Selecting patient-reported outcome measures for a patient-facing technology

Abstract Objective This article provides insight into our process and considerations for selecting patient-reported outcome measures (PROMs) designed for self-reporting symptoms and quality-of-life among breast cancer (BCA) patients undergoing oral anticancer agent treatment via a patient-facing technology (PFT) platform. Methods Following established guidelines, we conducted a thorough assessment of a specific set of PROMs, comparing their content to identify the most suitable options for studying BCA patients. Results We recommend utilizing the combination of EORTC QLQ-C30 + EORTC QLQ-BR45 as the preferred instrument, especially when developing a dedicated “breast cancer-only” application. Discussion When developing and maintaining a dashboard for a PFT platform that includes multiple cancer types, it is important to consider the feasibility of interface design and workload. To achieve this, we recommend using PRO-CTCAE+PROMIS 10 GH for the PFT. Moreover, it is important to consider adding ad hoc items to complement the chosen PROM(s). Conclusion This article describes our efforts to identify PROMs for self-reported data while considering patient and developer burdens, providing guidance to PFT developers facing similar challenges in PROM selection.


Background and significance
Patient-reported outcome measures (PROMs) are standardized questionnaires that patients use to report their health status, particularly relevant for cancer patients undergoing treatment to document cancer-related symptoms and treatment side effects.Patient-facing technologies (PFTs), frequently integrating electronic patient-reported outcomes (ePRO), aim to improve patient engagement, enhance patient experience, facilitate self-care, and ultimately contribute to better health outcomes. 1These technologies have gained growing importance in healthcare, with the emergence of desktop, mobile, and web applications demonstrating notable benefits.During our efforts to develop a PFT to assist patients with breast, lung, prostate, and colorectal cancer, we faced the challenge of choosing the most appropriate PROM that could efficiently assess patients' health status while minimizing the reporting burden.Additionally, creating a dashboard to display PROM data and depict the evolving trends in PROs over time 2 proved to be a complex task, intended to enhance patient self-assessment and communication with healthcare providers. 3Furthermore, this PFT is designed to adapt and generate personalized, magazine-style patient education and self-management suggestions based on PROs.
5][6] This article adopts a systematic approach to compare prominent PROMs designed for breast cancer (BCA).Additionally, we delve into the complexities of PROM selection, discuss the strengths and weaknesses of the PROMs we have chosen, and explore potential avenues for future research on PFTs that support multiple cancer types.Using BCA PROMs as a case study, this article aims to provide insights for fellow researchers and PFT developers encountering comparable difficulties in PROM selection.

Objectives
The primary goals of this study were to determine the most appropriate PROMs for self-reported symptom assessment and quality of life evaluation among BCA patients receiving home-based care through a PFT.Additionally, we explored the reporting burden and the feasibility of design considerations for a broader range of cancer types.

Patient population group
Our focus is on adult BCA patients receiving home-based care, especially those undergoing treatments such as oral anticancer agents (OAAs).These patients exhibit an adequate performance status that allows them to actively engage in self-reporting.Furthermore, they have experienced a shift in their care from medical facilities to home-based settings within the past year.The selected PROMs/items must proficiently track and evaluate the symptoms associated with their particular treatment(s).

Principles for PROM selection
To systematically select the most appropriate PROMs for our application, we adopted the 6 principles proposed by Luckett et al. 7 The first principle states that the population characteristics should be carefully considered in PROM selection. 7The second is to use proximal PROs (such as symptoms) as the primary outcome and distal PROs (such as QoL) as the secondary outcome. 7The third and fourth principles state that PROMs should be identified primarily based on the content and structure of subscales within the PROM, and they should be supported by the adequacy of evidence for psychometric properties (such as reliability and validity) and track records of use in similar populations.It should be noted that psychometric evidence must be documented for the ePRO mode of administration. 8The fifth principle discusses feasibility regarding respondent burden, cost, availability of translations, etc., while the final one states that ad hoc or supplementary items must be used minimally and can be appended at the end of the main PROMs and should be scored separately. 7jor PROMs for consideration Many published questionnaires are available for measuring symptoms and quality of life in cancer patients, including PROMs specific to BCA. 4 A comprehensive list can be found on the PROQOLID (Patient-Reported Outcome and Quality of Life Instruments Database) database (accessed July 21, 2023, https://www.qolid.org/).5][6] A comprehensive list of PROMs used in BCA studies has been listed in Appendix SA.Amongst the select PROMs, the QLQ-BR23/BR45 and FACT-B are the only 2 quality-oflife instruments explicitly developed for BCA patients "facing different disease stages and treatments". 4Both are widely used in multiple BCA treatments and act as supplements to their general cancer PROMs, namely the EORTC QLQ-C30 and FACT-G.EORTC QLQ-C30 is a 30-item cancer-specific QoL instrument with items on function, global QoL, and symptoms. 6The BR23/BR45 addresses several concerns specific to BCA. 9 FACT-G is also a cancer-specific QoL instrument with 4 well-being subscales and includes one item on global QoL.Its BCA supplement FACT-B (which, unlike BR23, subsumes the FACT-G) has an additional scale covering items of concern specific to BCA. 10 Another highlighted instrument is the Patient-Reported Outcome-Common Terminology Criteria for Adverse Events (PRO-CTCAE).PRO-CTCAE has been the gold standard instrument for reporting symptomatic adverse events in many cancer types for more than 30 years and provides a comprehensive assessment of cancer-related symptoms through a library containing 124 items representing 78 adverse events, divided into 14 categories by organ system. 11ther prominent PROMS include BREAST-Q and PROMIS, which we have reviewed but excluded for further consideration.BREAST-Q is focused on breast surgery patients instead of the BCA population.The PROMIS suite lacks a BCA-specific module.Nonetheless, PROMIS 10 SF could be used as a QoL PROM.We also reviewed the BCA PROM recommendations made by 2 consortia, namely International Consortium for Health Outcome Measures (ICHOM), 12 and Translational Breast Cancer Research Consortium (TBCRC). 13We have verified that our PROMs cover the core outcome set described in the recommendations.
In summary, we examined the 3 PROMs mentioned above (EORTC QLQ-C30þBR45, FACT-B, and PRO-CTCAE) and conducted a comparative analysis to identify the most suitable option(s) for our application, complemented by ad hoc items.

Item-wise comparison of selected PROMs
Table 1 provides a detailed comparison of the selected instruments (namely, the PRO-CTCAE, C30þBR45, and FACT-B) by items covered, and the interpretation is discussed in the next section.The latter 2 instruments also measure function/ wellbeing, and QoL.EORTC-CIPN20 and FACT-ES have also been listed to support the discussion on ad hoc items later.

Instrument of choice
The PRO-CTCAE provides a comprehensive library of cancer-specific symptomatic adverse events; however, in its entirety, the PRO-CTCAE raises feasibility issues when considering respondent burden due to its large size.Not all items are relevant to every cancer type, and a customized selection of items is recommended by both PRO-CTCAE developers 11 and researchers. 14Furthermore, several concerns pertinent to BCA, such as arm and breast symptoms, weight gain, and body image, are not included in the PRO-CTCAE.On the other hand, the BR45 and FACT-B are shorter instruments that were developed keeping BCA disease and treatment in mind and thus provide a tailored collection of BCA-relevant symptoms, including items not covered by the PRO-CTCAE.
Thus, we recommend the use of EORTC QLQ-C30 þ EORTC QLQ-BR45 as the primary instrument for symptom and quality-of-life reporting for a PFT specific for BCA, preferred over FACT-B due to its more inclusive content.Ad hoc items may be derived from the PRO-CTCAE, EORTC QLQ-CIPN20, FACT-ES, and patient free-text entry, based on provider/researcher suggestions, as discussed below.
When expanding the PFT design beyond BCA, we chose to integrate PRO-CTCAE into our PFT, guided by the fifth principle of feasibility regarding respondent burden, cost, etc.The logic behind this choice stems from our application's user base, which encompasses patients with 4 distinct cancer types.Employing separate instruments for each cancer type would impose a substantial workload on developers tasked with designing customized dashboards.To simplify and streamline this process, we opted for the straightforward symptom severity items derived from PRO-CTCAE, which can be universally applied across all cancer types.
In our research, we discovered that the most suitable instrument might vary depending on the cancer type.For instance, BR45 proves effective for BCA while FACT-P works well for advanced prostate cancer.Implementing specific instruments for each cancer type would entail additional efforts for developers in designing tailored dashboards.Since PRO-CTCAE focuses solely on symptom assessment and does not measure QoL, we included PROMIS 10 Short Form in our battery of measures.This addition includes 2 items pertaining to global QoL to ensure a comprehensive evaluation of patients' well-being.

Principles applied for the instrument selection
Both the PRO-CTCAE and the EORTC instruments follow the first 4 principles in that they are relevant to the patient population group and measure symptoms as primary outcome.The third principle is also satisfied in the EORTC and PRO-CTCAE batteries, in that symptoms are measured followed by QoL.6][17][18][19][20] Using the C30þBR45 in its entirety rather than selecting items prevents possible jeopardization of the construct validity of established scales of the instrument, honoring the third and fourth principles.
Patient burden is a crucial factor, as extensive questionnaires often lead to missing values due to nonresponses.It is generally recommended that a survey should require no more than 10-20 min 11 for patients to complete, equating to approximately 100 items or less.The C30þBR45 survey (N ¼ 75) falls within this limit.A similar outcome can be achieved with PRO-CTCAEþPROMIS 10 SF (N ¼ 134) if we utilize a subset of PRO-CTCAE, which we intend to do in our study.All the instruments listed in Table 1 have free permission for use, except the chemotherapy-induced peripheral neuropathy (CIPN)-20.

Ad hoc items
Assuming QLQ-C30 þ BR23/45 as the primary instrument, there are multiple sources of ad hoc items, which can be added/deleted from the main battery throughout the period of patient enrollment.First, we suggest adding the 2 cardio/ circulatory items of the PRO-CTCAE to track heart failure.It is now recognized that there is an increased risk of heart failure in BCA patients due to the effects of treatment and the overlap of risk factors between the 2 diseases. 21Other PRO-CTCAE symptom items not extensively covered by C30/ BR45 (such as cutaneous and genitourinary items) may be offered as a checkbox-style library for selection by the patients during each survey, and the items selected by individual patients can be added to the personalized batteries in subsequent surveys.Secondly, we recommend employing the EORTC QLQ-CIPN20 22 as a tool to assess sensory, autonomic, and motor peripheral neuropathy in patients undergoing treatment with specific chemotherapeutic agents, notably taxanes, which are recognized for their potential to induce peripheral neuropathy.
Note: Highlighted portions indicate the items we would include in a BCA application battery.
JAMIA Open, 2023, Vol. 6, No. 4 Peripheral neuropathy is a common toxicity observed in long-term BCA survivors treated with taxane-based chemotherapy. 23It is worth mentioning that the PRO-CTCAE includes just 2 items related to sensory neuropathy.Similarly, patients treated with oral endocrine therapy may present unique vaginal symptoms such as vaginal discomfort and pain or dryness vagina during sexual activity not tracked by the BR45.For such symptoms, items from the FACT-Endocrine Symptoms (ES) may be used as ad hoc items. 13imilar suggestions may be made for mental health and a more granular assessment of pain and other symptom clusters.Last but not least, patients should have the option to make free-text entries to report their concerns through the PFT, and such reports can be used to invent more ad hoc items for uncommon symptoms in subsequent surveys in an individualized manner.It is important to acknowledge that ad hoc items not only add to the developer burden but also raise psychometric concerns.Therefore, we suggest exercising caution and using them sparingly.

Discussion
The article summarizes a systematic approach to PROM selection for a PFT and recommends a precise battery of instruments for use with BCA patients.Literature on PROMs used for routine cancer care in PFTs often does not describe their criteria for PROM selection or how the selected PROMs/items meet the needs of a heterogeneous patient population.In this article, we have delineated our PROM selection process and how ad hoc items can support reporting tasks by a heterogeneous population, that is, patients under chemotherapeutic/hormonal/targeted OAA therapy.Our battery includes the core outcome sets listed in the consortia recommendations.Further research is needed to document ePRO psychometric evidence for the above instruments.Time points of assessment, scoring, and thresholds for clinical alert generation need further discussion.

Conclusion
We applied the 6 principles to justify our choice of PROMs for BCA patients in a PFT.Combining PROMs such as EORTC QLQ-C30, BR45, and ad hoc items can potentially yield comprehensive PRO data encompassing symptoms and QoL from BCA patients.However, opting for PRO-CTCAE in our PFT, which serves 4 different cancer types, proves to be more practical and feasible.The simplicity of symptomseverity items facilitates visualization, tracking, and trend analysis without requiring extensive technical complexities in dashboard design.Additionally, feasibility considerations related to dashboard design can be proposed as an additional subprinciple within the fifth Luckett principle that developers should consider when selecting a PROM.Further research is necessary to explore symptoms associated with OAAs during care transitions.Moreover, there is a need to identify PROMs that can efficiently capture self-reported data while considering the burdens on both patients and developers.

Table 1 .
Selected PROMs and their outcome coverage.