15. Electronic Patient-Reported Outcome Measures (ePROMs) in Rheumatology

The technological infrastructures facilitate real-time and long-term systematic patient-centered data collection as integral components of care [5, 29, 30]. Today’s information technologies allow intelligent orchestration of PROMs data collection, analyses, and reporting and thus provide a wide array of exciting challenges and opportunities not only for routine patient care but also medical research [31]. Various (non)pro prietary applications simplify data acquisition and accelerate information transfer between patients and physicians by eliminating intermediate data collection and processing steps (e.g., double data entry), increase flexibility of data capture (e.g., frequencies and locations), facilitate clinical decisions, and can thus improve the efficiency of clinical workflow [30, 32‐37]. Changing specific personal or disease-specific needs, treatment regimens, or phases might require different assessments. ePROMs might simplify context-based customization of the assessments [38]. They allow real-time flagging of important, clinically relevant symptoms. Frequent monitoring and reviewing of patient-centered issues and needs is enabled; thus, ePROMs offer an ability to enhance clinical care and quality assurance [29, 33]. Multiple follow-ups of definable time periods may be seen at a glance and/or compared with control groups, allowing adaptation of treatment plans and processes without the need for additional staff or equipment [39, 40]. As ePROMs improve the accuracy of data collection of symptoms indicating poor conditions and outcomes, they might even help to triage patients who need more extensive care, for example, in terms of evaluation and (non)medical interventions [2, 20, 30]. Electronic patient-centered monitoring may also facilitate patients’ management across care transitions [41]. ePROMs can enhance cost-efficiencies [20]. More ePROMs in EHRs could even widen the capacity to undertake population-based research [42].

Assessments with the help of ePROMs (e.g., in electronic diaries) allow to document time-stamped entries and thereby might positively influence compliance [43]. Visual feedback for patients—integrated into the Electronic Recording of Outcome Measures for Inflammatory arthritis and Ankylosing spondylitis (EROMIA) system—had a positive and significant impact on the disease activity control [44]. Recently, a strong correlation was reported between the use of a software application (Rheum-PACER) and disease control [45]. ePROMs give the opportunity to obtain a broadened view on disease courses and patterns also for the empowered patients who proactively participate in the management of the course of their disease [46]. In contrast to initial beliefs, patients have a positive attitude toward the use of PROMs in computerized systems [33]. As the public and thereby the patients become more experienced and familiar in the use of new technologies, ePROMs gain higher acceptability and are commonly preferred by patients over paper-based versions [33, 40, 47, 48]. However, recently active ePROM use in a Web portal was only performed by less than half of the patients [49].

Not only advantages (pros) but also disadvantages (cons) of ePROMs are summarized in Table 15.2.

Before broad introduction of ePROMs into clinical routine, careful comparison of data obtained by paper–pencil and computerized versions of the assessments was and is crucial, because equivalence of data obtained by the two acquisition methods cannot be taken for granted [18, 19, 36, 50, 51]. Detectable (test–retest) correlations between the modes of administration need to meet methodological requirements for demonstrating reliability and validity [50, 51]. Assessments on electronic devices need to be able to detect changes over time [17]. Coons et al. published a general framework for the transfer of paper-based PROs to electronic devices [52].

Meanwhile many traditional scales have been evaluated [5]. Gwaltney et al. showed in their meta-analysis that computer and paper measures produce equivalent scores. Subjects’ computer experience and age did not influence this result when “small mean differences” were not regarded as clinically relevant [50]. Similarly, Campbell et al. published a review that summarizes 55 studies investigating 79 instruments. It provides a good overview for rheumatology and other disciplines [47]. The authors stated that paper-based and electronic formats are usually rated to be equivalent, and that study participants prefer electronic assessments [47]. However, they recommend further validations of electronic versions, taking into account that data assessed electronically should produce figures that are equivalent or superior to those retrieved via paper-based versions [47, 52].

Electronic PROMs are available not only in EHRs and rheumatology-specific patient documentation systems, but also in (online) registries and other partly “registered user-restricted” online applications developed in a number of countries worldwide (e.g., http://www.​medal.​org and http://​www.​nihpromis.​org/​?​AspxAutoDetectCo​okieSupport=​1#2; [12, 32, 33, 44]). In addition, (non)profit organizations have implemented systems that allow ePROMs assessments (e.g., http://​c-path.​org/​programs/​epro/​ and https://​www.​parexel.​com/​solutions/​informatics/​clinical-outcome-assessments/​epro/​). Studies showed that use of health information technology may be associated with better outcomes (e.g., reductions in mortality, complications, and costs) [53]. Developed systems have become more affordable and feasible to implement [5]. Patients’ and others stakeholders’ integration into the development and implementation process is necessary and will lead to better acceptance of the systems [54].

The heterogeneity of systems allowing to document ePROMs becomes obvious from a cancer research review published in 2013. This paper identified at least 33 unique systems [41]. However, each of the systems differed in features and characteristics [41]. Considering publications on this topic, a similar number of systems is expectable in rheumatology. As evaluation data is not available for all systems, their number might even be underestimated in a meta-analyses.

Due to the rise in connectivity and the applicable devices, the range of times and locations where patients can complete assessments (e.g., at home, waiting room, or drug store/pharmacy kiosks, or use of their smartphone/tablet) has been enlarged [5]. There are different electronic devices that can assist gathering ePROMs: traditional computers, tablet PCs, smartphones, and online platforms.

Computer/Web-based documentation software-systems allow the collection and documentation of self-reported ePROMs data and/or physicians’ clinical findings (e.g., http://​www.​raintreeinc.​com/​rheumatology-emr/​ and http://​dgrh.​de/​rheumaedv.​html) [44]. Some German outpatient clinics and rheumatologists in private practices also use computer/Web-based documentation software-systems, such as ARDIS, DocuMed.rh, and RheumaDok, that allow linkage to EHRs and/or software systems in outpatient clinics and private practices [55]. Schacher et al. examined the usability of these three systems, concluding that they provide valid data with better data quality than the paper versions [56]. Additionally, documentation software systems bear the potential to support tight-control concepts.

There is an upcoming use of ePROMs and their related applications on mobile electronic devices such as tablet PCs or smartphones [35, 37, 57]. These devices are nowadays widely spread and highly valued by individuals, usually remain turned on and are kept with the owner during the whole day; hence, they offer the opportunity of using medical applications as electronic diaries in real-life contexts [58]. mHealth mobile software programs, known as applications (“apps”) on smartphones and other wireless devices, rely on the advantages of being personal, intuitive, user friendly, and portable [57, 59]. Thus, ePROM assessment by mHealth apps can become more convenient to the technically equipped patients and liberate them from filling out questionnaires at the physician visit [17, 47]. This becomes even more important as today’s digital natives will be the rheumatology patients of tomorrow, and consequently, it is important to start developments of such useful applications as soon as possible [57].

The use of mobile electronic devices, such as tablet PCs, simplifies data acquisition at the time and location of clinical decision making and its use has been evaluated positively in rheumatoid arthritis (RA), systemic lupus erythematosus (SLE), and spondyloarthritis (SpA) [33]. Thus, mobile-accessed ePROMs are a good and capable option in routine patient care [33]. Meanwhile, further supplementary apps running on tablet PCs also allow patients direct data entry in the waiting rooms (e.g., ScoreCheck^® Rheuma, see https://​www.​grandcentrix.​net/​portfolio/​roche-pharma-patientenberatun​g/​). The linkage to existing patient documentation systems is enabled.

Smartphones also support mobile medical and public health practice and give great opportunities [57, 60]. Apps offer new forms of patient (self-)management [61]. For example, free-of-charge electronic diaries including ePROMs were developed for different devices (e.g., smartphones, tablet PCs) as well as for different operating systems (such as iOS, Android, and Windows) [62]. They allow remote ePROM documentation whenever wanted and in even more asynchrony with their patient visit to the caring physician and might lead to a fundamental change in patient–physician interaction and directly influence therapy [62]. Azevedo et al. give a current overview on smartphone apps for self-management of rheumatic diseases and related problems that at least partly include ePROMs [57].

Physical disability as a sequel of chronic rheumatic diseases might be considered as a handicap for the use of mobile devices. Nevertheless, RA patients use electronic diaries and self-report their symptoms, major restrictions were not yet reported [61‐63]. Compliance with computerized diaries is said to be much higher than the compliance with paper diaries, and additionally computerized PRO assessment can increase patients’ compliance [50, 64, 65].

Website-based systems permit the gathering of ePROMs [5, 49]. For example, the international METEOR Project (http://​www.​meteorfoundation​.​com/​) developed a Web tool that allows registered user-restricted online documentation of PROs, physician-derived parameters, and an online access for patients to look in their electronic patient record [66]. Koevots et al. reported a high interest in online self-surveillance of the disease [67]. Although less than half of the patients used the system in their feasibility and acceptability study, the authors conclude that an autonomous online registry is feasible in daily clinical practice [67]. As a prerequisite for online data acquisition a cross-sectional study including ePROMs performed via a Web-portal showed that data assessed online is equivalent to paper-based data [68, 69]. As an initial example of a new approach to tight-control concepts, Walter et al. evaluated whether tightly controlled disease activity is possible with provided online PROMs [70].

However, collected ePROMs might be useless without being reviewed by physicians or other staff members. Thus, whenever ePROMs are incorporated in the care process, data should be shown to the physician during regular personal consultations. Alternatively, secure data transfer and exchange as well as remote reviewing should be provided to the stakeholders. Further research on the development and evaluation of applications are warranted [57, 59]. This includes remote ePROM documentation via apps integrated into the popular treat-to-target strategies and other new care management concepts. Other disciplines already have described that apps might improve treatment accessibility [59].

Several studies have been performed on the large armamentarium of ePROMs available not only in rheumatology but also in other disciplines, such as cancer and neurology, that use similar PROMs as rheumatologists [41, 71, 72].

As disease-specific assessments are more common in RA and SpA, studies predominantly address these patient groups. Examples are briefly summarized in chronological order: Ryan et al. showed that 44 % of RA patients were more likely to skip a question or mark more than one answer to the given question in paper-based questionnaires when comparing a paper-based and an electronic version of the SF-36 General Health Questionnaire [36]. In patients with systemic lupus erythematosus and vasculitis, an electronic version of the SF-36 correlated well with the paper version [18]. Schaeren et al. validated the North American Spine Society outcome-assessment instrument for the lumbar spine (a valid and reliable tool for measuring the outcome in patients with low back pain) in a touchscreen format. The computerized version was as reliable as the paper–pencil version and nearly two-thirds of the patients preferred the computerized version [40]. Evaluations by Bent et al. not only showed a high degree of agreement between paper and computer-administered versions of the Quebec Scale, the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), the Bath Ankylosing Spondylitis Functional Index (BASFI), and the Bath AS Patient Global Score (BAS-G) on a computerized touchscreen system but also noticed small systematic differences for the Quebec Scale and in the BAS-G results [48].

A comparison of self-reported health status measures (pain, fatigue, and global health on visual analogue scales [VAS]; rheumatoid arthritis disease activity index; modified health assessment questionnaire; SF-36) of paper-based questionnaires and electronic versions on a personal digital assistant (PDA) in RA patients was published in 2005 [73]. The authors reported acceptable agreement between the scores and patients’ preferences of the PDA version [73]. A study with touchscreen computers showed that assessment via computer was as fast as paper-based versions [32]. The authors investigated the Rheumatoid Arthritis Quality of Life Questionnaire (RAQol); the Stanford Health Assessment Questionnaire (HAQ); VAS for pain, fatigue, and global arthritis activity; as well as a joint assessment. Their touchscreen questionnaires produced similar results to the applied paper–pencil versions; age and computer experiences did not influence the results [32].

Thumboo et al. also reported that patients preferred computerized versions of the PROMs evaluated (EQ5D, the Health Utilities Index Mark 2 [HUI2] & 3 [HUI3], and the Family Functioning Measure [FFM]) [74]. They showed that differences in the mean scores (interviewer versus touchscreen) did not reach statistical significance with the exception of the EQ-VAS. The authors concluded that computerized PROMs may have great advantages for the conduction of clinical trials and cohort studies as they may lead to smaller sample size requirements as well as reductions in cost and recruitment time [74]. Richter et al. published data on the evaluation of the feasibility of electronic data capture of Hannover Functional Ability Questionnaire (FFbH)/HAQ, BASDAI, and SF-36 using a tablet PC connected to a patient documentation system [33]. The study showed no significant differences between the electronic and the paper-based assessments [33].

In patients from the DANBIO register, PROMs on a touchscreen were investigated. The ePROMs (BASDAI, BASFI, HAQ, and VAS for pain, fatigue, and global health) generated valid results in ankylosing spondylitis and rheumatoid arthritis patients [75]. Newman et al. successfully developed a touchscreen questionnaire and a Web-based dashboard (Patient Centric Electronic Redesign [PACER]) for the collection, scoring, storing, and presentation of PROs at the point of care [76]. Recently, again BASFI, BASDAI, and the Ankylosing Spondylitis Disease Activity Score assessment in a touchscreen system was feasible, well accepted by patients, and showed good data quality, reliability, and score agreement [77].

Although kids are nowadays digital natives, reports for pediatric rheumatology are scarce [78]. Having implemented and positively evaluated a Web-based application to monitor quality of life-related problems in pediatric rheumatology, the authors recommend implementation of ePROs in pediatric clinical practice [78].

Meanwhile, ePROMs in apps used on smartphones have been evaluated: Toruos et al. showed that patients with depressive disorders are able to use the Patient Health Questionnaire-9 (PHQ-9) in an app on their personal smartphones to self-assess their symptoms of depression and that app scores strongly correlated with traditionally administered PHQ-9 scores [63]. Recently, Richter et al. compared RA patients’ mobile data entry of a set of PROs (FFbH/HAQ, RADAI) using an app on a smartphone to paper–pencil versions. The authors demonstrated that patients are able to complete ePROMs in a mobile medical app (mApp) on a smartphone and that scores obtained by patients direct data entry on the smartphone did not differ significantly from the paper–pencil scores [79]. Strengths and weaknesses of apps and mobile health in the routine rheumatology service have been summarized by El Miedany [80].

Telemedicine bridges a spatial distance and involves information technologies such as interactive audio and video communications, email, smartphone, and other forms of telecommunications technology [81]. Telemedicine respectively telemonitoring applications are complex innovations offering many evaluated opportunities: They showed positive effects on hospitalization, quality of life, and costs in other chronic diseases [82, 83]. Telemedicine developments are regarded as a healthcare alternative for patient remote monitoring even in more rural areas [84‐86].

In 2012 “telerheumatology” was promoted as a solution to the national shortage of rheumatologists in Australia, which might serve as a model for other regions with shortages in manpower [85]. Additionally, a telemonitoring approach to self-managed kinesiotherapy sessions for the rehabilitation of hand function in patients with systemic sclerosis and RA has been evaluated positively [87]. In fibromyalgia, Salaffi et al. assessed an Internet-based home telemedical surveillance system. The system effectively evaluated pain and other health outcomes. The authors concluded that telemonitoring proved to be an easy to-use solution for patient-centered data acquisition [88]. From the patients’ perspective, preliminary quality attributes important to telemedicine encounters have been published [89]. This underlines that telemedicine might provide further possibilities, but still needs further evaluation [82].

Systems need to be user-friendly and intuitive and should pose minimal burden on the staff to reach successful integration and sustained use [41, 90]. Electronic PROMs questions can be depicted in different ways, for example, from one question to various questions per screen at a time, allowance of the “not applicable/unknown options,” ticks, or radiobuttons. Software applications can be programmed differently, for example, they require an item response, represent missed items, and/or respect patients’ rights to leave questions unanswered. The resulting varying design features may result in a different answering behavior and thus might affect ePROMs’ results [41, 50]. Another biasing factor that needs to be considered is whether patients answer the ePROMs on their own or are assisted by relatives or support personnel; some PROMs offer even two different versions [91].

While most apps are developed without taking all stakeholders’ needs and requirements into account, Herschman et al. published a methodology for developing a health app for patients with systemic lupus erythematosus to generate guidance to mobile app developers [92]. Besides well-known issues for software development (as easy navigation, informational content, etc.) aspects of gamification and options for customization were identified as being relevant in the development process [92]. As patient groups may have different technology comfort levels (e.g., of the graphical user interface) design needs to be evaluated by patients and adapted [51, 93]. In addition, as discussed for Internet sites that provide information on diseases, apps need to be flagged as high quality when their intended use is to enhance the interactions with the stakeholders [94, 95].

Systems might handle missing data differently. This given situation is complicated by the fact that not all PROMs score algorithms comment on how to handle missing data. Thus, there may be errors in the calculation of the scores and this may have an effect on the clinical decision making. System validation studies are indispensible, as the user must be able to rely on correct score calculations.

Assessments of ePROMs implicate a large amount of infrastructure-related issues as data security, analytic, and practical issues. These issues may differ in the systems but include from the following: given IT infrastructure and related logistic issues at the clinic/private practice, link to patient health record or other data storage systems, data storage issues (databases locations and interactions, time periods, and copies), data access and user rights, pseudonymization issues, self-determinable data storage/withdrawal policies, and standardized analytic and reporting methods [93, 96]. Systems need to consider security issues: If ePROMs are presented to the patients on their own, secure log-ins need to be realized and levels of data encryption need to be determined as well as implemented [41]. Institutional support might influence ePRO assessment [93]. Computer-adaptive testing (CAT) allows less fixed-item testing and might improve data quality and collection efficiency [97].

IT solutions and applying stakeholders need to be aware of self-reported severe health problems: they need to be flagged by the system and/or at least reviewed and judged by qualified health personal (see earlier) and require feedback loops (e.g., for provision of personalized feedback and/or motivational support) that need to be defined as well as presented to all process participants. Systems’ automatization processes vary [41]. For example, patients may complete PRO assessments when they want to, others rely on providers’ selection of assessment frequency and topics. In addition, some systems alert patients, others do not, these different approaches result in specific education processes for all stakeholders [41].

Interactive educational work for patients about PROMs is required when the new terrain for health conversation is entered in rheumatology [98]. Patients might and will be unfamiliar with ePROMs scores depicted respectively communicated to them; thus, it is necessary to make resulting and represented scores intelligible and to inform on the consequences of the results and the minimal clinical important differences. In addition, explanation of the need for follow-ups and reminders is mandatory. (Online) education programs that need to be developed in cooperation with patient representatives might be of help and should consider European League Against Rheumatism (EULAR) recommendations for patient education [99]. The added value of the electronically generated data needs to be clarified to all stakeholders to achieve long-lasting beneficial effects on the treatment process. Other disciplines already have developed and evaluated programs for training clinicians to effectively use PRO data in routine practice [100]. However, training of the staff besides the physician and the patients is also mandatory, for example, nurses might play a key role in ePROMs’ sustained implementation and utilization [20, 93].

When implementing ePROMs into IT solutions, complex licenses and legal aspects need to be considered [101]. “Copyleft” licensing has been proposed as a solution [101].

In addition to the legal aspects mentioned in the previous “Technical and Data Security Aspects” section of this chapter, patient’s consent for electronic data transfer (e.g., via wireless/local area networks) and storage needs to be obtained regardless of the (mobile) IT solution. In general, data avoidance and minimization issues need to be taken into account, but the extent might change according to the underlying rheumatic disease and the responsible regulatory authorities [17, 102].

There is already a large number of medical apps on Apple’s App Store and Google marketplace. Although many of them would probably need to be regarded as medical devices, they rarely carry, for example, the CE mark that signifies that the product conforms with all European Union (EU) directives or EU regulations that apply to it [103, 104]. One of the main intentions of these regulations is to prevent patients from the risks and potential harm that might overcome them from apps that are not running the way they are intended.

Various healthcare systems in the developed countries put a major focus on high quality of care that includes patients’ perspectives and outcomes as “modern” quality of care indicators. Meanwhile regulatory authorities recommend PROMs assessments; even value-based reimbursements that rely on PROMs assessment in the physician office have been implemented [5, 105‐107]. Software applications that integrate such quality indicators are capable to assist in monitoring and management as they might facilitate and improve the delivery process of the required information to the regulatory authorities and other key stakeholders [29]. To encourage the acceptance of ePROMs and IT solutions in rheumatology, it is necessary to actively promote the contribution of the use of technologies to outcomes and obligations to all stakeholders [108].

PROMs supply information on health-related topics patients know best. Electronic assessments of PROMs allow bringing the patient perspective into real-time clinical routine care and facilitate patients to participate in their healthcare process immediately. The implementation of new technological developments has become more feasible. Future systems will provide opportunities for automated integration of PROMs tailored to individual needs. The broad adoption of new applications on mobile devices and their connection to existing patient documentation systems might lead to more frequent and continuous documentation of the key outcome measures and thus to new possibilities for sustained implementation of treat to target and other patient management concepts. New business models need to be developed to reimburse ePROMs’ assessments apart from the physician visit. However, all stakeholders need to align their interests and enhance their engagement in the multilateral partnership.