Mobile Health Usage, Preferences, Barriers, and eHealth Literacy in Rheumatology: Patient Survey Study

Background Mobile health (mHealth) defines the support and practice of health care using mobile devices and promises to improve the current treatment situation of patients with chronic diseases. Little is known about mHealth usage and digital preferences of patients with chronic rheumatic diseases. Objective The aim of the study was to explore mHealth usage, preferences, barriers, and eHealth literacy reported by German patients with rheumatic diseases. Methods Between December 2018 and January 2019, patients (recruited consecutively) with rheumatoid arthritis, psoriatic arthritis, and axial spondyloarthritis were asked to complete a paper-based survey. The survey included questions on sociodemographics, health characteristics, mHealth usage, eHealth literacy using eHealth Literacy Scale (eHEALS), and communication and information preferences. Results Of the patients (N=193) who completed the survey, 176 patients (91.2%) regularly used a smartphone, and 89 patients (46.1%) regularly used social media. Patients (132/193, 68.4%) believed that using medical apps could be beneficial for their own health. Out of 193 patients, only 8 (4.1%) were currently using medical apps, and only 22 patients (11.4%) stated that they knew useful rheumatology websites/mobile apps. Nearly all patients (188/193, 97.4%) would agree to share their mobile app data for research purposes. Out of 193 patients, 129 (66.8%) would regularly enter data using an app, and 146 patients (75.6%) would welcome official mobile app recommendations from the national rheumatology society. The preferred duration for data entry was not more than 15 minutes (110/193, 57.0%), and the preferred frequency was weekly (59/193, 30.6%). Medication information was the most desired app feature (150/193, 77.7%). Internet was the most frequently utilized source of information (144/193, 74.6%). The mean eHealth literacy was low (26.3/40) and was positively correlated with younger age, app use, belief in benefit of using medical apps, and current internet use to obtain health information. Conclusions Patients with rheumatic diseases are very eager to use mHealth technologies to better understand their chronic diseases. This open-mindedness is counterbalanced by low mHealth usage and competency. Personalized mHealth solutions and clear implementation recommendations are needed to realize the full potential of mHealth in rheumatology.


Introduction
Rheumatoid arthritis, psoriatic arthritis, and axial spondyloarthritis are complex pathogenetic chronic diseases. These disease entities also require a complex treatment structure with interdisciplinary care by various specialists (rheumatologists, dermatologists, gastroenterologists, ophthalmologists, physiotherapists, etc), and intensive and regular disease monitoring is essential. With this complexity, affected patients often report a lack of understanding of their disease [1].
Mobile health (mHealth) holds promise to improve health care delivery and outcomes for people with chronic diseases [2]. Ideally, mHealth solutions, such as mobile apps and wearable sensors could empower patients, provide individual support, and lead to better outcomes than those available through standard care. Patients with chronic rheumatic diseases already have access to a broad range of mHealth solutions, starting from symptom checkers [3] or referral tools [4]. Once a diagnosis is established, mHealth tools enable patients to better monitor their symptoms passively through sensors [5] and actively by entering data [6,7]. Furthermore, electronic medication reminders can increase medication adherence [8,9], and supporting digital therapy can reduce pain [9] and improve important comorbidities (eg, depression) [10].
Recently the European League Against Rheumatism and Working Group Young Rheumatology of the German Society for Rheumatology (Arbeitsgemeinschaft Junge Rheumatologie der Deutschen Gesellschaft für Rheumatologie) published recommendations for the development of mobile apps in rheumatology [11,12]. The early integration of patients in the app development process was stressed in both papers. However, little is known about the patient perspective on mHealth solutions, as current literature focuses on rheumatologists [11,13,14]. To successfully integrate the various mHealth solutions into clinical routine, it is essential to identify barriers and the needs of patients.
The aim of this study was to explore mHealth usage, preferences, barriers, and eHealth literacy reported by German patients living with rheumatic diseases.

Methods
Between December 2018 and January 2019, consecutive patients seen at one rheumatology outpatient clinic of the University Hospital Erlangen were asked to complete a paper-based survey. This study was approved by the Ethics committee (No. 418-18B) and conducted referring to good clinical practice. All patients provided informed consent.
To create the survey (Multimedia Appendix 1), a broad literature review was carried out. Previous mHealth patient surveys in oncology [15] and rheumatology [16] served as a starting basis. The survey comprised four main parts: (1) sociodemographics and health characteristics, (2) mHealth preferences and usage, (3) eHealth literacy, and (4) communication and information preferences. The survey was pilot-tested with 10 patients to detect necessary formatting and wording changes. Minor revisions were made accordingly. Foreign words and technical term explanations were provided in a footnote. Inclusion criteria were patients (1) aged ≥18 years, (2) who were literate in German, (3) who had the physical and mental ability to fill out a structured questionnaire, and (4) who fulfilled classification criteria for rheumatoid arthritis [17], axial spondyloarthritis [18], or psoriatic arthritis [19].
The sociodemographic and health characteristics included age; gender; residence; diagnosis; disease duration; patient global assessment of disease activity; and current usage of smartphones, tablets, activity trackers, and social media.
The mHealth preferences and usage section included questions about the preferred time and frequency for using a rheumatology app. Patients were asked to rate their preference of app features (5-point Likert scale) and rate the importance of app characteristics (10-point Likert scale). Inquiries were made on internet usage and perceived usefulness, willingness to share recorded app data, general perception about the utility of medical apps, web-based services for improving patient's health, and telemedicine.
Patients' eHealth literacy was measured using the validated German version [20] of the eHealth Literacy Scale (eHEALS) [21]; eHEALS has been translated and validated in multiple languages [20][21][22]. It is based on a 5-point Likert scale and includes 8 statements concerning self-perceived eHealth literacy.
Patients were asked to state their preferences concerning medication reminders, medical information format, digitally provided information structure, patient diary type, and physician communication type. Rankings did not have to be unique. Characteristics were summarized using means, standard deviations, counts, and percentages as appropriate. We used Pearson correlation to explore relationships between continuous variables. The relationship between the eHEALS score and internet use frequency was examined using a linear regression model with the eHEALS score as the dependent variable and internet use as an ordinal predictor encoded using orthogonal polynomials to characterize nonlinear effects. Relationships between eHEALS score and binary preferences were examined using logistic regression. All models included age and gender adjustments. Two-sided P values less than .05 were considered significant. We used Excel (Microsoft Corp) and R (version 3.5.3; R Foundation for Statistical Computing) for data manipulation and analyses.

Patient Characteristics
In total, 224 patients were recruited. Only complete surveys (N=193) were considered for the final analysis. The number of patients rejecting participation was not measured. The study sample's demographics are shown in Table 1

Medical App Acceptance and Willingness to Provide mHealth Data for Research Purposes
Preferences and attitudes regarding potential mHealth apps and data flow were addressed ( Table 2). More than two-thirds of the patients (132/193, 68.4%) believed that medical apps are helpful for their health; however, only 4.1% (8/193) patients currently used medical apps, of which none were rheumatology specific apps. Increasing eHEALS scores were associated with a higher probability of expressing belief that apps were helpful after adjusting for age and gender (odds ratio [ Measured on a scale of 10, the most important app characteristics were security (mean 8.9, SD 2.5) and usability (mean 8.5, SD 2.5) ( Table 2). Regarding preferred app functions, patients were most interested in information about medications and diseases and were least interested in direct exchange such as chats with peers with the same disease ( Figure 1).

Internet Usage and Perceived Usefulness
To address habits regarding information search, patients were questioned concerning internet usage. The majority of patients (168/193, 87.0%) had previously used the internet to obtain health information (     Table 4 shows the distribution of responses to the 8 eHEALS items. The majority of the patients agreed that they know how to use the internet to answer their questions about health (82/193, 71.5%), and a considerable proportion of patients (82/193, 42.5%) felt uncomfortable using information from the internet to make health decisions. A lower eHEALS score was associated with a decreasing frequency of internet usage; the regression analysis shows that, after adjustment for age and gender, the negative association with decreasing frequency of internet use and eHEALS score followed a second-order polynomial (linear: -8.87, 95% CI -11.97 to -5.76; quadratic:-5.64, 95% CI -8.37 to -2.92) meaning that each step of decrease in the use-frequency categories was associated with, not an equal, but a progressively greater decrease in the eHEALS score ( Figure 4). Figure

Main Findings
Patients with rheumatic and musculoskeletal diseases are ready and willing to use mHealth technologies. Patients believe in the potential of mHealth and are open to participating actively by sharing health data with their physicians and researchers. Structured electronic data acquisition holds great promise to increase data quality, increase quantity, and reduce missing data and bureaucracy. Patients preferred filling out questionnaires before clinical visits, using video calls, and receiving electronic doctor's letters. These measures could drastically improve the clinical experience for patients, cutting down waiting time and long drives to the hospital. For clinicians, automatic data import could reduce administrative efforts. Furthermore, continuously obtained mHealth patient data increases the basis for shared personalized clinical decision making. Some patients, however, prefer personal contact with their physician and were concerned about data storage and transfer. Also, at the time of the study, no single patient was using a rheumatology specific app, only 4.1% were using medical apps at all, and 11.4% were aware of useful rheumatology websites or apps. The currently scarce mHealth usage and low eHealth literacy highlight an important need for structured mHealth guidance and patient-adapted information and education. The majority of patients clearly stated this by calling for official app recommendations from the national society of rheumatology. The most popular app features were information about medication and rheumatic diseases. The fact that 87.0% previously searched for various health information on the internet, and that the internet is the most frequently used source for health information, further supports the need for more, better, and personalized medical information. This information should be accurate and reliable, requiring rheumatologists and societies to actively lead and supervise mHealth in rheumatology. The opportunity to support therapeutic online health programs is currently not being fully utilized, despite promising study results [23]. Furthermore, our study shows an enthusiasm on the patients' side to improve medication adherence via mobile apps and diaries. Sharing of this mHealth data with a large research registry would represent an enormous potential to improve treatments and rheumatology research. These results may help patients, developers, clinicians, and researchers to use the full potential of mHealth in rheumatology.

Limitations
The cross-sectional design, self-reported data, and sampling method with relatively small sample size were the most important limitations of this study. Therefore, the results might not be generalizable, and actual mHealth usage might, for example, differ compared to self-reported usage.

Comparison With Prior Work
To our knowledge, this is the first work depicting a German mHealth patient perspective in rheumatology. This work may inform mHealth policy recommendations and adds to the growing body of eHealth rheumatology knowledge [11][12][13][14]24,25] by providing detailed patient preferences, needs, and barriers. Hence, we believe that the results of this study could help in devising mHealth solutions that can be integrated into the clinical routine of patients with rheumatic diseases. The importance of including patients in the app development process is stressed in various recommendations [11,12]. Rheuma-Auszeit was reported as the only app developed with major patient involvement and scored as the highest quality app [11]. It was also shown that the participation of patients in app development leads to high usage; for example, the ArthritisPower app was used by >18,000 patients with rheumatic diseases in 2018 [26].
Health literacy is critical for patient empowerment [27]. Our work confirms the lack of adequate eHealth skills in rheumatology patients [28], reflected by the borderline mean eHEALS score of 26.3 out of 40. Cutoffs for the eHEALS score show considerable variation in the literature [29][30][31]; however, a value of 26 was set in a previous study as the cutoff for low eHealth literacy [31]. eHealth literacy was associated with a greater belief in the usefulness of medical apps, and usage of the internet to obtain health information, similar to findings reported by Noblin et al [32]. A previous study [33] showed that younger age and higher eHealth literacy correlate with perceived effectiveness of medical apps. eHealth literacy can change over time [34], and supporting programs should be implemented to increase eHealth literacy [35]. Many mHealth challenges could be overcome if more support was provided by health providers [36].
Patients demanded clear recommendations from the national society for rheumatology and expressed the highest confidence in an app developed by a rheumatic disease scientific society [25]. Our results suggest that patient app entries should take no longer than 15 minutes and should not be requested more often than weekly. Data transfer should be clearly explained to patients to eliminate this barrier. Customizable app features are needed, as preferences differ, and a one-size-fits-all approach seems to be less effective. App building blocks (videos, information, tools) should be provided so that redundant work is reduced, and a large variety of features can be provided. Ideally, the app content should be created in a joint effort by all stakeholders.
The discrepancy between app use and general belief in usefulness and interest reflect the currently unmet need of effective mHealth solutions, guidance, and education. This discrepancy is also highlighted by the dominance of conservative communication preferences. These results should carefully be interpreted, as most patients currently do not have experience with app-based communication.
In an effort to address this need, the French society for rheumatology recently developed the free patient app Hiboot, which provides patients with trustworthy information about medication and answers to frequently asked questions [37]. App usage was very low among patients (4.1%) compared to medical app usage among German rheumatologists (49%) [13] and an international group of rheumatic patients (79/394, 20%). In contrast to rheumatologists [13], and an international group of rheumatic patients (188/394, 47%) [25]; patients were not aware of any rheumatology specific apps. These results could partly be explained by the paper-based nature of the survey that enabled us to include elderly patients that might have been reluctant to join the study otherwise.
The lack of patient interest was previously identified as a major barrier to app prescription among general practitioners [38]. Besides education and careful evaluation of current mHealth solutions [11], solid evidence for the effectiveness and usability is needed to overcome current barriers and increase app prescription rate.
Concerning app feature preferences, our results were very much in line with previous research [16,25]. Patients do not prefer patient-to-patient communication features; however, patients are very interested in other app features, particularly those providing information.
As reported in previous studies [25,28], our work clearly shows that most patients regularly use the internet to retrieve medical information. In a previous study [3] published in 2016, 47% of the patients consulted the internet to investigate their symptoms. This proportion seems to be increasing, as in our study 67% and in another recent study [25] 95% of patients stated they did so. Compared to German oncology patients interviewed in 2016 [15], more rheumatic patients seem to use mobile devices regularly (69.6% compared to 91.2%); however, the information quality is very heterogeneous which could be misleading.
The obvious enthusiasm of patients to share app data for research purposes and their wish for such data to be made available to the treating rheumatologist for routine care underlines that an improvement to the science and practice of rheumatology could be realized using mHealth. A digital approach allows the inclusion of well-structured patient-generated data to improve clinical shared decision making and clinical research [39]. The interoperability of systems is crucial for the success of such mHealth tools. A part of patients want rheumatologists to review their app input. This could easily cause information overload and an additional workload to physicians; however, rationing appointments, earlier detection, and response to flares would also be possible [40]. In combination with wearable sensors and algorithms, patient monitoring could further be improved.
The medical community, public health system, and private sector need to increase their efforts to improve eHealth competencies and to provide safe and effective digital tools to leverage the way of mHealth into routine rheumatology care.

Conclusion
To our knowledge, this is the first study capturing a detailed mHealth perspective of patients with rheumatic diseases, which could guide rheumatology app development and implementation. Most patients included in this study possessed smartphones and believed that using medical apps could be beneficial for their health. A substantial majority was also willing to share app data for research purposes. The current usage of mHealth among rheumatic patients is, however, very limited and eHealth literacy was rather poor. We could successfully identify unmet needs and patient priorities, which can be used to accelerate and guide the way of mHealth into routine rheumatology care.