Toward <scp>e‐Scales</scp>: Digital Administration of the International Parkinson and Movement Disorder Society Rating Scales

Medicine is becoming increasingly digital with widespread use of electronic devices including tablets, smartphones, and online platforms for collecting health-related data.1 The digitalization of medicine has the potential to reduce the burden of care and make standard clinical data available in real-time for clinical and research purposes2, 3 while lowering study duration and costs. One consequence of this digitalization is the electronic adaptation of validated paper-based clinical scales (e-scales). The International Parkinson's Disease and Movement Disorders Society (MDS) has led the development of clinical rating scales of movement disorders for both clinical and research settings. Rating scales are standardized instruments that provide a common language to quantify disease severity and progression, measure the response to interventions, and ensure intra- and inter-individual comparability.4 A practical advantage of clinical scales is the integration of multiple manifestations into a single scoring system.5 MDS scales have demonstrated satisfactory clinimetric properties6, 7 and an important example is the MDS-Unified Parkinson's Disease Rating Scale (MDS-UPDRS).8 Despite the increasing importance of patient-reported outcome measures that reflect the functional status of the patient,9 the MDS-UPDRS has been used as primary endpoint in more than 180 clinical studies on Parkinson disease in the last 5 years (source: clinicaltrials.gov). Digitization of clinical scales must guarantee that the e-scale's output preserves the original clinimetric properties of the paper version. The use of e-scales has many advantages, such as readability, accessibility, scalability, accuracy, and completeness of data collection. Additionally, e-scales may enable easier remote administration, real-time monitoring of the data by clinicians and researchers, as well as prompt feedback to patients and study participants. Certain MDS-owned scales are clinician-rated (eg MDS-UPDRS-III or the MDS-Non-Motor Rating Scale [MDS-NMS]),8, 10 while others are patient-rated (eg MDS-UPDRS Parts 1A and 28 or the autonomic SCales for Outcomes in Parkinson's disease [SCOPA-AUT]).11 The latter category will benefit most from the remote administration of e-scales, facilitating an extension of care and research science into the home, which is also occurring in other areas of medicine.12 The MDS Rating Scales Electronic Development Committee and the MDS Taskforce on Technology have begun to promote and harmonize this “analog-to-digital” transition of rating scales and to provide recommendations for the future use of e-scales owned by the MDS. One of the first steps in the digitalization of medicine was the introduction of the electronic data capture systems (EDC) at the end of the 1960s. At that time, the scarcity of computers and their high costs limited their applicability.13 Additionally, although computers enabled the gathering of information and helped improve medical decisions, concerns that these systems could replace clinicians prevented widespread adoption.14, 15 In the 1990s the presence of more affordable computers and the introduction of online health information set the stage for wide-spread use of EDC.16 In recent years, EDC have been increasingly and efficiently used in clinical practice, industry, and research.17 In clinical settings, the digitization of clinical reports enables the integration of data for and from multi-site longitudinal databases and research studies. Information can be shared almost instantly with multiple health professionals through EDC systems, improving patient care and saving time and money.18 Furthermore, in clinical trials, digital outcome measures have been shown to be reliable and accurate.19 Electronic Case Report Forms have increased data quality and completeness, enhanced database processing, and shortened study duration.18 Further, the benefits of EDC in clinical trials have been documented by using different data collection platforms: from websites to laptops and tablets with digital pens.20, 21 For example, the commonly used EDC system REDCap22, 23 has been employed to obtain electronic consent in clinical trials, replacing paper-based consent and supporting federal guidance for electronic informed consent.24 Finally, electronically delivered survey questionnaires25, 26 and clinical diaries27, 28 have been successfully implemented in EDC systems. The disadvantages of EDC include the start-up time and costs for development and implementation, and the lack of paper source documentation against which to verify miscoded data.17 Additionally, some EDC are online applications which cannot be deployed in localities where network coverage is limited.29 One benefit of the digitalization of health care is the increasing use of e-scales. Surveys conducted by our group30 and others,31-33 have highlighted that many of the known advantages for EDCs translate to e-scales, including: (1) easier data collection and handling of missing data, (2) elimination of double-data entry and associated costs, (3) reduction of transcription errors, (4) promotion of accurate and complete data collection, and, most importantly, (5) easier remote administration of the scales and data monitoring, as well as real-time data transmission. Further, e-scales can be used in clinic practice and enable both the input of new e-scale data as well as providing simultaneously access to prior evaluations. These benefits have also been shown, for example, in a phase three pimavanserin clinical trial where the psychosis scale (SAPS-PD) was administered by a blinded, central rater to minimize variability in the assessments.34 The advantages of remote evaluation have been amplified by the recent COVID-19 pandemic where many trials have had to interrupt in-person evaluations.35 The transition to e-scales needs to be done carefully in order to preserve the clinimetric properties of the original scale. Scales have already been transformed into e-scales for implementation in research settings.36 However, differences in administration from the previously validated, paper-format, including the formatting and user interaction, can affect the interpretation, validity, and subsequent quality of the data.37 In the following sections, we suggest a framework for the digital development of MDS Rating Scales (Fig. S1). Clinimetric properties of the e-scales should be comparable to those of the original scales with the adherence to the guidelines set forth here. It is important that the wording and presentation of the original version is maintained as much as possible in the electronic format.37 Previous studies have shown high correlation between e-questionnaires/scales and the paper versions, with similar levels of sensitivity and specificity.38 Nonetheless, factors associated with the mode of administration may undermine reliability and validity as the interview is transferred from one mode of administration to another (Table 1). The design may not be as flexible as developers might prefer because e-scale adaptation must adhere to the original instrument's properties for the reasons explained above. However, some modifications are needed to adapt the scale for electronic use39 and illustrative examples of acceptable modifications are presented in Table 1. (1) Minor changes in format (eg use of larger fonts or high contrast colors) (2) Minor changes in wording in text intended for the administrator or subject that do not alter interpretability (e.g., using “select item” instead of “underline item”) In the next paragraphs we highlight some of the essential needs to account for in the conversion and implementation of MDS-owned scales into an electronic format. As care and treatment strategies become increasingly patient centered, e-scale development needs to take into greater consideration both the physician and the patient's user perspective. This is important for the patient's emotional reaction and attitude to the assessment experience. The concept of experiential equivalence (the way the test-taking process is experienced by the test taker) is important as it may modulate the patient's willingness to participate in further assessment sessions.40 Straightforward instructions on how to complete the e-scale and the anticipated length of completion are critical. Ease-of-use will contribute to the successful completion of e-scales by patients via remote administration such as the home environment. Thus, an e-scale requires clear interface, careful consideration of the number of questions per screen, large fonts to read and interact with, high contrast colors, and clear and concise language (critical for newly developed instruments; existing scales will need to replicate word-by-word language for e-use)30, 41 (Fig. S1 and Table 1). A considerable benefit of e-scales is that raters can bridge the communication gap that exists with patients who have limited communication skills in English. E-scale software needs to accommodate properly validated translations of the scale allowing patients to complete scales in their native language and then translating the patient's responses back into English for the clinician's review when needed (Fig. S1). An accessible interface would ideally provide a “progress report” numerically or visually during e-scale administration. A reminder system would also be useful for reviewing the information before completing the assessment and submitting the data (Table 1). The implementation of e-scales requires simplified training for their administration to ensure data validity and integrity, and match the technological literacy of the user. Therefore, when designing e-scales it is important to ensure that they can be used with minimal training and supervision to facilitate their integration into clinical and research practice through widespread portable devices (e.g., tablets). The impaired ability of movement disorders patients to navigate an application needs to be considered for scales that are self-administered. Therefore, using a computers or tablet rather than a smartphone as the primary platform for e-scale administration would facilitate its use30 (Table 1). Additionally, the implementation of voice recognition could partially overcome this problem,42 namely in those patients with tremor, although, it is worth emphasizing that many patients with Parkinson's disease or other movement disorders are troubled by speech disorders. The use of e-scales may be a challenge in patients who have non-motor phenomena such as visual and cognitive difficulties; however, it is noted that these non-motor symptoms may equally affect paper-based assessments. As such, the development of e-scales should ideally take into account that a third person (caregiver) can support the electronic administration of a scale, as often found with paper-based assessments. It is also important to guarantee the participation of patients which these manifestations in studies conducted electronically and that these barriers are accounted for. These practical aspects may be particularly relevant when paired with complementary mobile health technologies that include wearable sensors such as smart devices and other monitoring technologies.43 Increasingly, smart devices are used by clinicians for the evaluation of disease status. The naturalistic data collection with these devices immersed in a patient's daily life may translate into reduced need for direct patient input and could also help overcome these barriers.26 It is important to note, the use of electronic scales poses challenges and risks. In particular, the lack of real time supervision could lead to inappropriate use or missing/inappropriate data. An additional risk to data security is of paramount importance when implementing e-scales, and one must ensure that adequate privacy and data protection standards are met (HIPAA/GDPR for the US and EU respectively44 and similar regulations in other jurisdictions). For example, e-scale software needs to ensure compliance with local regulatory standards and strict data security, but still allow clinicians and study investigators access to these data. These risks should be carefully monitored and controlled in future studies using any electronic instruments. Finally, e-scale applications need to have an audit trail for changes when used for research in certain regulated settings such as clinical trials (i.e. 21 CFR part 11 FDA regulation for Electronic Records). To illustrate the process of digitalization, each of the original MDS scales (eg MDS-UPDRS, MDS-UDyRS, QUEST, MDS-NMS, NMSQ, UMSARS) were reproduced in an electronic format. For this purpose we developed a visual and content representation of each section of the scale using HTML language. Each representation included the font size (12 points), the structure of the questions (option buttons with associated categories), the other font features (eg bold, underlined, spacing, among other features), and more importantly, ensured that all the content was a word-by-word representation of the original scale without substantial changes. An example of a minor modification comes from the MDS-NMS, in which the score categories were explained prior to each questions and not at the beginning of the scale as found in the paper version. These changes were intended to facilitate the electronic administration and were determined to be unlikely to affect the scale properties according to the guidelines presented in Table 1. The main challenge of this process was to ensure a similar visual representation of the scales's original format. Once the scale was developed, the electronic build was reviewed by the MDS-Rating Scale program Chairs and additional MDS staff who provided comments during Pilot Testing I. Design and formatting issues were identified in this phase, for example, there were extra line breaks in the scale text and bolded formatting on some phrases was missing. Once these errors were corrected the scale was used by external users in Pilot Testing II. Additional errors were reported by the external users (developers). For example, there were some issues when using the first version of the MDS-UPDRS REDCap template in a different geographic location to where it was developed. This issue was due to the different writing systems used among these two locations. Unicode standard was implemented to ensure an appropriate representation of the scale in different countries and areas. Once those errors were corrected, a fully functional electronic version of the assessment was produced. Future studies are being planned to evaluate the equivalence of the paper and electronic versions to confirm the comparability of their properties. Additionally, based on the proposed framework (Fig. S1), usability testing study with movement disorders patients needs still to be done to ensure the usability of the electronic version of the MDS-own scales. To leverage the value of previously developed MDS scales for the digital era and assist movement disorders specialists, health-care providers, researchers, patients and caregivers, we proposed a framework for the standardized digitization and faithful paper-to-screen reproduction of MDS-licensed scales (Fig. S1). By following this step-by-step framework for the design, testing, production and implementation in each scale, we will be able to ensure an accurate electronic representation of MDS-owned e-scales. The initial framework has been established on REDCap to ensure all the e-scale requirements described above are met (Fig. S1). The MDS has developed a portfolio of e-scales in this platform (e.g., MDS-UPDRS, MDS-UDyRS, QUEST, MDS-NMS, NMSQ, UMSARS) and will continue to develop additional scales. The REDCap versions of the MDS scales are available on the Society website and applicable copyright and appropriate licensing arrangements can be found there. (1) Research Project: A. Conception, B. Organization, C. Execution; (2) Statistical Analysis: A. Design, B. Execution, C. Review and Critique; (3) Manuscript: A. Writing of the first draft, B. Review and Critique. M.H.G.M.: 1C, 2A, 2B, 2C, 3A, 3B R.L.M.F: 1A, 1B, 1C, 2A, 2C, 3B E.C.: 1A, 1B, 1C, 3B T.A.M.: 1A, 1B, 1C, 3B A.H.T.: 3B J.C.S.: 3B S.A.:3B L.M.C.:3B K.D.: 3B C.F.A.: 3B J.Y.: 3B B.B.:·3B F.B.H.:3B A.M.:3B R.R.:3B S.S.P.: 3B E.R.D.: 3B A.J.E.:·3B P.M.M.: 1A, 1B, 3B G.T.S.: 1A, 1B, 3B A.S.F.: 1A, 1B, 1C, 2A, 2B, 2C, 3A, 3B Ethical Compliance Statement: The Clinical Research Ethics Committee of Hospital Universitario HM Puerta del Sur approved the study (18.06.1256-GHM) in June 2018. The participants provided written consent for participating in the interviews referenced in the website and cited in this article. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this work is consistent with those guidelines. Funding Sources and Conflicts of Interest: This work was covered under the MDS Rating Scales Program committee expenses. All authors have no conflicts of interest to declare. Financial disclosures for previous 12 months: MHGM is employee of the Autonomous University of Madrid and HM-Puerta del Sur University Hospital. RLMF is an employee CHDI Management, Inc. EC has served on the advisory board for Abbvie and Allergan. TAM has served on the advisory board for AbbVie; has received honoraria from International Parkinson and Movement Disorders Society, University of Ottawa, AbbVie, and Ipsen; has received grant support from University of Ottawa Medical Associates, Parkinson Canada, Parkinson Research Consortium, LesLois Foundation, PSI Foundation, University of Ottawa Brain and Mind Institute, and Michael J Fox Foundation; and has been employed by University of Ottawa Medical Associates. AHT has received honoraria from Novartis and grant support from the Toray Science Foundation. LMC research support from the Michael Fox research support for a clinical trial by and from reports with University Medical is an employee of as for the of Parkinson's has received honoraria from on the advisory board for AbbVie, and has received for at from AbbVie, and and has received research support from the for the Michael Fox Foundation, AbbVie, the Parkinson the the Parkinson's Foundation, the and and the Parkinson has interest in a with in for movement disorders. This has been reviewed and by has received grant support from and as for and received support from the Development has received grant support from and has received honoraria from and is and of the Institute, a research on clinical and research in and a clinical research providing in clinical trials and provided advisory board clinical trial for and the Disease Foundation. reports from and by the University of has received honoraria for at of and University of received for from AbbVie, Associates, of Disorders and Research Associates, and research support from AbbVie, Foundation, and Foundation, Michael Fox Foundation, of of Disorders and Science Foundation, Outcomes Research Institute, Foundation, University of for and with integration and or received from the of of the of of Michael Fox Foundation, Foundation, and received honoraria from and was to of and and is an advisory board of the for Parkinson's as the of the MDS Technology has received grant support from the and the Michael J Fox Foundation; as a advisory board for and from University and and honoraria from and reports from for the study of the Scale for of Disorders in Parkinson's Disease and from the International Parkinson and Movement Society for the Pilot of the MDS-Non-Motor reports from the International Parkinson and Movement Society for of missing and of the has received from the Program for and number has also received and honoraria from and Novartis common in a with in a Medical developed for is also an of a and for that has been to an by the of S1 for the development of e-scales. for and implementing e-scales. It a into steps and The Design in the of an electronic build of the scale, by using a HTML the information of the scale. The electronic format of the paper-based scale needs to ensure the of the scale, to ensure of data and to its clinimetric properties. The Testing two different Pilot testing is the first of the design by the developers to ensure with the original scale. This process that the design and content of the e-scale are to the original The next is Pilot testing which is done by users in a controlled environment. The the content by the Movement The last is the of the At each of these any with the format is it would take users back to the Design to The is not for the content or of any supporting information by the than missing should be to the for the article.