Historically, clinical outcomes, such as survival and recurrence, were used to assess quality of care in oncologic surgery. Clinical outcomes continue to be important; however, they are no longer sufficient when assessing outcomes in breast surgical oncology. More recently, additional performance metrics have been focused on structure and processes of care as well as outcomes that patients find most important.1,2 Patient-reported outcomes (PROs) are defined as any report of the status of a patient’s health condition that comes directly from the patient, without interpretation of the patient’s response by a clinician or anyone else.3

PROs provide insight into how patients perceive health and treatment effects, how treatments impact outcomes, and are helpful to determine how disease and surgical interventions impact many aspects of patients’ lives.4 As described in Table 1, commonly utilized metrics include survival and disease control, degree of recovery and functional status, access to treatment, treatment-related complications, health-related quality of life, and long-term consequences of therapy.5 PROs not only have the potential to monitor changes in individual patients throughout the course of their treatment but also may provide information to support patients in shared decision-making, serve as a measurement of patient-centered care, allow hospitals to monitor outcomes of care over time, play a role in value-based payments, and identify targets for quality improvement efforts.4,6,7

Table 1 Classes of patient-reported outcomes
Full size table

PROs in Breast Surgical Oncology

An evaluation of randomized, controlled trials (RCTs) in cancer patients identified 447 RCTs with PROs identified as an endpoint from 2004 to 2013.8 Of those studies, the most common disease site found to have PROs as an endpoint was breast with 123 RCTs. PROs were the primary endpoint in 24% of RCTs and the secondary endpoint in 76% of RCTs. Howell and colleagues similarly performed a systematic review of PROs utilized in routine cancer clinical practice and found that PROs were most commonly utilized with breast cancer patients.9 The PRO used most often was the European Organization for Research and Treatment of Cancer (EORT-Qol C30), which was commonly paired with disease-specific modules and the Hospital Anxiety and Depression Scale. A review of the BREAST-Q measure, which addresses outcomes related to breast reconstruction, found increasing annual numbers of publications using BREAST-Q 5 years after the development of the measure (from < 5 to 20).10

A review of 10 PRO metrics in oncologic breast surgery by Chen et al. summarized the PRO measures, methodology for development of measures, and psychometric properties.11 We have listed commonly used PRO metrics in breast surgery in Table 2. Most measures were developed using expert opinion, literature review, and patient focus groups or interviews.4 A weakness of some metrics was lack of patient involvement in PRO development, while strengths of breast surgery PROs included focus on surgery specific measures and formal psychometric evaluation of the metrics.11

Table 2 Commonly used patient-reported outcomes in breast surgical oncology
Full size table

Benefits of PROs

Implementation of PROs have been found to result in improvements in clinical outcomes, symptom management, resource utilization, and patient engagement.1 A RCT to evaluate 12 PROs in cancer patients found a trend toward improved survival in the PRO group versus standard care (75 vs. 69%, p = 0.05).12 The authors also noted longer adherence to chemotherapy (mean 8.2 vs. 6.3 months, p = 0.002), decrease in hospitalizations (45% vs. 49%, p = 0.08), and fewer visits to the emergency room (34% vs. 41%, p = 0.02) in the PRO group.12 Furthermore, health-related quality of life (HRQOL) at 6 months improved more often in the intervention group (34 vs. 18%, p < 0.001) compared with the standard of care group.

A literature review demonstrated a positive correlation between the use of PROs and improved patient satisfaction in 13 of 16 studies.9 Additionally, two RCTs of PROs demonstrated a positive association with patient satisfaction, although neither study reached statistical significance.9 When PRO data are provided to patients in an easy to interpret format, it will allow patients to monitor their own outcomes over time, which may improve patient engagement and satisfaction.13

Monitoring of patient symptoms with PROs has been found to increase provider response to symptoms with increased referrals for psychosocial issues and treatment for pain. Furthermore, improvement in patient symptoms persisted at subsequent follow-up visits.9 Collection of PROs before clinic visits has the potential to set priorities for the clinical encounter and allow providers to prioritize patient concerns to make the most of patient-clinician communication during office visits.13 Figure 1 demonstrates how PROs can be tracked by clinicians over time. This data can be reviewed at each clinic visit to stimulate conversation about patient symptoms and satisfaction, which may lead to more effective and efficient patient-provider communication.

Fig. 1
figure 1

Provider review of BREAST-Q patient-reported outcome data. Data can be reviewed at each clinic visit to stimulate conversation about patient symptoms and satisfaction, which may lead to more effective and efficient patient-provider communication

Full size image

A RCT evaluating PROs vs standard of care demonstrated an improvement in HRQOL at 6 months in the intervention group (34 vs. 18%) and found fewer declines in quality of life in the intervention group (38 vs. 53%, p < 0.001).12 The mean HRQOL also decreased less in the PRO group compared with standard of care (1.4 vs. 7.1 point drop, p < 0.001).

PROs and Reimbursement

As the focus on oncologic outcomes shifts from survival and recurrence to long-term outcomes and patient-centered care, hospital reimbursement has already been linked to patient experience and outcomes, such as complications, readmissions, and reoperations.14,15 It is likely that reimbursement also will be tied to PROs as hospital reimbursement shifts from fee-for-service to value-based care. PROs will not only become a meaningful way to capture quality-of-life issues that are important to patients but also will allow payors to monitor value of care more broadly.

In 2015, Congress passed the Medicare Access and Children’s Health Insurance Program Reauthorization Act (MACRA) with the goal of tying Medicare reimbursement to outcomes and value of care as opposed to volume.16 The Quality Payment Program (QPP) focuses on enhancing value of care by continuing to improve patient outcomes, while also reducing costs by standardization to improve cost efficiency. Currently there are seven claims-based measures reported through MACRA; however, in the near future additional nonclaims-based PRO measures may be added. It is imperative that providers and hospital administration learn about PROs and implement them before it is mandated and linked to reimbursement.

Development of PROs/Methodology

Figure 2 demonstrates an outline of the steps to developing and validating PRO measures. Some studies focus on development of metrics that are sensitive and specific to the intervention of interest (i.e., nipple-sparing mastectomy), whereas others try to create PROs that are more generalizable across populations.17 Involvement of all potential stakeholders, including providers, staff, family members, and most importantly, patients is key when developing PRO metrics.6

Fig. 2
figure 2

Development of patient-reported outcomes. Outline of creation of PRO metrics, data collection, analysis of metrics, and how results can be interpreted and reported

Full size image

It is important to understand gaps in resources and information use before data collection so the process can be streamlined.6 Choosing consistent measures and scales for metrics also is important in ensuring that the data collected will be useful.6 Instrument psychometric properties, including validity, reliability, and responsiveness, also should be measured and validated.2,4,7 Other important factors to report when considering PRO methodology include: compliance, timing of assessments, and how missing data was handled.2

The Patient Reported Outcome Measurements Over Time in Oncology (PROMOTION) registry evaluates RCTs, PRO methodology, analysis, and study validity.18 The PROMOTION registry has been able to improve the methodology of PRO assessment in prostate cancer RCTs over time and continues to monitor for areas where PRO methodology is poor or needs improvement.18 An addition to the Consolidated Standards of Reporting Trials (CONSORT) for RCTs, CONSORT PRO also provides guidance where PROs are primary or secondary endpoints in RCTs.19 The CONSORT PRO guidelines indicate that PROs should be identified as an endpoint in the abstract, a description of the hypothesis of PROs and relevant domains should be provided, evidence of the instrument’s validity and reliability should be provided or cited, statistical approaches for dealing with missing data should be explicitly stated, and PRO specific limitations of study findings and generalizability of results should be discussed.19

When interpreting PROs, the clinical significance and presentation of results should be taken into consideration.2 There should be a standardized framework and process for analyzing and scoring data.6,20 Data also should be made available for research and quality improvement work.6,21

Challenges to Implementation of PROs

There are many challenges to the implementation of PROs, and no standardized method for integrating PROs into the usual workflow have been established.22 There is a potential for a substantial burden to patients, with varying number of questions and time needed to complete PRO measures. Patients with multiple health conditions or who utilize the health care system frequently may receive many different surveys resulting in survey fatigue. Patients also may have difficulty completing surveys due to poor health.4,6,13 Additionally, patients may not have access to the technology required to complete PROs and predefined times for data collection may not align with clinic visits requiring patients to complete PROs from home.6,13 While PRO data may be used by patients to track their health outcomes, if there is a lack of focus on the patient as the end user they may have difficulty viewing and interpreting their data.13

There is a financial burden to institutions who wish to implement PROs, and there are currently no financial incentives for widespread implementation of PROs.22 Furthermore, there is an increased burden on providers and staff with implementation of PROs and the associated cost of staff training and time to monitor PRO data. Data management and analyses is resource-intensive. To implement PROs, the clinical workflow needs to be reworked, and providers who can dedicate time to evaluating PROs need to be identified.4,13 More importantly, there is a need to respond to severe or worsening symptoms in a timely fashion to ensure patient safety both during normal clinic hours and on nights and weekends.13

Concerns related to collection of data include that most PROs are collected through patient portals, which many patients do not use and which may not be readily captured by the electronic health record.22 There also is no commonly accepted terminology for PRO metrics and some have raised concerns about the reliability of patient-generated data.13,22 In addition, there are many PRO surveys to choose from making it difficult to compare data across studies, and while there are many established measures, there are metrics that are important to patients that are not well represented in PRO measures (i.e., end of life care).

There are many challenges to interpreting PRO data. There is a need for normative data for comparison.7 There also is a potential for response bias and undersampling of patient populations who may not have access to the technology required to complete PROs, such as patients of low socioeconomic status.7,9 In addition, missing data needs to be acknowledged and accounted for when reporting data. Providers need to have the ability to easily view PRO data to respond efficiently to patient symptoms and effect changes for both individual patients and identify areas for quality improvement.22

Also, an unintended consequence of PRO data collection may be an increased focus on patients as consumers with a focus on short-term satisfaction (i.e., overprescribing pain meds or reluctance to discuss patient conditions such as obesity or smoking) rather than long-term cancer care outcomes. Patient satisfaction as measured by the Hospital Consumer Assessment of Healthcare Providers survey has not been found to correlate with patient outcomes and has recently been tied to hospital reimbursement.23,24 Because PROs are likely to be tied to reimbursement in the future, it is imperative that measures focus on specific outcomes that are important to patients.

Keys to Success/Future Work

While there are many inherent challenges, PROs have already been successfully implemented. One of the keys to success has been physician or team champions with support from institutional leadership.20,22 To ensure that PROs are patient-centered, patients should be integral in the development, implementation, and interpretation of data.13 Both patients and providers may benefit from having immediate access to PRO data for participation in shared decision making.7,20,25 The burden of PRO data collection can be minimized for both patients and providers by optimizing workflows, keeping surveys short, and minimizing the number of surveys for patients with multiple medical conditions and high hospital utilization.9,20

Utilization of technologic advances, including ability to complete measures electronically and integration of PROs in the EHR, may facilitate both completion of PRO measures by patients as well as review of data by providers.7,13,25 The ability to link PRO metrics with other EHR data will enable researchers to evaluate effectiveness of interventions, assess quality of care, and monitor pharmacovigilance.22 Automation of PRO workflow can allow providers to be notified if worsening or severe symptoms are reported so that patients can be contacted and interventions can be undertaken in a timely fashion.13

PROs are adding a much needed focus on patient centered care to the traditional measures of quality of care.7,25 It is likely that in the future PROs will be tied to reimbursement as a quality measure, PRO data may be incorporated into transparent performance measures, and patients could use PRO data to guide their selection of surgeons and institutions.1,26 If PROs are tied to provider reimbursement, there should be appropriate condition-specific measures and scales to evaluate performance. In addition, reimbursement models should be transparent to providers.1,20

In breast surgical oncology, PROs have been utilized and most published studies focus on overall satisfaction with their breasts postoperatively. To improve comprehensive evaluations of outcomes in breast cancer patients, more work is needed to assess psychosocial, sexual, and physical well-being domains.10 Normative data has been published with the BREAST-Q scale; however, there is a need for normative data for other breast surgical oncology specific measures.27

Conclusions

Because there is an increased emphasis in providing high-value care in cancer patients, the addition of metrics, such as PROs to traditional clinical measures of high-quality care, are becoming increasingly more common. While there are many challenges to implementing PROs, they are not insurmountable and the benefits of using PROs to monitor symptoms in individual patients, assist in shared decision making, implement quality improvement work, and has the potential to play a role in value-based payments for surgical oncology care. Keys to the successful implementation of PROs requires strong physician leadership and institutional support, utilization of technological advancements, engagement of patients and family members, and automated and rapid return of PRO data to providers to respond to patient symptoms quickly and accelerate quality improvement work.