown to improve communication between provider and patients, enhance patient well-being, and improve the perception of continuity of care. PROs also have been shown to be prognostic for improved survival in patients with advanced cancer.7-9
Despite the interest in PRO use in clinical practice, several questions and dilemmas remain for those who seek to implement PROs in their practices. These include how to use and interpret measures, which measures/instruments to use, and how to integrate PROs into the clinical workflow in ways that enhance rather than detract from patient care.
To address questions such as these, the User’s Guide to Implementing Patient Reported Outcomes in Clinical Practice was first published in 2011 to assist clinicians with PRO implementation.6 A key recommendation of these guidelines is to first and foremost define the goal of PRO collection up front. This is of great importance because PRO collection may serve several purposes. For example, is the data to be used purely for patient care, or will it also be used for research purposes? Will the data be used to screen for symptoms of disease or treatment, monitor response to treatment, enhance communication between the team and the patient, or some other purpose? Are patients to have access to the PRO data in addition to the clinicians which may support a goal of patient-centered care? Is PRO data to be used to improve the quality of care provided? Determining short- and long-term goals for PRO collection is the foundational step to implementing PRO measurement in clinical practice.10
In the hematology oncology practice at Mayo Clinic, we aimed to improve utility in clinic by obtaining data from the patient prior to the appointment, that informs about possible toxicity and/or disease progression, and to detect any psychosocial or functional issues that may otherwise not be discussed in routine visits. In this article, we share our experiences and discuss the key considerations for incorporating PROs into clinical practice.
Selecting a Pro Measure
Unfortunately, there is no consensus on the best PRO measurement to employ. This quandary reflects the vast heterogeneity of PRO measures that are currently utilized as well as the complexity of care. Keeping in mind the intended goals of PRO collection, as discussed above, the PRO measure selected should be valid, reliable, and easily administered and interpreted.10 Instruments may be disease-specific or generic, single- or multi-item, and static or dynamic; these variables need to be considered before selecting a PRO measure. Additionally, survey fatigue on the part of the patient is an essential consideration when selecting the best instrument to capture a specific outcome.
Defining which patient population will complete PRO instruments is also important, because this will drive resource requirements, including personnel, space, and technology, and will shape potential barriers to data collection.10 The clinician should define if the instrument will be used by adult patients, pediatric patient, or both; whether only patients with a specific disease or illness will be included; whether instruments used will require validated translated instruments or language proficiency will limit widespread use; and whether data collection will occur in an inpatient or outpatient setting, and the logistics associated with each. Ideally, the PRO instruments should be completed by the patients themselves, but this may not always be feasible depending on patient age, language proficiency, health literacy, or other issues. When a PRO instrument is completed by a proxy, such as a spouse or parent, this should be clearly indicated. In addition, the clinician should keep in mind that those responses may reflect the perceptions of the person filling out the instrument rather than the patient themselves.
In the hematology oncology practice at Mayo Clinic, it is standard for all patients to answer a 3-item questionnaire to rate their overall fatigue, pain, and QOL for the past week prior to their appointment. This patient-reported symptom screen uses the Linear Analog Self-Assessment (LASA) scale, a validated measure of QOL and other domains.11 On the LASA scale, patients assign each measure a score of 0 to 10 points, with a score 0 representing the best outcome for pain and fatigue measures and a score of 10 representing the best outcome for the overall QOL measure. Clinically meaningful deficiencies have been shown when patients’ scores are ≥ 5 for pain and fatigue or ≤ 5 for QOL.12
Pro Data Collection Method
The methods by which PRO data are collected also play a significant role in how the data are incorporated into clinical practice. The use of paper or computerized format—in the form of desktop or mobile devices—should be individualized to each practice’s preference, as should the logistics of how, when, where, and to whom the PRO results will be presented. Paper-based collection of PROs can be a lower cost method of obtaining data; this is our current method with the symptom screen. Patients complete a paper-based questionnaire, and values from the PRO assessment are recorded into the patient’s medical record. However, with the increasing use of electronic medical records, Mayo Clinic is moving to have PROs collected electronically and is currently piloting this process.
Several other centers have evaluated electronic PRO collection and have documented their success. Jensen and colleagues13 surveyed oncology centers using PROs and found 33 unique systems in use. Most systems allowed data collection within the clinic, whereas 37% allowed collection both remotely or in the clinic. More than 70% of the systems allowed scores to be immediately available to clinicians after the information was entered by the patients.13
Understanding of the information technology resources and support at any institution or practice is important when selecting an electronic PRO collection system. Many electronic health records do not have the capability of incorporating PRO data into the record and thus may require signing into a separate system to review PRO data; this may be a source of major dissatisfaction for clinicians. To increase the chances of successful implementation, every step of the clinical workflow should be assessed to decide how data collection use will impact that flow as well as personnel requirements.
Engaging Physicians and Patients to Obtain Buy-In
Gaining buy-in and engagement from health care staff is important for PRO acceptance and successful utilization in clinical care. In general, it is not recommended for clinicians to devise their own questions for evaluating PROs, but rather measures should be determined by a multidisciplinary group in a research setting.10 However, it is important to organize all stakeholders and obtain buy-in from anyone who will be involved. This includes patients, because the need to collect sufficient PRO information must be balanced against the patient burden of completing the PRO measure.
One of the greatest barriers to success is the perception of additional burden for physicians and exacerbating their limited time further. For this reason, clinicians should be allowed to provide input as to how many items they want to review during the clinical encounter. Clinicians may worry that introducing PRO data into the practice will significantly prolong the clinical encounter. However, studies have found that incorporation of PROs into the practice does not increase patient visit duration by an appreciable amount.14,15
The success of our PRO initiative was due to our conducting numerous focus groups of various staff and patients to gain an understanding of their opinions and ideas for how to use PROs in practice. In addition, the study group was multidisciplinary and focused on how participants could gain engagement with their peers for the project.
Physicians are often the most difficult to engage, due to varied data on the use of PROs and their success or lack thereof in other settings. One approach we tested was to highlight the current standard for symptom and QOL assessment, which is relying on the patient to bring up an issue at a clinical visit or hope that the provider inquires about their symptoms or QOL of their own accord. Frequently, no issues are brought up by the patients, and, as demonstrated in multiple studies in various patient populations, including patients with cancer, even if an issue is raised, providers often overlook or underestimate the issue/symptom and its impact on QOL.8,9,14,15 We found that presenting this data to physicians is beneficial and persuasive and improves the likelihood of acceptance of PRO by providers in routine clinical care.
Using Pro Data in Clinical Practice
Using PRO data within the clinical practice to address issues of importance to the patient can enhance communication and potentially increase the quality of care. The scoring of the PRO data and training of the clinical staff to interpret PRO results are key factors in whether incorporation of PRO collection into the practice will succeed.10 Questionnaires may be scored manually or electronically depending on practice capability. Result reporting will largely depend on the stakeholders in the clinical team, who will view the data, and their preference of data display such as bar graphs vs simple numerical scores. Regardless of the way it is reported, PRO data need to have scores that are clinically meaningful and will help guide clinicians in use of the data. Ultimately, discussion with the patient that highlight his or her response will lead to a better understanding of the patient’s issues and, in turn, how to address these issues.
The collection of PRO data using the patient-reported symptom screen has been a valuable addition to our care process. The completion of the measure allows patients to reflect on how they have been doing, and the results offer health care providers with a quick evaluation of the patients’ well-being at each clinical visit. This instrument also provides insight into potential toxicity from treatment or disease process. In doing so, it has also identified issues that may otherwise not have been raised.
A survey of oncologists at Mayo Clinic Rochester was performed following implementation of this PRO screen, and 86% reported using information from the screen in their practices.16 They also reported talking to their patients about the specific ratings.16 Furthermore, among the oncologists who responded, more than 90% stated that using the PRO data did not lengthen their clinical visit or add more work.16 This was a tremendous success for our institution, but it is credited to the involvement of all potential stakeholders who would be affected.
Conclusion
PRO collection in clinical practice is the subject of much research, but much work still needs to be done to promote widespread acceptance of PRO collection to enhance care delivery. Clinicians still need to be convinced that the effort for PRO collection does provide value over current routine care. Through continued study, clinician and patient engagement, and PRO implementation in the clinical environment, enhancement of quality of care through patient centeredness will continue to evolve.
Successful integration of PROs into our clinical society will require systemization and establishment of consensus between researchers, providers, and patients about the instruments and outcomes to measure. There are many challenges but this newfound emphasis on incorporating the patient’s perspective into practice; not relying solely on labs, pathology, and imaging, to make care decisions is an encouraging start. However, it will be key to publish practice improvement projects and data on utilization of PROs to add to the body of information available to support and guide successful implementation more broadly.
In addition, although PROs have been studied for their impact on clinical practice outcomes, especially in oncology practice, the use of clinical practice guidelines tied to PRO measures has not been thoroughly evaluated. For example, the use of an instrument to screen for depression in an oncology patient may be tied to a pathway such as the National Comprehensive Cancer Network (NCCN) distress guidelines. However, in a pilot of the use of PROs in radiation oncology clinical practice at Mayo Clinic in Arizona, the NCCN distress guidelines were not deemed to be a useful adjunct to PRI use.17 Therefore, there is a need for clinical guidelines or disease management pathways to address issues identified through PRO collection, so that clinicians can consider pathway use an enhancement of quality of care.
References
1. Institute of Medicine. Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, DC: National Academies of Science, Engineering, and Medicine; 2001.
2. Bergman J, Brook RH, Litwin MS. A call to action: improving value by emphasizing patient-centered care at the end of life. JAMA Surg. 2013;148(3):215-216.
3. Kerr EA, Hayward RA. Patient-centered performance manangement: enhancing value for patients and health care systems. JAMA. 2013;310(2):137-138.
4. Fayanju OM, Mayo TL, Spinks TE, et al. Value-based breast cancer care: a multidisciplinary approach for defining patient-centered outcomes. Ann Surg Oncol. 2016;23(8):2385-2390.
5. Berman AT, Rosenthal SA, Moghanaki D, et al. Focusing on the “person” in personalized medicine: the future of patient-centered care in radiation oncology. J Am Coll Radiol. 2016;13(12 Part B):1571-1578.
6. Acquadro C, Berzon R, Dubois D, et al. Incorporating the patient’s perspective into drug development and communication: an ad hoc task force report of the Patient-Reported Outcomes (PRO) Harmonization Group meeting at the Food and Drug Administration, February 16, 2001. Value Health. 2003;6(5):522-531.
7. Velikova G, Booth L, Smith AB, et al. Measuring quality of life in routine oncology practice improves communication and patient well-being: a randomized controlled trial. J Clin Oncol. 2004;22(4):714-724.
8. Velikova G, Keding A, Harley C, et al. Patients report improvements in continuity of care when quality of life assessments are used routinely in oncology practice: secondary outcomes of a randomised controlled trial. Eur J Cancer. 2010;46(13):2381-2388.
9. 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(6):557-565.
10. Snyder CF, Aaronson NK, Choucair AK, et al. Implementing patient-reported outcomes assessment in clinical practice: a review of the options and considerations. Qual Life Res. 2012;21(8):1305-1314.
11. Greenhalgh J, Long AF, Brettle AJ, Grant MJ. Reviewing and selecting outcome measures for use in routine practice. J Eval Clin Pract. 1998;4(4):339-350.
12. Cella D, Eton DT, Fairclough DL, et al. What is a clinically meaningful change on the Functional Assessment of Cancer Therapy-Lung (FACT-L) Questionnaire? Results from Eastern Cooperative Oncology Group (ECOG) Study 5592. J Clin Epidemiol. 2002;55(3):285-295.
13. Hubbard JM, Grothey AF, McWilliams RR, Buckner JC, Sloan JA. Physician perspective on incorporation of oncology patient quality-of-life, fatigue, and pain assessment into clinical practice. J Oncol Pract. 2014;10(4):248-253.
14. Pakhomov SV, Jacobsen SJ, Chute CG, Roger VL. Agreement between patient-reported symptoms and their documentation in the medical record. Am J Manag Care. 2008;14(8):530-539.
15. Abernethy AP, Zafar SY, Uronis H, et al. Validation of the Patient Care Monitor (Version 2.0): a review of system assessment instrument for cancer patients. J Pain Symptom Manage. 2010;40(4):545-558.
16. Fromme EK, Eilers KM, Mori M, Hsieh YC, Beer TM. How accurate is clinician reporting of chemotherapy adverse effects? A comparison with patient-reported symptoms from the Quality-of-Life Questionnaire C30. J Clin Oncol. 2004;22(17):3485-3490.
17. Halyard MY, Tan A, Callister MD, et al. Assessing the clinical significance of real-time quality of life (QOL) data in cancer patients treated with radiation therapy. J Clin Oncol. 2010;28:15s.
