Telehealth for Assessing and Managing Tardive Dyskinesia: Expert Insights from a Cross-Disciplinary Virtual Treatment Panel

Introduction: Publications on the integration of telehealth in the care of patients with movement disorders are increasing, but little has been presented regarding its use in tardive dyskinesia (TD), a drug-induced movement disorder associated with prolonged exposure to dopamine receptor blocking agents. This study was conducted to address that knowledge gap, based on insights from a panel of TD experts. Methods: In 2020, six neurologists, three psychiatrists, and three psychiatric nurse practitioners participated in individual semistructured interviews about in-person and virtual TD assessment and management in their practices. Two virtual roundtables were then conducted to consolidate findings from these interviews. Results: The panel agreed that despite the challenges of virtual TD assessment (e.g., technology issues, difficulty observing entire body, inability to conduct thorough neurological examinations), telehealth can offer benefits (e.g., fewer missed appointments, reduced time/cost, easier access to family/caregiver feedback). The panel also agreed that telehealth should be combined with periodic in-person visits, and they recommended an in-person TD assessment within 6 months before the first virtual visit and at least one in-person assessment every 6 months thereafter. Additional best practices for TD telehealth included implementing video, involving family/caregivers, and providing preappointment instructions to help patients prepare their technology and environment. Conclusions: Telehealth is not a substitute for in-person visits but can be a helpful complement to in-person clinical care. Clinicians can optimize virtual visits in patients at risk of TD by using targeted questions to identify TD and evaluate its impact and by providing education about approved TD treatments.


Introduction
T elehealth, which typically involves real-time or store-and-forward audio, text messaging, or video communication, 1,2 was recommended by the Centers for Disease Control and Prevention (CDC) in February 2020 as part of the federal response to the COVID-19 pandemic. 3 The pandemic-related stay-at-home orders and subsequent infection mitigation measures prompted wideranging changes to health care, including relaxed regulations for providing telehealth and temporary reimbursement parities between in-person and telehealth visits. [3][4][5] These factors led to a steep decline in in-person care and a rapid rise in telehealth use from March to June 2020. [3][4][5][6][7][8] Based on data from a FAIR Health database with 35+ billion health insurance claims, telehealth usage climbed from 0.2% of monthly claims in January 2020 to a peak of 13% in April 2020 before decreasing to 6% by July 2020 and slowly tapering to 4% in November 2021 (https://www.fairhealth.org/states-by-thenumbers/telehealth).
This uptake of telehealth was particularly evident in psychiatry: 81% of respondents to an American Psychiatric Association (APA) member survey indicated that they were continuing to use telehealth for 75-100% of patients as of January 2021. 9 According to a recent report from Trilliant Health based on a nationally representative health care claims dataset, *56 million Americans used telehealth in 2020 and 2021, with behavioral health diagnoses accounting for 58% of telehealth visits in 2021. 10 In light of recent trends and telehealth policy changes under consideration (as of February 2022) to extend or permanently adopt measures enacted during the COVID-19 pandemic, it is evident that telehealth will remain a part of the health care delivery system. [11][12][13] The potential benefits and challenges of telehealth depend on multiple factors, including clinicians' concerns, patients' needs, and reasons for the visit (e.g., initial diagnosis, followup assessment, medication adjustments or other treatmentrelated issues). [14][15][16][17][18][19][20] To that end, a panel of experts was convened to better understand how telehealth may be applied to assessing, diagnosing, and treating patients with tardive dyskinesia (TD), a hyperkinetic movement disorder associated with prolonged exposure to antipsychotics and other dopamine receptor blocking agents, such as metoclopramide. TD can impair a patient's physical, mental, and emotional wellbeing, leading to feelings of embarrassment, anxiety, and social withdrawal/isolation. [21][22][23][24][25] Patients' TD can also negatively impact their caregivers, who are often their family or friends. 26 Despite the availability of approved TD medications (e.g., vesicular monoamine transporter 2 [VMAT2] inhibitors, valbenazine and deutetrabenazine), identification, diagnosis, and assessment of this disorder remains complex, and these complexities can be compounded when care is delivered via telehealth. 22 Further education regarding appropriate screening, management, and treatment of TD is important, especially for telehealth visits where identification and diagnosis may be more difficult or impossible, as in the case of audio-only telehealth interventions.
In their 2020 treatment guidelines for schizophrenia, the APA recommends that all patients taking an antipsychotic should be screened for TD and other drug-induced movement disorders (DIMDs), with follow-up clinical assessments at all visits (e.g., visual observation) and a formal evaluation every 12 months with a structured instrument such as the Abnormal Involuntary Movement Scale (AIMS). 27 The APA guidelines also suggest that patients at high risk of developing TD be screened every six months. In a recent modified Delphi consensus study of the screening, diagnosis, and treatment of TD, consensus was reached that brief screening for TD should be performed at every clinical encounter in all patients taking antipsychotics, and that management of TD requires an overall assessment of psychopharmacologic treatment, including use of antipsychotics, anticholinergics, and VMAT2 inhibitors. 28 Here, we present insights from an expert panel on how to implement these types of recommendations during virtual visits.

Methods
Twelve health care providers (HCPs) were identified based on their research and/or clinical experience with TD, both in person and via telehealth. These HCPs were invited by the study sponsor (Neurocrine Biosciences, Inc., San Diego, CA) to participate in an expert panel on TD; all 12 individuals accepted the invitation. The cross-disciplinary panel included six neurologists, three psychiatrists, and three psychiatric nurse practitioners with diverse backgrounds, practicing in distinct U.S. regions, and from various types of practice settings involving the management of TD.
This qualitative study was conducted using an iterative approach developed by Metaplan Ò (Princeton, NJ) to elicit feedback on key topics regarding the multilayered components of TD management (i.e., screening, diagnosis, assessment, treatment). 29,30 The process began in July 2020 with 60-min online interviews that were conducted by Metaplan with each panelist. The interviews were semistructured, based on a question/discussion guide that was developed by the study sponsor, but representatives of the sponsor did not participate or influence the discussions in any way. Results from the initial interviews were collated using Metaplan's dialogue mapping format and analyzed to determine areas of convergence and divergence regarding the various key topics. These findings, which were distributed to all panelists, were then used to guide a second round of 60-min semistructured interviews conducted by Metaplan. Results from the second round of interviews were analyzed for convergence/divergence and used to generate two concurrent workstreams that culminated in two workshops facilitated by Metaplan in October 2020 that used visualization (e.g., virtual ''sticky notes'') and moderated discussions to identify key areas of consensus. One virtual workshop, which included four neurology specialists over two sessions, focused on the screening and diagnosis of TD. 31 The other workshop, which included five psychiatry specialists, focused on the assessment and management of TD in telehealth settings. Findings from both workshops were presented and discussed during a final virtual workshop in November 2020 that included 11 of the 12 expert panelists. Results from all interviews and virtual workshops were documented and analyzed by Metaplan. Telehealth-related findings are presented in the current report. No quantitative or statistical methods were applied.

INDIVIDUAL INTERVIEWS: TD SCREENING, DIAGNOSIS, AND TREATMENT
During their individual interviews, experts described some of the inherent challenges of assessing TD in both in-person and virtual settings (Fig. 1). Diagnostic challenges included difficulty differentiating TD from other DIMDs and the misuse of ''extrapyramidal symptoms'' (EPS) as an umbrella term for all DIMDs, which can be detrimental to patient outcomes, as TD is distinct from other DIMDs in phenomenology and treatment. TD identification and differential diagnosis can also be complicated by the presence of multiple movement disorders in the same patient (e.g., TD and parkinsonism), as well as the overlap of symptoms from multiple psychiatric and/or medical comorbidities.
Another challenge to arriving at the correct diagnosis is that some patients may not report their abnormal movements because of a lack of understanding of what caused these movements and how they impact their overall psychiatric or medical well-being, decreased awareness due to cognitive deficits from underlying chronic schizophrenia, or even inconsistency of TD symptoms (i.e., waxing and waning in severity and/or intensity).
In addition to the inherent challenges in the screening, diagnosis, and treatment of TD, telehealth adds another level of complexity. For example, the panelists described the technical challenges of virtual TD assessment, such as the segmented views of a patient's body on video (clinicians cannot diagnose or assess movements which they cannot see). Visualization can also be hindered by pixelation or inadequate lighting. Moreover, during virtual visits, it is difficult for HCPs to observe a patient's full-body movements or posture in an unobtrusive way (i.e., without asking the patient to adjust the camera and/or their position in front of the camera). Some patients try to control or hide their movements when they know they are being observed, which can hinder a clinician's overall assessment.
Finally, the panelists noted that many patients do not have smart phones and/or internet access, or they may have difficulty using this technology. In instances where only audioonly telehealth is provided, it is likely that screening for DIMDs is not possible and/or not conducted at all.
During their individual interviews, the panelists suggested strategies to improve TD assessment (both in-person and virtual), including HCP education on the phenomenology and differential diagnosis of TD, as this is a key tool for guiding treatment selection and improving patient outcomes (Fig. 2). In addition, terminology differences across specialties (e.g., misuse of EPS as a catch-all term) need to be addressed to ensure TD is appropriately diagnosed and treated. Panelists also suggested engaging with family members and caregivers to gather additional information about the presence and impact of abnormal involuntary movements. Family/ caregivers can also help patients with treatment decisions and help support adherence to any prescribed medications; thus, clinicians should educate patients and their families about approved TD medications (i.e., valbenazine and deutetrabenazine).
Finally, the experts suggested that when telehealth visits are used for patients who have or are at risk of developing DIMDs, they should be used in combination with periodic in-person visits. Although suboptimal when compared directly to inperson evaluations, video-enabled communication should always be incorporated into virtual visits. Audio-only interactions are completely inadequate for assessing TD.

ROUNDTABLE WORKSHOPS: TD TELEHEALTH
After the individual interviews were completed, the expert panel met for two roundtable workshops to discuss the potential benefits, challenges, and best practices for telehealth in the management of TD. Key findings from these workshops are presented in Figure 3.
The panelists agreed that patients like telehealth, which offers benefits such as less travel time and lower costs. For clinicians, benefits included fewer no-show visits, greater ease of soliciting collateral information from caregivers, and the ability to observe patients in their own environment. However, the panelists agreed that telehealth may limit the ability to perform a thorough neurological examination. Technical issues, such as poor connection, video quality, and/ or camera placement, can be obstacles for effectively diagnosing and managing TD. It can also be difficult to observe involuntary movements during a virtual visit because of segmented/obstructed views of the body or difficulty eliciting hyperkinetic movements. Furthermore, virtual assessment of other DIMDs which can coexist with TD (e.g., observing tone, gait and ancillary signs of parkinsonism) can also be more challenging.

BEST PRACTICES FOR TD TELEHEALTH
The panel agreed that telehealth should not be considered a substitute for face-to-face visits for patients who are at risk of developing DIMDs such as TD, as lack of periodic in-person interaction may compromise the doctor-patient interaction and potentially lead to suboptimal diagnosis and treatment (Fig. 3). However, when used in combination with periodic in-person visits, telehealth can be part of an overall strategy for improving TD diagnosis and treatment. The panel recommended that patients have an in-person TD assessment within 6 months before the first virtual visit and at least one in-person assessment every 6 months thereafter.
Additional best practices included implementing video during all virtual visits and encouraging family members and/or caregivers to participate in the visit to help better understand the overall impact of involuntary movements on both patients and their care partners. The panel also recommended providing patients with clear instructions about the technology set-up before their telehealth visit, as detailed below.
Specific recommendations to help patients prepare for their virtual appointments are presented in Figure 4. Patients should be educated about what will happen during the visit, including how examinations will be conducted and why the setup and environment are important. Providing a checklist can help patients prepare the technology and physical space for the virtual visit. In addition, nurses or other staff members should check in with the patient 24 h before the scheduled visit; during these check-ins, they can help patients test their technology and provide reminders about room and camera setup. If the patient is not able or comfortable conducting a virtual visit from their home, HCPs or other staff can offer an alternative examination location with appropriate technology, such as a telemedicine office or standardized telehealth room.

OVERALL STRATEGIES FOR IMPROVING TD SCREENING, DIAGNOSIS, AND TREATMENT
HCP education on the differential diagnosis of TD may be the best tool for improving patient outcomes, as this would ensure that TD is appropriately identified and treated with the right medications in both in-person and virtual settings (Fig. 5). In addition, alignment within the movement disorder community, particularly with regard to terminology and diagnostic protocols, is needed to help multidisciplinary clinicians (e.g., general

Discussion
Despite a growing body of published literature on the incorporation of telehealth into the clinical care of patients with movement disorders such as Parkinson's disease, there is little published insight on its use in patients with DIMDs, including TD. 14,15 Most studies have evaluated the feasibility of telehealth as a follow-up service to patients with an established diagnosis (usually Parkinson's disease), after an initial in-person evaluation with a detailed medical history and neurological examination. 14,15,[32][33][34][35][36] To date, there are only limited data on the use of telehealth in screening, diagnosing, and managing TD. This layered, qualitative study gathered information from TD experts on their experiences translating TD management (i.e., diagnosis, assessment, and treatment) to virtual care.
The expert panel agreed that telehealth can offer some advantages over traditional in-person visits: fewer no-show visits, better patient access (less time and travel costs), and easier for caregivers/families to participate and provide collateral information. Previous studies of telehealth in patients with movement disorders have shown increased patient satisfaction, reduced cost, and improved access for patients who have difficulty traveling or live in rural areas. [16][17][18]32,37 Despite these advantages, the panelists noted that the inherent challenges in the identification and accurate diagno-sis of movement disorders, including the differentiation of TD from other DIMDs, can be further complicated by the limitations of telehealth, such as segmented/ obstructed views of the patient, inability to conduct a complete neurological examination, poor connection, communication difficulties, and limited access to appropriate technology.
In a recent observational survey study on telehealth during COVID-19, neurology and psychiatry specialists reported similar challenges in telehealth assessment of DIMDs, including difficulty assessing gait/balance, insufficient training for clinicians and staff, patients with limited access to computers and telephones, and greater difficulty obtaining reimbursements. 38 Given these challenges, telehealth should not take the place of in-person visits for individuals who are at risk of TD; rather, it should be used as a helpful complement to clinical care. The panelists recommended that patients at risk for TD should have an in-person TD assessment within 6 months before the first virtual visit and at least one in-person assessment every 6 months thereafter, which aligns with the APA guidelines for TD screening and evaluation. 27 The expert panel agreed on several key points for TD assessments with telehealth, including the importance of incorporating video in all telehealth visits, as audio-only interactions are insufficient to evaluate TD. To encourage patient participation during virtual visits, panelists recommended familiarizing patients with the telehealth process and technology before the appointment.
In a study of patient-physician communication during telehealth and in-person consultations, unfamiliarity with technology and a perceived loss of personal connection may have contributed to lower patient participation during telehealth versus in-person visits. 39 Panelists also recommended engaging family and caregivers, as they can provide collateral information about the presence and impact of TD symptoms. In addition, family members can be a valuable partner in discussions about treatment options (e.g., timing of once-daily dosing [valbenazine] or twice-daily dosing with meals [deutetrabenazine]) and can help to improve treatment adherence.
Finally, the panel recommended continued HCP education on the clinical features of DIMDs. A useful resource to

EXPERT INSIGHTS ON TELEHEALTH FOR ASSESSING TD
facilitate recognition and appropriate treatment of TD is a recent review by Hauser et al, which includes detailed descriptions of the different movements associated with TD and other DIMDs, with accompanying links to videos illustrating TD and non-TD movements. 22 With proper training, TD screening can be done by any trained clinician (i.e., physician, nurses, allied health professionals). 40,41 Screening could include informal evaluations, such as the simple-to-administer rating scales that were used in the real-world RE-KINECT study. 21 These included clinician-and patient-rated severity of abnormal movements (''none,'' ''some,'' or ''a lot'') in four body regions (head/face, trunk/neck, upper extremities, lower extremities) and patient-rated impact (''none,'' ''some,'' or ''a lot'') of these abnormal movements in seven functional areas (continuing usual activites, talking, eating, breathing, being productive, taking care of self, and socializing). Although not tested in telehealth settings, it seems reasonable to expect that these scales would be appropriate for use during virtual visits or as part of a previsit patient questionnaire.
Some validated scales used to assess TD in person may also be adaptable to telehealth. Video rating for the AIMS is already common practice in TD clinical trials. [42][43][44][45][46] A structured AIMS examination can be reliably performed over video telehealth with a full-body view, 47 or a modified examination (6/7 regions) can be performed with an upper-body view. There is also room for innovation with virtual-friendly instruments, such as the recent MIND-TD questionnaire. MIND-TD guides clinicians through a short screen for abnormal movements, an interview about physical/functional difficulties, and visual observation with telehealth-appropriate activation maneuvers for a modified AIMS assessment.

Conclusions
The future of psychiatric and neurological care may be moving toward hybrid models of care that balance in-office visits with virtual visits. To ensure that patients with TD receive appropriate care, future policy for telehealth should weigh the benefits and risks of audio-only telehealth, improve access to appropriate video technology during virtual visits (an estimated 45-50% of telehealth visits during the pandemic were audio-only), 48,49 and safeguard continued access to and proper incentives for in-person visits. In addition, future studies should evaluate outcomes resulting from hybrid care delivery models. Finally, the development of new technologies, such as wearable devices or sensors, may enhance future virtual visits.