Virtual Cognitive Screenings and Interviews of Patients with Neurodegenerative Conditions Associated with Alzheimer’s Disease and Parkinson’s Disease

The current pandemic of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), commonly referred to as COVID-19, brings myriad challenges to research conducted among those more susceptible to the virus. According to the United States (US) Centers for Disease Control (CDC), eight out of ten reported COVID-19 deaths are among people > 65 years of age and older. Nonetheless, researchers must continue the crucial work of investigat-ing and understanding diseases that affect the elderly. The focus of this white paper is to assess the challenges associated with research within the elderly population with neurocognitive conditions. Specifically, this paper addresses the need for the standardized administration of performance measures (e.g., neurocognitive assessments) among a dementia population while ensuring the physical safety of participants. Consideration is given to the administration of performance measures and the availability and feasibility of adminis-tering these measures remotely to a population that may have difficulty using novel technologies. In implementing remote research assessments, it is suggested that researchers follow a GAMMA approach by: 1) establishing clear Guidance on remote visit expectations and processes; 2) establishing Appropriate exclusionary criteria in the development of the study design; 3) providing subjects Appropriate study Materials for visual processing; 4) incorporating Multiple data sources in the overall study design (e.g., caregiver input); and 5)


Introduction
Due to the widespread nature of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the number of deaths globally associated with COVID-19 continues to increase [1]. The groups most vulnerable to the virus in society are the elderly and those with underlying health conditions, resulting in higher mortality among these groups [2] [3] [4]. According to the World Health Organization [5], there are approximately 50 million people living with dementia worldwide, with 10 million new cases every year; this accounts for approximately 5% -8% of the general population over the age of 60 years [6]. While the elderly population may be particularly vulnerable to COVID-19, the need continues for research into Alzheimer's disease (AD) and other forms of dementia.
The United States (US) Food and Drug Administration (FDA) has published Guidance for the conduct of clinical trials during the COVID-19 pandemic [6].
The FDA guidelines recommend utilization of alternative methods of completing study visits, for example, virtual visits held by telephone, to ensure the safety of vulnerable participants. Since it remains crucial to collect screening and performance measures (e.g., the Dementia Severity Rating Scale (DSRS) [7], Clinical Dementia Rating (CDR) [8], and Mini-Mental State Examination (MMSE) [9]), and to evaluate disease severity, other options are needed for collecting qualitative and performance data from the dementia population and their caregivers. In addition to assessing cognitive performance using performance-based outcomes (PerfOs), outcomes data are often collected from these patients using other clinical outcome assessment (COA) measures to assess how they are feeling as well as their level of functioning and quality of life (QoL).
The use of virtual technology in research is becoming both more necessary and more acceptable. Due to the characteristics of dementia such as disturbed emotions, perception, motor activity, mood [10], and cognitive impairment [11], and since populations with cognitive issues may face additional challenges when moving away from traditional, in-person data collection, there are considerable hurdles to overcome and logistical complications to consider.
Virtual and telephone interviews have been used for some years in qualitative data collection among patient populations [12] and caregivers, for example in amyotrophic lateral sclerosis (ALS) [13]. A modified 26-item version of the MMSE has been validated for telephone use (TMMSE) [14] for patients with AD; the TMMSE omits items that are not possible to confirm during telephone completion (i.e., "What floor of the building are we on?") and adds new or revised items for the telephone administration (e.g., the examiner asks the patient  to follow a three-step command, "Say, 'hello,' tap the mouthpiece of the phone 3 times, then say, 'I'm back.'") [15]. Educational information has been delivered to caregivers of patients with dementia via the telephone for support in self-efficacy and assessment of caregiver burden [16]. This paper will discuss the use of technology among patient populations with AD, Parkinson's disease (PD), and mild cognitive impairment (MCI), and the use of remote data collection, specifically qualitative interviewing and performance tools, to assess disease severity.

Methods
Targeted Clinical Trial Review A review of ongoing and completed clinical trials conducted in neurodegenerative conditions including AD, PD, and MCI populations was conducted using the FDA clinicaltrials.gov database in April 2020. The search aimed to identify COA measures being used in these populations. Three separate searches were conducted using the following terms: "Alzheimer's disease (AD)," "Parkinson's disease (PD)," or "mild cognitive impairment (MCI)," Phase II and III, 2017 to present (past 3 years), adult, and older adults. Figure 1 illustrates the process used to determine the measures for review in the current paper.
We also searched products with patient-reported outcome (PRO) claims that had been approved by the FDA and European Medicines Agency (EMA), using the FDA and EMA databases to identify COA measures utilized to support label claims in AD, PD, and MCI.
Review of Electronic Platforms A manual internet search of available electronic data platforms and videoconferencing systems was performed using the terms "videoconferencing platforms" and "web-based platforms." Each identified platform was reviewed for suitability for use in data collection among patients with neurodegenerative conditions. Information about previous use of the platforms included whether they have been used in healthcare or research settings, and HIPAA compliance was assessed for each (Table 1). Other information such as video sharing capability, screen sharing, file sharing, technical support, and security was collected for each platform. The existing FDA Guidance allows use of software in the context of health care and research. To protect patient privacy, remote data acquisition, transmission, and storage must be secure. This includes electronic platforms designed for remote assessments that are required to have automated audit trails [6]. Although platforms for remote data collection are not considered to be medical devices as they are not specifically intended for use in the diagnosis, treatment, or prevention of disease, they must provide a secure way of communication [17]. When considering an electronic platform, it is critical to review its compliance with HIPAA requirements to ensure that the electronic health information collected through the software is protected. were identified as having been used most frequently across the three conditions, and were selected for further review of their suitability for completion in a virtual/remote, non-face-to-face setting with technological devices ( Table 2 and Table 3 measures were also mentioned as part of label claims in these conditions. Each of the sixteen COA measures was assessed for suitability for virtual/remote completion. Props and tools required as part of the assessment (i.e., paper, pen, stopwatch, and objects for item recognition) as well as worksheets and scoring methods were reviewed in detail.

Review of measures
Verbally Administered Measures for Virtual Administration The Clinical Dementia Rating Scale Sum of Boxes (CDR-SOB) is a global assessment dementia screener used to assess severity of dementia using a 5-point scale system. The CDR-SOB does not require the use of props for administration; the interviewer conducts separate interviews with both the participant and the study partner (e.g., caregiver), recording their responses for comparison using the score sheet. The CDR-SOB can be conducted via telephone or web-conferencing and, in a recent study, showed no significant statistical difference from in-person administration [15]. The Logical Memory Test (LMT), Hopkins Verbal Learning Test (HVLT), Free and Cued Selective Reminding Test (FCSRT), and Controlled Oral Word Association Test (COWAT) are widely used measures designed to assess episodic memory, verbal learning, and verbal   The ADAS-Cog has been conducted remotely in older adults with and without cognitive impairment successfully [20], with strong correlation between in-person and web-conference data collection in those with MMSE < 17. However, potential barriers could include access to and ease of use of technology [21]. Use of web-conferencing and cameras could enable props such as word cards to be shown to the participant. The 26-item TMMSE has been previously utilized in studies where data were not collected in person and showed strong correlation with the in-person administration of the MMSE [14].

Device-based Measures for Virtual Administration
The Cambridge Neuropsychological Test Automated Battery (CANTAB) and those from Cogstate Ltd are designed to detect changes in neuropsychological ability and are conducted electronically using a tablet computer provided to the participant during the interview. Devices could be shipped to participants ahead of the interview provided that participants are able to log into and use the device without assistance from the interviewer. Alternatively, a caregiver could receive training on how to set the assessment up for the participant. One potential risk is that sponsors and researchers may be loath to send out valuable hardware that may not be returned in a timely manner.
The MoCA has been modified for virtual use by the developer, with telephone, video-conference, and app-based options available [22]. The telephone option removes the visual elements of the test, while the video-conferencing and app options allow administration of the full measure [22]. A recent study among an elderly population of veterans indicated that in-person and video-administration methods showed comparable accuracy and inter-rater reliability [23] [24].
Additionally, the RBANS has been validated for virtual administration through the use of video-conferencing and interfacing iPads [23] [25]. Analyses have shown that scores obtained via virtual administration are highly correlated with those obtained via face-to-face administration, with minimal differences in measure index scores [25] [26].
Phone App Measures for Virtual Administration The 9 Hole Peg test and the 6-Minute Walk test are performance-based measures designed to detect hand dexterity and aerobic capacity. These measures are not widely available for remote completion; the challenge remains that they require equipment and observation by an interviewer. However, the 9 Hole Peg test has been developed for use in an app [27] using Apple's ResearchKit [28] and tested for use in patients with multiple sclerosis.

Review of Electronic Platforms
Online data and videoconferencing platforms were also searched using a manual internet search, and were reviewed for suitability for use in data collection in this population using a checklist developed by the study team (

Discussion and Conclusion
Our review indicates that there are indeed viable performance and other outcome measures available for virtual administration among a mild-to-moderate dementia population. Furthermore, secure data-transfer platforms are available to ensure data privacy. With the rapid rise in smartphone and computer use, global technology continues to advance and bring new opportunities for research and data collection. However, the elderly are often at a disadvantage when presented with new technology due to a lack of fluency with, or apprehension about, devices and program applications [31] [32]. In light of the current pandemic, it is incumbent upon researchers to provide easily accessible platforms and procedures to ensure the physical safety of this vulnerable population during study assessments, while concurrently ensuring that the assessments are conducted in a standardized fashion. These considerations can be adequately met by employing a combination of strategies, referred to by the authors as the GAMMA Approach. This approach is described subsequently and is a suggested guide for researchers as they implement remote research assessments among patients with dementia.
Guidance on expectations-Prior to assessing and/or interviewing patients, researchers should provide clear Guidance on what the patient can expect and how the assessment and/or interview will be conducted. A document outlining procedures and any technological requirements should be provided to patients well in advance of the study interaction. If possible, the document should also include screenshots and a step-by-step guide to ensure patients can easily navigate any technology that they are required to use and/or set-up prior to the assessment and/or interview. Some recent qualitative studies in older, healthy adults have shown that older populations are interested in adopting new technology [33] [34], but are hesitant to do so due to how quickly technology changes and concerns with data privacy [33].
Appropriate Exclusionary Criteria-Importantly, researchers must ensure Appropriate exclusionary criteria are used. Recognizing the severity of the disease and level of impairment should be a key driving factor as researchers develop study designs. The use of virtual platforms and advanced technology may prove too burdensome among patients with more advanced disease, and may result in increased agitation or frustration among patient participants. Patients with more severe cognitive deficits may not be the proper candidates for virtual assessments, and paper-and-pencil administration of performance measures may ultimately be the best choice for the more clinically advanced dementia populations. For data collection purposes, it is recommended that cognitive assessment of patients with severe dementia continue to be conducted in person.
Materials for Visual Processing-During the interview or assessment, patients should be provided materials for visual processing, either through a shared computer screen or a mailed hard copy of documents to be discussed. This allows an additional level of engagement and opportunity to process materials and can troubleshoot potential issues such as poor attention or mild hearing impairments.
Multiple Data Sources-The incorporation of multiple data sources should be considered in the overall study design. For example, including caregivers (e.g., spouses, adult children, professional caregivers, etc.) can provide an additional level of robustness to the data, as well as an additional level of comfort to the patient. Caregivers can be incorporated into the study design to allow assistance with technology and provide proxy-reports on the patient's level of cognitive functioning, either via interview or completion of an ObsRO. Additionally, the patient's referring clinician can provide another level of detail that neither the patient nor caregiver can convey. Specifically, the clinician can provide de-

Conflicts of Interest
The authors declare no conflicts of interest regarding the publication of this paper.