ATA2020 Session Summaries (Abstracts)

1. What You Should Know About Interacting with a Telehealth Resource Center

Elizabeth Krupinski¹ and Doris Barta²

¹Emory University and ²National Telehealth Technology Assessment Resource Center

The Health Resources & Services Administration Office for the Advancement of Telehealth National Consortium of Telehealth Resource Centers are comprised of 12 regional and 2 national Telehealth Resource Centers (TRCs) providing technical assistance, education, and various resources, with each having individual uniqueness, allowing them to provide a wide range of assistance targeting your regional needs. We describe our overall technical assistance model, which provides tiered levels of support ranging from free advice and information on dedicated topics of inquiry to in‐depth consultation on program design and implementation. We review data on the types of technical assistance providers to our telehealth customers, including but not limited to: number of assistance requests, types of provider and other organizations, assistance topic areas, commonly encountered questions.

Method: Descriptive

Classification of Research: Technical Assistance

Result: Since inception of the TRCs, our websites have received over 1 million hits, our webinars have been attended by of 7,500 people, regional conferences have attracted over 8,000 attendees, and we have engaged in over 10,000 direct technical assistance events with those seeking to develop or expand their telehealth programs.

Conclusions: The TRCs have established the National Consortium of Telehealth Resource Centers to strengthen their collaborative efforts. We have produced fact sheets, guides, templates, and other telehealth resources while improving brand recognition and strengthening relationships. The knowledge TRCs possess covers the full range of telehealth domains, reflective of their impact on the field, developing approaches to telehealth barriers from a variety of angles field.

2. Long‐Distance Connections: Metrics, Outcomes and Overcoming Challenges in Telepsychiatry

Hossam Mahmoud, Michel Tawil, and Mohammad Haidous

Regroup, TUFTS University School of Medicine

As the adoption of telepsychiatry continues to expand to meet the growing mental health needs across the country, there has been increased interest in examining the quality of telepsychiatry programs and assessing the impact of such programs. Research has attempted to examine not only clinical outcomes, but also population health measures, cost‐efficiency, access improvement and user satisfaction. This session begins by reviewing the approaches and methods that have been undertaken to evaluate the impact and outcomes of telepsychiatry programs. Next, we discuss a case study of a telepsychiatry program in rural Illinois that combines direct care with consultation services.

Method: Descriptive

Method – Other:

Classification of Research: Clinical Outcomes

Result: In this case study, we outline the program components, including synchronous consults, asynchronous consults, and direct patient care. We discuss the planning, implementation and outcomes of this program and compare outcome measures to those described in the literature. Specific metrics that will be reviewed include cost‐effectiveness, volume of patients served, wait list, patient satisfaction, clinician satisfaction, screening tools and clinical outcomes.

Conclusions: We then examine the advantages and limitations of current approaches to telepsychiatry program evaluations, including the significant variation in metrics and indicators that are used to assess outcomes. Finally, we explore approaches to overcoming some of these challenges in program evaluations, and we review recommendations and best practices for telepsychiatry program evaluation.

3. Tele‐transitions of Care: Evaluating the use of Telehealth for Triple Aim Objectives

Kimberly Noel, Catherine Messina, Wei Hou, Elinor Schoenfeld, and Gerald Kelly

Stony Brook Medicine

Telehealth has the potential to improve transitions of care, through enhanced connections among patients and their clinicians, during a vulnerable period after hospital discharge. To achieve triple aim objectives, reducing unnecessary hospital readmissions is desirable for payers and patients alike. However, poor transitions of care extends beyond the risk of increased hospital readmission rates. Poor transitions also lead to increased medical errors, poor outcomes and inappropriate resource allocation. This is a 12‐month randomized controlled trial, evaluating the use of telehealth (remote patient monitoring and video visits) versus standard transitions of care with the outcomes of over‐utilization, access to care, medication management, patient adherence and patient engagement.

Method: Randomized Controlled Trial

Classification of Research: Clinical Outcomes

Result: Compared with the standard of care, telehealth patients were more likely to have medicine reconciliation (p = 0.013) and were 7 times more likely to adhere to medication (p = 0.03). Telehealth patients exhibited enthusiasm (p = 0.0001), and confidence that telehealth could improve their healthcare (p = 0.0001). Telehealth showed no statistical significance on emergency department (ED) utilization (p = 0.691) nor for readmissions (p = 0.31). 100% of telehealth patients found the intervention to be valuable, 98% if given the opportunity, would continue using telehealth to manage their healthcare needs, and 94% reported that the remote patient monitoring technology was useful.

Conclusions: Telehealth can improve transitions of care after hospital discharge improving patient engagement and adherence to medications. This study proves the value of telehealth and it's feasibility. Although this study was unable to show the effect of Telehealth on reduced healthcare utilization, more research needs to be done in order to understand the true impact of telehealth on preventing avoidable hospital readmission and ED visits. Our study showed that patients who were readmitted had life‐threatening emergencies. We offer consideration as to whether hospital readmissions should be the desired endpoint in telehealth research.

4. A Clinical trial of asynchronous telepsychiatry in primary care: clinical outcomes and implications

Peter Yellowlees and Alvaro Gonzalez

UC Davis

Objective Asynchronous Telepsychiatry consultations have been suggested as a potential choice in a stepped care menu of consultation approaches to be made available to patients in primary care settings. In this randomized clinical trial Asynchronous Telepsychiatry (ATP) consultations were compared with Synchronous Telepsychiatry (STP) consultations, in a heterogeneous sample of treatment seeking patients referred by their primary care physicians (PCP).

Method: Randomized Controlled Trial

Classification of Research: Clinical Outcomes

Result: 184 participants, English and Spanish speaking, were enrolled and randomized, of whom 160 (80 ATP, 80 STP) completed baseline evaluations. Patients were treated by their PCP's in consultation with UCD psychiatrists using ATP or STP for up to two years. The clinical outcomes on psychiatrist and patient reported scales in English and Spanish speaking patients for ATP were similar to STP. Both patient groups improved over the course of the trial. Significantly greater levels of patient drop out than anticipated were experienced in both groups for numerous reasons, although less in the ATP group.

Conclusions: This trial is evidence of the clinical effectiveness of ATP in primary care patients. It should lead to the introduction of this form of consultation as part of a stepped series of mental health interventions available within the primary care treatment setting, and because of its efficiency may be part of the solution to the workforce shortage of psychiatrists. Funders should support payment for ATP consultations, and research using asynchronous video in disciplines other than psychiatry should be prioritized. ClinicalTrials.gov Identifier: NCT02084979Funded by AHRQ.

5. Leveraging Telehealth to Deliver Advance Care Planning Services to People Approaching End of Life

Connie Ducaine

Vital Decisions

Patients who are dealing with complex medical situations and approaching end of life benefit from developing an advance care plan. This type of plan ensures caregivers and providers are aware of patients' preferences so these stakeholders can facilitate the appropriate level of care in the event the individuals can no longer speak for themselves. Traditionally, Advance Care Planning (ACP) activities have been initiated in a physician's office with varied results. Physicians may fail to discuss end of life care with patients/families until it's too late to substantially impact the treatment experience or allow for optimal decision‐making about future care (during inpatient episodes for medical crises). When physicians attempt ACP conversations, effectiveness may be limited by time pressure, lack of training/skills, or a focus on specific procedures rather than on patients' values and quality of life goals. Leveraging virtual modalities for ACP has resulted in the delivery of services to more than 300,000 people in the United States with a positive impact on individuals' end of life experiences. Virtual models eliminate or address many barriers (e.g., timeliness of discussion, transportation, time to travel, childcare) that often prevent seeking these services. Engagement rates for the virtual ACP intervention were 60 – 70%.

Method: Survey/Qualitative

Classification of Research: Patient Experience

Result: To evaluate the patients' experiences of the ACP intervention, which is a series of telehealth sessions in which individuals explore, document and communicate their care preferences, a six question survey was administered. The survey results were tabulated by a third party. Patients and their caregivers have expressed their appreciation for, and positive experience with the virtual intervention. Survey respondents (N ∼ 2,000) have reported an overall satisfaction of 4.6 on a 5‐point Likert scale. This is equivalent to a Net Promoter Score of 59.

Conclusions: The efficacy of the telehealth model with advanced illness patients demonstrated by the engagement data and patient satisfaction results negates the perspective that delicate conversations are best addressed in‐person. Program results also suggest that patient activation and engagement can be developed via telehealth model utilizing a brief intervention model.

6. A Virtual Intervention to Reduce Behavioral Health Admissions from Rural EDs – Program Design

Jason Roberge, Christine Zazzaro, Amy Barrett, Pooja Palmer, and Wayne Sparks

Atrium Health

Hospital admissions are common among patients with mental illness resulting in a burden within the healthcare system. Significant morbidity exists in patients that are admitted to a psychiatric hospital from the emergency department (ED). Due to limited availability of behavioral health resources, ED providers often decide to admit patients to a psychiatric hospital. To better enhance the transition of care for patients with mental illness, Atrium Health has designed a behavioral health virtual patient navigation (BH‐VPN) program that helps coordinate services and follow‐up care, while facilitating the safe discharge of patients. Patients that present to an ED that have a telepsychiatric consult and are recommended for discharge are eligible for the service. The Behavioral Health Service line is expanding the program from urban EDs and assessing the effectiveness of the program among rural EDs.

Method: Randomized Controlled Trial

Classification of Research: Access to Care

Result: Using a randomized clinical trial, we will assess the effectiveness of the BH‐VPN program to reduce hospital admissions among patients presenting with a behavioral crisis at a rural ED. Patients who complete a telepsychiatric consult in the ED are enrolled to either the usual care or intervention arm based on a randomization scheme. Prior to discharge, a navigator will connect with the patient virtually in the ED and offer participation in the program.

Conclusions: Available evidence suggests that adoption and utilization of virtual care in tandem with wrap‐around services may reduce utilization and improve health outcomes. Here we will present the details of the program and share experiences from prior behavioral health virtual programs.

7. Precision Psychotherapy with Artificial Intelligence

Thomas Hull, Jeffrey Swigert, and Neil Leibowitz

Columbia University, Talkspace

Digital platforms for delivering psychotherapy and other types of medicine are growing. These platforms generate a large amount of data offering the promise of more sophisticated classification and predictive models. These models allow practitioners to go beyond underspecified categories and to match treatment to patient in a more precise way. In this talk we present the results of applying text‐based machine learning to specify patient presentation that is much more specific than diagnosis alone. We also present results on factors that identify therapists who deliver better care on average. This final model combines value‐added modeling from economics with natural language processing from clinical interactions and transcripts.

Method: Implementation Science

Classification of Research: Measurement Frameworks & Tools

Result: The presented machine learning model is able to accurately categorize 85% of patients based only on their age and gender combined with as little as two paragraphs of text from a natural therapy interaction. The value‐added model identified 6 key factors that separate highly effective therapists from the rest. Of these 6, 4 are modifiable through training or through better matching of patients with providers best suited to address their needs.

Conclusions: We conclude that the provision of behavioral health, a key factor in addressing modern disease burden, can be vastly improved through the use of big data and machine learning methods. The large data sets made available by telemedicine platforms offers a significant potential for improving not just access of care, but quality of care and value delivered to patients.

8. Design and Implementation of a Primary Care Telemedicine Elective for Medical Students and APRNs

Mark Rood, Kari Gali, Leighanne Hustak, and Matthew Faiman

Cleveland Clinic

Despite massive expansion of telemedicine, most programs listed as telemedicine sites by the AAMC are educating students via long distance learning, not training students to provide quality virtual medical care, including good webside manner, components of a video exam, and cutting‐edge innovations occurring in the rapidly changing field.

In 2016 Cleveland Clinic partnered with three medical schools in Northeast Ohio creating a 2‐week elective. In 2017 the elective was opened to all of the 13 Ohio nurse practitioner schools, in 2018 aligned with ACGME competency driven education, and in 2019 began a collaboration with a college to assist in the development of a Certification in Telemedicine education track. This elective provides experiential learning in the clinical application of Primary Care Telemedicine, as well as specialty opportunities including MyChart, eHospital, Telepsychiatry, remote monitoring and others. Guided by a Virtualist mentor, they complete a scholarly project on a cutting‐edge aspect of telemedicine in their field of choice. Students are required to have completed their core clerkships in Family and Internal Medicine, Pediatrics, Psychiatry and General Surgery before applying for this rotation for the medical students. The nurse practitioner student must be in their last semester, typically completing their capstone.

Method: Mixed method (observational, descriptive and survey/qualitative)

Classification of Research: Other

Classification of Research: Education for future clinician experience

Result: Students are assessed in areas such as Knowledge for Practice, Interpersonal and Communication Skills, and Systems Based Practice. As we have grown we have formalized our structure with enrollment of students, involved more NP's in the instructional curriculum, enhanced the use of complimentary instructional videos, and continued our relationships with eHospital, Telestroke, Telepsychiatry, eVisit and MyChart teams. Feedback from stakeholders helped shape the program into a robust, interactive, competency based elective.

Conclusions: With a shift from fee‐for‐service to value‐based care, healthcare providers are being challenged to consider delivering care through a virtual platform without any formal training. Just because providers can do this, it doesn't mean they will do it well. Integrating synchronous and asynchronous telemedicine opportunities allows students to merge the didactic learning into clinical practice, enhancing care for patients in the future. It is essential for health care systems to innovate ways to educate providers on delivering a level of virtual care that maintains consistency with the same high‐quality standards to which they are held in the non‐virtual world.

9. Changing Hospital Culture through the Implementation of a TeleNeurology Program

Tejal Raichura, Anthony Noto, and David Fletcher

Geisinger

Geisinger is a large integrated health system in central Pennsylvania and New Jersey comprised of 13 hospitals. Geisinger Medical Center (GMC) in Danville, PA is Central Pennsylvania's only quaternary‐referral, academic institution with all major medical, surgical, transplant, obstetric, neonatal, pediatric, neurosurgical, and trauma‐related specialties represented and serves as the central hub of Geisinger's telehealth network.

Utilizing a hub‐and‐spoke model, the Neurosciences Institute leveraged an on‐site neurology hospitalist model to offer non‐stroke, tele‐neurology consults to 5 spoke sites (Pennsylvania only) within the Geisinger system; this is in addition to an already robust telestroke program. The intent of this program was to mitigate the need to transfer patients to the main hub, reduce the cost of the transfer, and review overall patient outcomes.

Method: Observational

Classification of Research: Access to Care

Classification of Research – Other:

Result: The total number of teleneurology consults completed coincided with a reduction in transfer rates from spoke hospitals to the hub; additionally, patients were able to be seen by a neurologist faster than in the previous period as the consult was often completed at the time of admission. Finally, neurology admissions started trending downward post‐intervention, however, the admissions proceeded to increase again due to a change in culture. Furthermore, the contribution margin per inpatient transfer was calculated at $10,000. Taking our average number of transfers avoided per month, we saw a reduction in costs of approximately $250,000 per month.

Conclusions: The implementation of the teleneurology program has started to shift the culture of the local emergency department teams who feel more comfortable admitting the patients locally knowing a telemedicine consult is available as needed; it has also allowed the hub neurology ICU to keep their admissions instead of transferring to a Medicine ICU. As a result, the savings from avoiding transfers to a higher acuity hospital in addition to admitting patients to facilities with the most appropriate level of care for their condition has led to cost savings across the organization.

10. Emergency provider tele‐medicine hours associated with decreased reported burnout symptoms

Anisa Heravian, Erica Olsen, David Kessler, and Bernard Chang

Columbia University Irving Medical Center

Whereas 45% of the 1 million physicians in the United States report symptoms of burnout (i.e., emotional exhaustion, depersonalization, and reduced personal accomplishment), an astonishing 70% of emergency department (ED) providers report burnout symptoms. ED overcrowding and related factors has been found to increase psychological stress in not just patients but also emergency providers, potentially increasing one's risk for the development of adverse professional and psychological outcomes such as clinician burnout. Telemedicine, may offer a unique complement to this practice environment, allowing providers to administer care in a more controlled environment without many of the other existing acute environmental stressors. We hypothesized that providers working more telemedicine hours would be associated with lower rates of clinician burnout compared to providers not working telemedicine.

Method: Observational

Classification of Research: Clinician Experience

Result: Fifty (n = 50) emergency providers participated. Eighteen (n = 18) individuals did telemedicine shifts and 32 individuals did not do any telemedicine. Overall, no differences in sex, age or years of practice were found between providers performing tele‐medicine and those. Tele‐medicine providers performed on average 9.7 ± 4.5 hours of telehealth weekly. Individuals in the telehealth group had significantly lower scores on burnout measures (emotional exhaustion subscale) compared to the non‐telehealth group (8.4 ± 2.2 vs 11.4 ± 3.5; t = 2.64, p < 0.05). A multiple regression model, adjusted for age, years of practice and sex, found that hours of telemedicine per week significantly predicted emotional exhaustion (beta: −0.35, t = ‐2.22, p < 0.04)

Conclusions: In addition to improvements in patient outcomes, telemedicine may also improve provider psychological well‐being. Future work exploring the integration of telemedicine shifts into clinical scheduling may be associated with improvements in provider well‐being and career longevity.

11. A telehealth approach to reduce emergency utilization by combining AI with tailored interventions

Sara Bersche Golas, Jorn op den Buijs, Mariana Simons, and Gary M. Garberg

Partners Connected Health Innovation

By 2030, 73 million adults will be over the age of 65 in the United States (U.S.). Chronic disease is prevalent in this population: 80% have at least one chronic disease; 77% have at least two. Chronic disease management accounts for 75% of annual U.S. healthcare system spending. Moving toward value‐based care, organizations are identifying ways to lower healthcare costs by reducing emergency and hospital utilization. Many independently‐living older patients use a Personal Emergency Response System (PERS) to signal for help in case of incidents, e.g. falls, breathing problems. Using remotely‐collected PERS data, we developed artificial intelligence approaches to identify individual risk of emergency transport, allowing early intervention and care in lower‐cost settings. We describe a 180‐day randomized controlled trial combining risk predictions with tailored interventions, demonstrating reductions in emergency and hospital utilization.

Control and intervention groups used a PERS service comprised of a wearable device with automated fall detection and 24/7 response center access. In the intervention group, PERS service data were collected and processed via predictive models to indicate imminent emergency transport risk. A study nurse triaged individuals with high‐risk scores using needs assessment questionnaires and tailored intervention recommendations (e.g. personalized remote education, primary care referral, condition‐specific telehealth).

Method: Randomized Controlled Trial

Classification of Research: Clinical Outcomes

Result: A total of 333 patients were analyzed ‐ 173 in the control and 160 in the intervention groups. While there was no statistically significant difference between the group's demographics and clinical characteristics, the following clinical outcomes were statistically different: compared with the control group, the intervention group had 61% fewer 90‐day readmissions (p = 0.015) with corresponding triple decrease of proportion of patients with any 90‐day readmission (10.4% control vs. 3.1% intervention group, p = 0.009); 46% fewer 180‐day readmissions (p = 0.038); and 49% fewer 180‐day Emergency Medical Services (EMS) encounters (p = 0.006).

Conclusions: This randomized control trial provides clinical evidence that combining actionable predictive analytics with personalized interventions can greatly reduce emergency incidents experienced by older patients in the home, thereby reducing hospital transports and readmissions. Such solutions combining predictive analytics, remote patient engagement, and telehealth facilitate the delivery of value‐based care, improve patient health outcomes, and decrease healthcare costs.

12. Avoidable Emergency Department Outcomes in a Health System‐Partnered School Telemedicine Clinic

Carlene A. Mayfield, Tiffany Effinger, Jennifer Villafane, Sam McGinnis, Patsy A. Fisher, Brisa Hernandez, Alisahah J. Cole, and Patty Grinton

Atrium Health

Families living in Cleveland County, North Carolina, experience a cluster of social and economic determinants of health including high rates of poverty, unemployment, and lack of insurance coverage. Additional healthcare access barriers including transportation, system navigation, and parental work schedules, result in inappropriate utilization of the emergency department (ED) for nonemergent or primary care. Atrium Health, one of the largest integrated health systems in the region, and its facilities‐ Kings Mountain Hospital, Cleveland, Shelby Children's Clinic, and Department of Community Health‐ partnered with the County's Public Health Center and the local school system to develop School Based Telemedicine Clinics (SBTCs). Some program outcomes (i.e. reducing early school dismissal) can be tracked using school records and/or during the SBTC visit. Other outcomes, including reduction of ED utilization, requires metric specificity (i.e. isolation of nonemergent and avoidable ED visits) and the enrichment of primary program data with electronic medical record (EMR) data. Our project developed and tested a protocol to track avoidable ED utilization among SBTC patients using a scalable, semi‐automated metric available through EMRs. Avoidable ED utilization was measured using the New York University Algorithm, a validated classification system that predicts the probability of a visit being avoidable using discharge codes.

Method: Secondary Data Analysis

Classification of Research: Clinical Outcomes

Result: The ED records of 80 patients with an SBTC visit on October 10, 2017 through August 31, 2018 were extracted from Atrium Health EMRs, resulting in an analytic sample period of 1 year pre‐ and post‐visit. Visits with a probability ≥50% were classified as avoidable, according to standard practice. Preliminary descriptive results show an overall 13% reduction in avoidable ED visits btween the pre‐visit (88 ED visits; 47 or 53% avoidable) and post‐visit (63 ED visits; 25 or 40% avoidable) periods.

Conclusions: Measuring the outcome of avoidable ED utilization using an automated algorithm is an efficient and sensitive metric to track clinical outcomes tied to SBTC project goals. Initial tests show this method is valid in our sample. Leveraging health system partnerships to access existing EMR data resources can inform quality improvement and better tracking of program outcomes. Attendees of our session will learn our protocol for accessing, testing, and applying the New York University Algorithm to examinations.

13. Improving Patient Outcomes & Cost Savings by Leveraging Remote Monitoring Technology to Accelerate & Scale Care management for Medicare Advantage Members

Carla Moore Beckerle, Erin Stamm, and Robert Mattson Peters

Esse Health

Shifts toward improving outcomes in value‐based care systems have prioritized managing high and rising‐risk patients with chronic disease proactively and efficiently. Remote monitoring with digital and telehealth tools has been shown to lead to proactive engagement. Our organization implemented a text message‐based remote monitoring program for eight months with a single care manager and scaled to over a thousand active Medicare Advantage members at a time. Real‐time automated monitoring allowed us to restructure our CM program by focusing resources to the right members at the right time as staff operated at top‐of‐license and patients were empowered to actively manage their health. Daily targeted check‐ins provided timely patient health data, the automated feedback loop notified the care manager with opportunities to escalate care and proactively engage the member to coordinate care prior to routine outpatient appointments or eventual ED visits. As operational feasibility and scalability was immediately evident, the next step was to evaluate if the program led to reductions in healthcare utilization and member costs. To mitigate confounding factors, we evaluated outcomes using Intention to Treat (ITT) per‐member‐day methodology to compare differences in all‐cause ED utilization, inpatient hospitalization, and insurance claims pre and post program enrollment.

Method: Cost Analysis

Classification of Research: Cost Analysis

Result: We analyzed claims data for 1,527 patients. Wilcoxon matched‐pairs Signed Rank tests showed a 38% reduction in paid claims per‐member‐per‐day associated with emergency care (Mean ‐$5.99, SEM $3.10, 95%CI ‐$12.01 to ‐$0.09, p = 0.053); 51% reduction associated with hospitalizations (Mean ‐$7.56 SEM $3.06, 95%CI ‐$13.56 to ‐$1.56, p = 0.014). All‐purpose claims analysis showed a 19% reduction (Mean ‐$11.20, SEM $2.96, 95%CI ‐$17.01 to ‐$5.39, p < 0.001). Mean differences in HbA1c ‐0.51% (SEM 0.13, 95%CI −0.76 to −0.27, p < 0.001, n = 107), sBP of ‐14.75 mmHg (SEM 2.38, 95%CI −19.50 to −10.00, p < 0.001, n = 92); dBP of ‐7.55 mmHg (SEM 1.71, 95%CI −10.96 to −4.15, p < 0.001).

Conclusions: Implementing a text message‐based remote monitoring tool allowed our care management program to scale by utilizing one full‐time care manager to engage and monitor over 1,500 members with chronic conditions with over 170,000 automated “touches” that generated 4,100 opportunities for proactive outreach. Proactive patient outreach resulted in significant reductions in ED use, inpatient admissions, and healthcare costs/expenditures. The results of claims data analysis affirmed the associated reductions in non‐outpatient healthcare utilization and biometric measures. Current and future endeavors include expanding beyond the Medicare Advantage cohort, increasing diabetes & hypertension management cohort, and beginning medication adherence and behavioral health programs.

14. Medicaid ACO: Digital Innovations and Start‐ups drive biggest ROI and Clinical Impact

Heather Meyers

Boston Children's Hospital

Boston Children's Hospital (BCH) formed an Accountable Care Organization (ACO) for 150,000 Medicaid lives in March 2018. Six months prior to launch, BCH started a digital strategy initiative with a goal of improving outcomes while lowering cost over a 3‐year span. Year one spend request was $1M in state DSRIP funding, $600,000 year two and $300,000 in year three to result in sustainability in year four.

Six months before launch, industry interviews were conducted, a Steering Committee and Advisory Working Team was formed, and a 3‐year plan created factoring in staffing needs, legislative environment and enterprise strategy. Year one, 2018, focused on specialty and primary care virtual visits and non‐emergency medical transportation to address social determinants of health. These were “quick wins” due to current population implementation and partnerships. Year two, 2019, focused on behavioral health virtual visits in BCH Affiliated Primary Care Practices. These practices account for 65% of the ACO population with an office variance of 2% to 70%. Digital automation for asthma was also included to target the second highest risk cohort patients. Year three, 2020, focuses on urgent care visits for select specialties and primary care offices as well as remote patient monitoring.

Method: Cost Analysis

Classification of Research: Access to Care

Result: Year one, 30% under budget with following outcomes:

Virtual Visits

Non‐Emergency Medical Transportation

Conclusions: The 2018 initiative results show patients have increased compliance with appropriate care when digital initiatives are incorporated, are less likely to use higher cost of care when unnecessary and 100% of patients personal smart devices giving them access to these tools. However, more staff engagement and patient education was needed in the higher populated vs. lower Medicaid offices and issues such as app storage requirements and use of data could limit the number of total patients engaged.

15. Office of Virtual Health: Moving virtual health forward in British Columbia

Kathy Steegstra

Provincial Health Services Authority, Office of Virtual Health

The Office of Virtual Health (OVH) is a Provincial Health Services Authority (PHSA) initiative mandated by the BC Ministry of Health to enhance virtual health as part of the care continuum. At four times the size of the United Kingdom, British Columbia's geography presents challenges for many patients to receive equitable access to care. Virtual health supports people to receive care and stay well by using digital innovation to connect them to care seamlessly, when and where they need it.

In 2017 the OVH was created to facilitate work with all clinical programs across PHSA to support their virtual health strategies. Four clinical priorities were identified:

The OVH initiated demonstration projects in the priority areas to integrate virtual health solutions at the point of care, following the initiative life cycle (engagement, discovery, planning, execution, evaluation and scale).

Method: Demonstration projects

Classification of Research: Access to Care

Result: The OVH collaborates with clinical, operational and corporate partners to facilitate transformation, and currently has 13 active demonstration projects across 7 clinical programs and 3 regional health authorities:

Conclusions: Early evaluation and feedback indicates patients and clinicians strongly value the time, cost and effort saved through virtual health care delivery. OVH integration support includes process redesign, change management, project management, evaluation and reporting. The next step is to incorporate learnings from the demonstration initiatives to scale virtual health solutions across PHSA. The OVH team is collaborating with key stakeholders across the health system to ensure we consider all components.

16. Implementation of Telemedicine Medical School OSCE

Walkitria Smith and Folashade Omole

Morehouse School of Medicine

The future of medicine is based on the training of future providers. In addition to basic science, honing of interpretation skills and development of clinical competencies is imperative for future health professionals and this must include telemedicine as part of the medical experience. Morehouse School Medicine is investing in the advancement of health equity through the application of innovative patient care delivery through introducing third year medical students to telemedicine. As a part of their formal evaluation for the Department of Family Medicine and Fundamentals of Medicine, third year medical students are graded by clinicians in a simulation lab designed to standardize the content and scoring of specific medical procedures. The purpose of the OSCE (Objective Structured Clinical Examination) is to provide a controlled environment for patient interaction, deliver feedback on clinical skills and enhance the medical student's development of interpersonal communication skills. In January 2019, the first cohort of students were graded on formal “OSCE Telemedicine Cases.” The cases included an obstetric patient seeking care for vaginal bleeding and a patient receiving follow‐up results to a procedure requiring additional education and history solicitation.

The goals of these cases are to teach new modalities designed to improve access to care.

Method: Survey/Qualitative

Classification of Research: Clinician Experience

Result: Students are randomly assigned to one of two OSCE cases. Participating students received feedback on their collection of essential historic data which includes social, sexual, medication and past medical history. This information was provided by colleagues and attending physicians. At the end of the OSCE experience they are able to rate the effectiveness of their telemedicine case. Students completed a telemedicine‐focused survey that probes their knowledge of telemedicine, utilization and subjective opinion on the experience.

Conclusions: Morehouse School of Medicine medical students value the inclusion of telemedicine when training in a controlled environment focused on improving communication and medical procedures. Medical students taking OSCE value the benefits of the introduction of telemedicine and they believe it will improve care.

17. Advancing Adoption of Teleophthalmology in Primary Care through Stakeholder‐Engaged Implementation

Yao Liu, Alejandra Torres Diaz, Julia Carlson, Nicholas Zupan, Todd Molfenter, Jane Mahoney, Mari Palta, Deanne Boss, Timothy Bjelland, and Maureen Smith

University of Wisconsin‐Madison, Department of Ophthalmology and Visual Sciences

Teleophthalmology is an evidence‐based method of diabetic eye screening recommended by the American Diabetes Association. Vision loss from diabetes is highly preventable with early detection, but remains the leading cause of blindness among working age United States (U.S.) adults due to low screening rates. Teleophthalmology has been greatly underused in part because it is challenging to successfully implement in U.S. primary care clinics. We tested the hypothesis that a systematic healthcare process improvement framework, NIATx (www.niatx.org), could be adapted to increase teleophthalmology use and diabetic eye screening rates in a rural primary care clinic.

Method: Implementation Science

Classification of Research: Clinical Effectiveness

Result: Mile Bluff Medical Center (MBMC) is a rural U.S. health system that established a teleophthalmology program in 2015, but the program was very underutilized. We adapted NIATx to guide stakeholder meetings and used PDSA (plan‐do‐study‐act) cycles to test strategies for increasing teleophthalmology use tailored to an individual clinic's needs and resources. Nine patients and 23 clinical stakeholders participated in meetings from May 2017‐October 2018. Teleophthalmology use increased 5‐fold at the implementation clinic compared to 0.4‐fold at the other clinics (p = 0.03). Overall screening rates at Mile Bluff increased from a baseline of 47.4% in 2015 to 64.0% in 2018.

Conclusions: We created an implementation toolkit (I‐SITE) to facilitate adoption of teleophthalmology in U.S. primary care clinics. Engaging stakeholders to tailor implementation sustained increased telemedicine use. Our freely‐available online toolkit (https://hipxchange.org/I-SITE) can help facilitate widespread adoption of teleophthalmology and may be applied to other primary care‐based telemedicine programs.

18. Validation of computer‐aided diagnosis of diabetic retinopathy from retinal photographs of diabetic patients from Tele‐camps

Sheila John, Sangeetha Srinivasanb, Keerthi Ramc, and Mohanasankar Sivaprakasamc

Department of Teleophthalmology, Sankara Nethralaya, Chennai, India

Diabetic retinopathy (DR) is a microvascular complication of diabetes and causes blindness. An algorithm developed by Healthcare Technology Innovation Centre, IIT Madras aided in the screening of retinal images of diabetic patients to detect the presence or absence of diabetic retinopathy. The technology was validated in fundus images of diabetic patients from teleophthalmology camps to find the screening performance for diagnosis of diabetic retinopathy.

Methods, 939 eyes of 472 diabetic patients underwent nonmydriatic fundus photography (40–45 degree posterior pole/each eye) from Mobile Teleophthalmology camps in Thiruvallur and Kanchipuram districts, Tamil Nadu, India, over the two‐year study period from Jan 2015 to May 2017 Fundus images were obtained for all patients using a nonmydriatic fundus camera (model Topcon Retinal Fundus Camera TRC‐NW8F with Accessories) by the fundus photographer. The fundus images were evaluated for the presence or absence of diabetic retinopathy using the computer‐assisted algorithm, by the Ophthalmologist at Sankara Nethralaya (reference standard) and by the fundus photographer.

Method: Implementation Science

Classification of Research: Clinical Effectiveness

Result: Compared to the ophthalmologist, 49% images were ungradable by the algorithm. There was an only slight agreement in terms of image gradeability between the ophthalmologist and algorithm, Kappa = 0.019 (95%CI = ‐0.008 ‐ 0.046). The sensitivity of the algorithm was found to be 85% and specificity was found to be 80% in detecting DR compared to the ophthalmologist.

Conclusions: The algorithm developed was able to detect the presence or absence of diabetic retinopathy in diabetic patients and will help to reduce the workload for human grader in remote areas. Keywords. diabetes, diabetic retinopathy, algorithm, screening, retinal photographs.

19. Validation of a wearable electromyography sensor for the remote management of swallowing disorders

Cagla Kantarcigil, Minku Kim, Bruce A. Craig, Chi Hwan Lee, and Georgia A. Malandraki

Northwestern University, Department of Communication Sciences and Disorders and Purdue University, Department of Communication Sciences and Disorders

Swallowing is a complex biomechanical process. One critical component of this complex process is hyolaryngeal excursion. This action involves the anterior and superior movement of the hyoid bone and the larynx during swallowing and occurs mainly by contracting the suprahyoid muscles, consisting of the mylohyoid, geniohyoid, and anterior belly of digastric. Surface electromyography (sEMG) of the suprahyoid muscles is typically used in clinical practice to provide real‐time biofeedback to patients during swallowing treatment. However, most sEMG devices are bulky, expensive, and only available in large medical centers. Due to lack of existing user‐friendly and cost‐effective sEMG devices for swallowing, patients typically receive no biofeedback when they complete their exercises at home. In addition, clinicians who work in rural areas typically do not have access to these types of devices, thereby decreasing the quality of care provided to patients who live in rural and underserved areas. To start addressing this need, we developed an inexpensive, portable, and ultra‐thin wearable sEMG sensor which was specifically designed to conform to the anatomy of the suprahyoid region. The purpose of this study is to compare the utility of this wearable sEMG sensor in monitoring suprahyoid muscle activity during swallowing with a conventional sEMG sensor.

Method: Randomized Controlled Trial

Classification of Research: Clinical Effectiveness

Result: Forty healthy adults participated (24F, mean age = 67.5). Clinical factors: No pain or adverse effects were reported. Satisfaction was significantly higher with the wearable sensor (p = 0.0476). Sensor‐related factors: Mean signal‐to‐noise ratio of the wearable sensor was not inferior to the mean signal‐to‐noise ratio of the conventional sensor (p < 0.0056). Similarly, baseline amplitude values obtained with the wearable sensor were not inferior to the baseline amplitude values obtained with the conventional sensor (p < 0.0001). The normalized amplitude values were deemed equivalent for all swallows (5mL: p < 0.05; 10mL: p < 0.0012). The duration of sEMG burst during swallow trials was also deemed equivalent (5mL: p < 0.0001; 10mL: p < 0.0001).

Conclusions: Our findings indicate that the newly developed ultra‐thin wearable sEMG sensor conforms to the curvilinear surface of the suprahyoid area seamlessly and its technical performance is similar to the performance of widely‐used conventional sEMG sensors. Results of the safety and clinical factor comparisons further support that the wearable sensor is safe to use and healthy older adults are satisfied with it. Upon optimization and further validation of this sensor in patients with swallowing disorders, this new technology has the potential to improve the quality of care provided to patients with swallowing disorders who live in rural, remote, and underserved areas.

20. Physician adherence to clinical guidelines using virtual care platform dependent upon time of day

Lisa Ide

Zipnosis

Two 2019 studies published in JAMA found clinical decisions were impacted by the time of day at which a decision is made. The studies showed that (1) clinician ordering of cancer screening tests significantly decreased as the day progressed and (2) even within an individual physician's schedule, clinical decision‐making for opioid prescribing varied by the timing of appointments.

However, analysis of the time of day impacting adherence to clinical guidelines is underreported for virtual care. The purpose of our study was to determine the clinical adherence rates of virtual care visits based on time of day, and thus better understand how telehealth impacts clinician burnout.

35,127 virtual visits over a 1‐year period (July 2018 ‐ June 2019) were analyzed for adherence to clinical guidelines and for the time of day at which the visit took place. First, the total adherence rates of the virtual visits were calculated. Then, visits were isolated into two groups: one containing 7,343 visits from 8:00 to 10:00AM, the other containing 3,836 visits from 2:00 to 4:00PM. The adherence rates for each group were determined, and a two‐sample t‐test was conducted to measure the statistical significance of the difference in adherence between the groups.

Method: Secondary Data Analysis

Classification of Research: Clinical Outcomes

Result: When using virtual care, change in physician adherence to clinical guidelines depending on the time of day was statistically insignificant. In other words, when using virtual care, physicians are relatively as adherent to clinical guidelines in the morning (8:00 to 10:00AM, when they are most energetic) as they are in the afternoon (2:00 to 4PM, when they are more likely to be fatigued). In the morning, across 7,343 visits using virtual care, physicians were adherent 90.51% of the time.

Conclusions: While in‐person visits may see lower physician adherence to clinical guidelines later in the day, there is no statistically significant change in physician adherence to clinical guidelines using virtual care when comparing the time of day. Asynchronous virtual care aids in removing variation in the quality of care patients receive, regardless of physician fatigue.

21. A Collaborative Telehealth Analytics Platform based on Intelligent Data Licenses

Najib Ben Brahim, Cory Pitt, Edward Zyszkowski, and Alan Pitt

Ignis Health

Disparate data silos prevent whole‐picture care delivery as well as visibility into a health system's telehealth services performance. The current lack of responsible data mobility prevents clinical effectiveness and ultimately diminishes the patient and provider experiences. The problem persists beyond the delivery of care, with operational insights suffering from the lack of transparency and interoperability, specifically regarding enterprise‐wide analytics and insights from disparate sources. Given the nature of telehealth in terms of spanning across multiple clinical specialties and a myriad of vendors, the need for a collaborative analytics platform is critical. We propose a system for cross‐enterprise data unification through a blockchain‐enabled joint network of intelligent data licenses. The intelligent data licenses leverage a shared record of credentials and access logs for the network participants to access, unify, and utilize disparate data sources across health systems. The system lowers the financial and operational barriers associated with traditional hub and spoke analytics models through a seamless process of permissioned and transparent federated data analysis. We have enabled centralized access to unique telehealth insights across the enterprise, specific to each clinical service line. A telehealth ROI model was then modeled based on the combination of financial, clinical and operational data.

Method: Cost Analysis

Classification of Research: Cost Analyses

Result: We've evaluated an implementation of the network in use at two academic medical centers and measured its effectiveness in analyzing operational effectiveness and sustainability for telehealth programs accessing disparate hospital data. Using the ROI analytics model and telehealth insights, the average growth year over year in two of the leading health systems in the United States is 90% in terms of budget, 170% in terms of the number of patients and 200% in terms of clinical specialties offered to their patient populations.

Conclusions: This research provides insight into the potential of collaborative health information transmission and analytics systems. The empirical evidence reinforces our belief that the financial and operational barriers associated with traditional hub and spoke analytics models can be lowered through a seamless process of permissioned and transparent federated data analysis. As demonstrated by the year over year performance improvements to our case studies, we stress the benefits of a health information infrastructure based on intelligent data licenses, as well as its applications for responsible machine learning in achieving improved digital health strategies.

22. Telehealth‐based health coaching using an employee population is effective for weight loss

Michelle Alencar and Kelly Johnson

CSULB; inHealth Medical Services Inc.

Obesity is a public health issue and is associated with other chronic conditions. With the heightened escalation in healthcare costs, employers are trying to find effective ways to improve the health of their employees, while also reducing employee healthcare costs. Currently, employers are leveraging health coaches to provide lifestyle support for employees. Therefore, the purpose of this study was to evaluate the effect of a telehealth‐based health coaching program on weight loss and perceived program value using an employer population.

Method: Observational

Classification of Research: Clinical Outcomes

Result: A total 303 participants were included in an ANCOVA with 111 women (61% obese) and 192 men (62% obese). The average Perceived Program Value (PPV) was 8.4 ± 1.8. The covariate, PPV, reached statistical significance F(1,298) = 12.73, p ≤ 0.001 indicating that average WL% varied across PPV value with weight loss by 0.43% per unit increase in PPV. The average PPV‐adjusted %WL for the program was 3.0% (2.5–3.5%.). There was no statistical difference in %WL between men and women (p = 0.958) nor an interaction between Gender and BMI (p = 0.859).

Conclusions: This study examined the effects of an employee telehealth‐based health coaching program on weight loss. The present study found on average that employees lost on average 3.0% of weight loss, and those that had higher initial BMI lost 1.5% more weight. Overall, telehealth‐based coaching is an effective approach to reducing body weight that is also valued by employees.

23. Telemedicine Enters the Home: Addressing the Needs of Complex, High‐Cost Patients

Neil Solomon and Alexander Li

MedZed

In the United States, approximately 5% of the population accounts for 50% of health spending. Many of these high‐need, high‐cost patients have multiple serious medical conditions compounded by social determinants of health (e.g., homelessness, poverty, transportation) that limit access to care in traditional settings and leave them lacking strong connections to health care providers. Home care models, while effective in serving this population, are difficult to scale due to drive times and provider preferences. A telemedicine enabled home care model can remove these barriers, driving improved clinical outcomes, enhanced quality of life, and decreased costs of caring for this vulnerable population.

LA Care, a large Medicaid insurer, teamed up with MedZed Physician Services, a mobile primary care medical group to redesign the care experience for LA Care's high‐need members. Working from utilization‐based referrals from LA Care, nurses equipped with mobile telemedicine units visit patients in their homes and link to remote PCPs, with the same clinical team responsible for the longitudinal care of the patient.

Method: Cost Analysis

Classification of Research: Clinical Outcomes

Result: Six‐month outcomes showed reduction of inpatient costs from $4,100 PMPM to $1,900 PMPM, which was attributed to improving patients' health, strengthening self‐management skills, providing emotional support, and addressing social needs. Preliminary repeat analysis, to be presented at ATA, shows sustained improvement. Patient recruitment rates and satisfaction are high. Patients graduate when their health improves and they are able to navigate the office‐based delivery system.

Conclusions: For this vulnerable population, frequent ED utilization and excess hospitalization can be a symptom of the failure of the health care system to meet these patients where they are and address their needs comprehensively, inclusive of not only medical, but also behavioral and social needs. A redesigned, telemedicine‐based home care model utilizing mobile nurses and remote PCPs offers a significant opportunity to improve outcomes and reduce costs in patients with complex medical and social needs.

24. Telehealth Outcomes Research: Show Me the Data

Jillian Harvey, Kathryn King, Ryan Kruis, Dee Ford, James McElligott, and Rebecca Beeks

Medical University of South Carolina

Telehealth is often accused of lacking ‘data.’ As a federally recognized Telehealth Center of Excellence, MUSC is charged with creating an outcomes measurement toolkit. There is a need for more rigorous evaluations of telehealth. To date, most of the available measurement guidance assumes a fully implemented telehealth service. As a result, many telehealth programs and quality improvement projects rely on simple counts of programmatic data, and never advance to assess patient or population‐level cost and quality outcomes. Due to the unique data challenges of telehealth delivery in the real‐world setting, telehealth research and evaluation requires innovative data collection and analysis techniques. Utilizing existing conceptual frameworks of quality and outcome monitoring domains for telehealth evaluation, we will present a staged approach to measurement based on the varying levels of telehealth service maturity.

Method: Measurement and Evaluation

Classification of Research: Measurement Frameworks & Tools

Result: The four stages of telehealth service maturity include: 1) pilot, 2) scaling‐up, 3) established program with predictable volumes, and 4) optimized program. For each stage of telehealth service maturity, we identify appropriate metrics, data sources and evaluation tools. To demonstrate the utility of the measurement framework we will describe three active telehealth services (telestroke, school‐based, and tele‐ICU) to demonstrate and report metrics and analysis tools that can be implemented across the various stages of telehealth maturity. The following data sources will be utilized: program tracking data, EHR, and claims data.

Conclusions: The applied telehealth measurement framework adapts the National Quality Forums' telehealth measurement domains, program evaluation models such as RE‐AIM, and the Telehealth Maturity Model to create a measurement tool for telehealth service evaluation. Established outcome domains reflect what is currently recommended in the literature and acknowledge the human variability and process improvement nature of telehealth implementation. The measurement framework is generalizable to all telehealth modalities (synchronous, asynchronous, store‐and‐forward, remote patient monitoring, etc.). In addition, the measurement framework is adaptable to any telehealth service line, technology, or disease condition. The framework will allow for disease/condition specific metrics and guideline‐based measurement.

25. The Digital Therapeutics Effectiveness Chain: An Industry Reference Architecture

Anand Iyer, Malinda Peeples, and Vinayak Shenoy

Welldoc

The interest in and application of digital health and digital therapeutic solutions for the management of chronic conditions is rising exponentially. But, the purveyors of digital health solutions are often faced with unique customization requirements which hinder the inherent scalability that's expected from such SaMD (software as a medical device) solutions. Compounding this conundrum and implementation complexity is the heterogeneity of the customer implementation environment, which can include but are not limited to large payors/insurers, self‐insured employers and integrated health delivery networks (IDNs. We hypothesized a reference architecture that could be used to describe the implementation of digital health using common processes that can be configured to meet the needs of the different operating environments. Associated with the architecture is a set of metrics that can be used across different implementations to measure and benchmark effectiveness.

Method: Identification of a taxonomy of implementation & measurement process for digital health

Classification of Research: Measurement Frameworks & Tools

Result: The digital therapeutics experience chain was described as 6 macro processes: Target, Outreach, Activate, Engage, Support and Report. Top and mid‐tier metrics were defined for each macro and sub‐level process in several customer models. Increases in activation rate (2.6 times), engagement rate (1.7 times) and persistence rate (1.9 times) were noted when outreach and activation first involved physicians and care team as compared to health plan and employer program outreach.

Conclusions: The digital therapeutics engagement chain can serve as an industry reference model to allow standard implementation configurations to be more effectively deployed given the unique nature of the healthcare delivery environment they are targeted for. Future work can leverage the collection of standard metrics across different models as well as similar models in different therapeutic areas, to create a basis of best‐in‐class effectiveness that can be used by others.

26. Value Scorecards as a Tele‐ICU Evaluation Tool

Dee Ford

Medical University of South Carolina

South Carolina (SC) is largely medically‐underserved with limited or no access to intensivists. For this reason, the Medical University of South Carolina (MUSC) sought to develop an outreach tele‐ICU program for rural hospitals. MUSC is one of two national Telehealth Centers of Excellence and tele‐ICU is a pillar program. Our tele‐ICU model uses continuous remote coverage with robust information technology support and two‐way audiovisual communication. This high‐cost program currently supports 10 rural hospitals. Our objective was to evaluate the value, defined as outcomes relative to cost, of participating hospitals with respect to tele‐ICU. We used the balanced scorecard framework to guide evaluation. This business framework holds that no single domain is definitive and that multiple domains and measures should be triangulated to ascertain program value. We used program quantitative data (clinical and financial) as well as qualitative data derived from semi‐structured interviews with stakeholders. We examined the programmatic, clinical, financial, and strategic value of tele‐ICU. Individual measures mapped to each domain were developed and sites received point value attributions for measure that were then tallied by domain. Each site was assigned a composite value score, as well as category‐level scores detailing performance across our four domains.

Method: Implementation Science

Classification of Research: Quality Improvement

Result: We found that all sites derived value from the tele‐ICU program. However, there was substantial variation in relative value with the highest site scoring 85 and the lowest 32. Qualitative factors were key drivers towards better quantitative outcomes. Effective hospital leadership, clear and consistent communication, and an engaged and positive physician leader were key program factors that led to better clinical and financial performance. Strategically, tele‐ICU was a powerful physician recruitment tool and helped extend medical staff capacity. Among quantitative drivers of value, growth in ICU volume, reduced length of stay, and increased patient acuity were integral to financial performance.

Conclusions: Tele‐ICU can provide significant value to rural SC community hospitals in clinical, financial, programmatic, and strategic domains. A strong physician champion represents the single most critical driver of success. The variation we observed in relative value among our four sites highlight the need to develop strategies to optimize tele‐ICU programs. Future work will apply the value scorecard prospectively to identify sites likely to derive maximal benefits from the costly and resource investment in tele‐ICU.

27. Advancing Telehealth at HRSA ‐ The Office for the Advancement of Telehealth

William England

Office for the Advancement of Telehealth, Federal Office of Rural Health Policy, Health Resources and Services Administration

Since 2002, the Office for the Advancement of Telehealth (OAT) in the Federal Office of Rural Health Policy (FORHP) of the Health Resources and Services Administration (HRSA) has funded hundreds of grantees building telehealth networks to connect thousands of rural sites to deliver telehealth services. OAT also funds Telehealth Resource Centers to provide public information and technical assistance to telehealth providers, a Telehealth Research Center to advance the telehealth evidence base, two Telehealth Centers of Excellence that are developing best practices for telehealth, and license portability models to simplify multi‐state telehealth practice.

Method: Descriptive

Classification of Research: Regulatory & Policy Research

Result: In recent years, telehealth at HRSA has expanded well beyond OAT, to include 1,370 HRSA grant awards in 92 different programs that mention or allow telehealth as a component of the award. Most of these awards include telehealth in keyword searches for pending funding opportunities at Grants.Gov.

Conclusions: This presentation will discuss the breadth of funding opportunities for telehealth in HRSA programs as well as a detailed look at OAT programs and how they help clinicians, facilities, and health systems considering, starting, or expanding telehealth services.

28. The Results Are In! School‐Based Telehealth and Emergency Department Utilization

Kathryn King and James McElligott

Medical University of South Carolina

School‐Based Telehealth (SBTH) has become a popular modality for improving access to care, however supporting evidence for improved clinical effectiveness and associated changes in care utilization patterns has been slow. A SBTH in South Carolina was designed to target childhood asthma as it is one of the most common and costly chronic diseases of childhood. This quasi‐experimental study utilized publicly available data from the South Carolina Medicaid database. A 6‐year longitudinal child‐month panel of data from 2012–2017 was constructed which included children ages 3–17 who were enrolled in SC Medicaid, had at least one primary or secondary diagnosis of asthma and lived in the intervention county or one of the four control counties with no SBTH program. All county‐specific time invariant confounders and secular trends similarly affecting all counties were controlled for. Having an ED visit in a given month in which the primary or secondary diagnosis of asthma was then compared for each student in the pre‐telehealth period (2012–2014) vs the post‐telehealth period (2015–2017) using a linear probability model with county fixed effects and time (quarterly) fixed effects in which unit‐of‐analysis was child‐months. Standard errors were adjusted for heteroscedasticity and a sensitivity analysis was conducted using child fixed effects.

Method: Quasi‐experimental

Classification of Research: Clinical Effectiveness

Result: The SBTH program was associated with a 6.89‐percentage‐point, or 22% overall reduction (95% CI: −11.84, −1.94 P < 0.01) in ED visits from 2015–2017 and 11.07‐percentage‐point reduction, or 37% overall reduction (95% CI: −18.12, −4.02 P < 0.01) in ED visits by the 3rd year. Child fixed effects produced consistent estimates.

30. Addressing Barriers to Adoption for Clinicians in Pediatric Primary Care Telehealth

Meher Kachwala, Vandna Mittal

Stanford Children's Health

Stanford Children's Health (SCH) has 28 primary care locations and >100 providers within its PCHA network that span over 125 miles. In FY18, SCH began enabling telehealth across these locations and has completed >400 telehealth visits since then. During the same time, the specialty side completed >4,500 telehealth visits. Why the difference in speed to adoption? Telehealth adoption in primary care pediatrics has been challenging compared to adoption in specialty pediatrics. Compared to when offered as a stand‐alone service, the adoption of primary care telehealth is accelerated when it is utilized as an enabler for programs with pre‐identified use cases for defined patient populations and their needs. In FY18, SCH introduced telehealth virtual visits as a convenient solution that met clinical needs. In combination with trainings, providers received recommended visit reasons to try out virtual visits. In FY19, SCH took a different approach to encourage telehealth adoption. Provider champions designed programs, namely Virtual PrEP and eConsults, to address specific patient needs and enthusiastically embraced digital health tools to address these needs. In the end, these programs led to telehealth being a key enabler.

Method: Descriptive

Classification of Research: Clinician Experience

Result: In FY20, the target is 1500 telehealth virtual visits in pediatric primary care coupled with new care delivery models. Initially in FY18, telehealth was promoted mainly as virtual visits across a variety of diagnosis and the adoption was slow. E.g. In February 2019, only ∼2% of possible visits that were actively “recommended” for telehealth were converted to virtual visits. In FY19, the approach was to design programs that addressed specific patient needs and populations. Telehealth here was proposed as a value‐add component and enabler.

31. Utilizing Telehealth to decrease hospital admissions in Spinal Cord Injury

Broderick Flynn

Veterans Affairs

Telehealth was utilized to reduce the amount of emergecny room visits for a patient with a high readmission rate. The patient was being admitted for a non‐emergent issue multiple times a year and costing the facility valuable time and resources. Telehealth was used to provide in home monitoring and education and the patient had a significant drop in readmission and bed days.

Method: Observational

Classification of Research: Quality Improvement

Result: Before telehealth intervention patient bed days totaled 102. After telehealth intervention was instituted bed days dropped to 32 days. This was conducted over 6 months and a retrospective review of 6 months. Patient admission total before telehealth intervention was 8 total admission. After telehealth intervention total number of admissions was 5.

Conclusions: Telehealth is a safe and effective way to monitor and triage patients while in the home and is an effective way to reduce frequent readmissions.

32. Telehealth: A New Innovative Approach in Anticoagulation Management

Maria Rebecca Bernstein, Libiny John, Susan Sciortino, Elise Arambages, Danielle Auletta, Andrew Tucci, and Alex Spyropoulos

Northwell Health at Lenox Hill Hospital

Anticoagulation Management Services (AMS) are known to improve warfarin management in the outpatient setting. The guideline recommendations are well established and indicate that patients who receive a Vitamin K antagonist (VKA) should be under the care of an experienced and specialized anticoagulation clinic. The goals of the Telehealth Anticoagulation Management Service are to: 1) increase patient satisfaction 2) reduce turnaround time for results by providing the International Normal Ratio (INR) results in real‐time 3) increase patient compliance to INR testing and 4) improve system wide anticoagulation care by increasing center Time in Therapeutic Range (cTTR) metrics to high quality metrics such as 65% and above. The overarching goal of our health system Telehealth program was to develop a collaborative care telemedicine INR model in collaboration with core laboratories (Patient Service Centers [PSCs]) in order to improve patient quality metrics on warfarin. In this model, the Telehealth Anticoagulation and Clinical Thrombosis Service (Tele‐ACTS Center) team was able to provide distant care for the outpatient population on warfarin maintenance therapy using a virtual telemedicine INR model located in a distant location to the PSC. Using this model, we were able to improve center‐based TTR by 45.73%.

Method: Implementation Science

Classification of Research: Quality Improvement

Result: Individual TTR's (iTTR) from eight patients were analyzed prior to enrollment and ranged from 18.0%‐50.7%, with a mean cTTR of 32.12% and a standard deviation of ±10.57. Post enrollment, iTTRs ranged from 61.14% ‐ 90.4%, with mean cTTR of 77.85% (SD ±9.68). (Figure 2). The Telehealth INR pilot project was able to improve delta center‐based TTR by 45.73% (SD ±14.5) as of March 1, 2019 pilot project data results.

Conclusions: Incorporating key elements of Anticoagulation Management Service into our Tele‐ACTS INR pilot program showed a center‐based improvement of TTR (cTTR) by 45%. Using a Telehealth INR model is a new, innovative approach for AMS that has potential to yield excellent quality metrics over UMC. This Telehealth ACTS INR pilot project is a “proof of concept” which has potential to promote workflow efficiency while decreasing the time and cost of care. The role of Telehealth INR in revolutionizing anticoagulation health care delivery of patients on chronic warfarin management needs further investigation and validation.

33. FOXTROT Forward Operating Base EXpert Telemedicine Resource Utilizing MObile Application for Trauma

Jennifer Stowe and Gary Legault

United States Army Aviation Research Lab

During Operation Iraqi Freedom (OIF) and Operation Enduring Freedom (OEF), 10–15% of combat‐related trauma injuries involved the eye. There were 170 ocular trauma cases reported in 2018 occurring in deployed locations. The military ophthalmologist's primary mission is to be prepared to manage ocular trauma, especially in an austere environment; however with deployments across the globe, access to ophthalmic care is not easily accessible. The primary purpose of our research is the development of an operationally secure, Health Insurance Portability and Accountability Act (HIPPA) compliant, mobile application (mApp) to provide ophthalmic care to any remote deployed location through a teleophthalmology called ‐ Forward Operating Base EXpert Telemedicine Resource Utilizing MObile Application for Trauma (FOXTROT). The development of this application will effect the Aerospace Medical community by reducing the number of MEDEVACs needed in theater operations. In addition, future buildout of this application will involve developing a platform for aviation flight physicals. Currently in the military, we have one primary method for teleophthalmology involving pagers and non‐secure or secure Defense Switched Network (DSN) phone, satellite phone, or cellular phone if Wi‐Fi is available.

Method: Implementation Science

Classification of Research: Quality Improvement

Result: FOXTROT is currently being fielded in Afghanistan field hospitals. Based on an independent assessment, FOXTROT saved the Military $2.4M by preventing costly medical evacuations out of theater.

Conclusions: There is currently limited access to ophthalmic care at forward‐operating bases, especially with ocular trauma. Teleophthalmology is currently limited in the military and in the civilian sector. FOXTROT teleophthalmology mApp will improve and extend ophthalmic trauma care in remote deployed environments.

34. Nurses Place at the Innovation Table: One Universities Experience

Tina Gustin

Old Dominion University

The health care sector of the United States is plagued with high costs, variable access, uneven quality, and some of the poorest outcomes among developed countries. Industry leaders have identified healthcare innovation as a method to turn these negative numbers around. Healthcare programs have been challenged to include innovation education in the curricula for healthcare administrators and leaders. While nurses have always been innovators at the bedside, they too often do not see themselves as innovators or design thinkers. Unfortunately, nurses are not prepared to be thought leaders and innovators in the telehealth industry. Several schools of nursing are now offering combined degrees in health innovation or biomedical technology for graduate and doctorate level students. This is not practical for all schools. This presentation will discuss innovative teaching methodologies and student experiences that have leveraged the spirit of innovation and design thinking. Project outcomes future directions will be discussed.

Method: Observational

Classification of Research – Other: Educating healthcare professional for innovation and design thinking

Result: Since opening 8 faculty from nursing, engineering, and computer science, and 12 students from nursing, clinical counseling, engineering, and computer science have participated in innovative product design. The following projects are in active development: 1) home sensors aimed at allowing adults with autism to live independently, 2) a patient controlled robot aimed to engage medically isolated children, 3) a training platform for teaching telehealth etiquette skills using artificial intelligence and video overlay, 4) a mobile application that will be used for trauma providers at a local children's hospital, and 5) remote patient monitoring project for children on hospice care.

Conclusions: The experiences and projects that are in development would not have occurred without the deliberate collaboration. In each of these experiences students from various professions have been able to work with faculty. Prior this collaboration the engineering department outsourced projects. The students have learned to embrace innovation and realize the process of innovation. C‐TIER is beginning to build purposeful interprofessional innovation content as a result of this early work.

35. Telehealth Implementation Evaluation: A Framework for Measurement Strategies and Tools

Emily Johnson, Katie Sterba, Claire Macgeorge, Kathryn King, Ron Teufel, Ryan Kruis, Kathryn Hale, Annie Andrews, and Dee Ford

Medical University of South Carolina

Telehealth offers an ideal platform for implementation science evaluation, as outcomes regarding telehealth integration into practice have been mixed, despite clinical efficacy in specific populations. In this case study, we utilized implementation science methodologies to apply an evidence‐based framework to identify barriers and facilitators to telehealth service implementation. The evaluation team within the Telehealth Center of Excellence at a medical university developed a library of evidence‐based implementation evaluation tools, based on existing frameworks. These measures included implementation tracking logs (monitor implementation outcomes), site surveys (assess organizational demographic and structural characteristics), staff/champion surveys (measure readiness, teamwork, barriers) and interview guides. We describe a mixed methods case study approach to an evaluation of the implementation of a school‐based telehealth asthma program in rural schools throughout South Carolina. This study was guided by an adapted version of the Exploration, Preparation, Implementation, and Sustainment (EPIS) model, which evaluates implementation processes across outer (external to organization) and inner (within organization) context levels. Three data sources were triangulated to compare perspectives and describe inner/outer context factors associated with implementation outcomes, based on factors in EPIS model. Implementation tracking log data classified schools as high, medium or low performance based on number of completed asthma telehealth services.

Method: Implementation Science

Classification of Research: Measurement Frameworks & Tools

Result: Focus groups were conducted with medical university telehealth program providers and staff (N = 5) and school nurses and administrators from 8 South Carolina schools (N = 11). Surveys were also completed by state school nurses (N = 33). School nurses reported high satisfaction with the asthma program and highlighted the key benefits of providing timely access to care and medications for students. A key outer context facilitator to implementation was the collaborative engagement style of the medical university in leveraging resources for schools. Outer context barriers noted were lack of parental involvement and understanding of the program and lack of community resources (pharmacies, transportation).

Conclusions: Key inner context facilitators to implementation included nurse commitment and leadership, yet competing demands in the school environment was considered a significant barrier. This novel approach to telehealth implementation evaluation, guided by implementation science principles, was beneficial in describing factors of successful implementation of an asthma telehealth school program. The approach can be utilized across different telehealth services, to guide development of strategies and tools to support telehealth engagement and provision of high‐quality care to populations in need.

36. Food and Exercise as Virtual Medication for Diabetic Treatment

Bao Tran

Tran & Associates

Diabetes is a major disease:

Prevalence: In 2015, 30.3 million Americans, or 9.4% of the population, had diabetes. Approximately 1.25 million American children and adults have type 1 diabetes.

Undiagnosed: Of the 30.3 million adults with diabetes, 23.1 million were diagnosed, and 7.2 million were undiagnosed.

Prevalence in seniors: The percentage of Americans age 65 and older remains high, at 25.2%, or 12.0 million seniors (diagnosed and undiagnosed).

New cases: 1.5 million Americans are diagnosed with diabetes every year.

Prediabetes: In 2015, 84.1 million Americans age 18 and older had prediabetes.

Deaths: Diabetes remains the 7th leading cause of death in the United States in 2015, with 79,535 death certificates listing it as the underlying cause of death, and a total of 252,806 death certificates listing diabetes as an underlying or contributing cause of death.

Method: Observational

Classification of Research: Patient Experience

Result: We trialed various high starch food in the morning after fasting for at least 10 hours, measure the glucose level with a novel bioimpedance sensor that can infer the glucose level, validate the result with a CGM device, and also tested the impact of exercise on the glucose level. The sensor is a watch device that uploads data where a care provider can monitor and advise the patient remotely.

Conclusions: Glucose level can be controlled using food and exercise. A remote coach enables a non‐professional user to control his glucose level, leading to diabetic control using food, exercise, and remote feedback through telemedicine approach.

37. Interdisciplinary Telehealth: Chronic Pain and Opioid Management

Rachel Wong, Kimberly Noel, Patricia Ng, Tracey Spinnato, Erin Dainer, Mark Lerman, Amit Kaushal, Alice Fernan

Stony Brook Medicine

Patients with chronic pain and psychosocial complexity are challenging to manage in primary care. We need to train physicians to provide interdisciplinary care using tools that facilitate collaboration. Technology such as telehealth can reduce barriers in patient access to interdisciplinary care, but effective use of telehealth will require additional competencies in virtual visits, electronic communication and virtual conferencing. Our aim was to (1) create an interdisciplinary, longitudinal simulation which integrates team‐based telehealth care of complex patients with chronic pain and (2) develop evaluation metrics for synchronous and asynchronous communication skills with interdisciplinary team members.

Method: Survey/Qualitative

Classification of Research: Quality Improvement

Result: Fifty‐six internal medicine residents completed the simulation, with a survey response rate of 94–100%. Post‐OSCE, confidence improved in adjusting the tele‐visit camera from 32% to 85%, communicating effectively from 32% to 79% and establishing rapport from 39% to 91% (p < 0.0001) and likelihood of using telehealth in future practice increased from 54% to 84%. Post‐teleconference, confidence improved from 33% to 84% in electronic care coordination, 26% to 79% in facilitating a teleconference and 32% to 82% in addressing interdisciplinary needs (p < 0.0001). Specialist evaluation of resident EMR consultation identified gaps in residents asking a specific question and documenting plan for follow up.

Conclusions: A longitudinal, interdisciplinary simulation addresses the team‐based nature of chronic pain management and can train residents in synchronous and asynchronous virtual patient care. Education using simulation can improve competencies in virtual visits and teleconferencing and may increase the likelihood that residents will use tele‐medicine in the future. Future interventions should focus on areas of improvement for resident electronic communication to specialists and creating additional longitudinal simulations that mimic real‐life primary care.

38. Augmented Intelligence in Dermatology: Purposeful Innovation to Promote Patient Care

Ivy Lee and Trilok Tejasvi

American Academy of Dermatology

Augmented intelligence (AuI) garners much hype for potential disruption in healthcare yet it is shrouded in mystery for many providers. Identifying and addressing gaps in knowledge, practice, and research may lead to efficient, effective development and implementation of AuI technologies.

Method – Other: Position Statement (Policy)

Classification of Research: Regulatory & Policy Research

Result: The Position Statement describes core concepts of AuI and principles of human‐centered design. This guiding document summarizes the steps of developing, deploying, validating, and integrating AuI technologies in Dermatology. Potential pitfalls and ethical challenges are discussed as well.

Conclusions: AuI has the potential to transform our collective and personal experience of health, healthcare, and wellness. To achieve this potential, deliberate and diligent efforts must be taken to engage and collaborate with stakeholders to create policies that promote AuI that is high‐quality, inclusive, equitable, and accessible.

39. Feasibility of televisits in a men's sexual and reproductive health practice

Jamie Pak and Peter Stahl

Columbia University Irving Medical Center

There have been several reports in urology supporting the use of televisits, which have led to reduced cost and time expenditures with high patient satisfaction. However, there is very little data describing the feasibility of integrating synchronous virtual visits (VVs) into clinical care for men with reproductive and sexual disorders. Our hypothesis was that the integration of televisits in a men's sexual and reproductive health practice would result in significant patient cost and time savings, while facilitating discussion of lab/imaging results and changes in medication with a high rate of subsequently scheduled VVs. This was a retrospective review of all VVs performed by a single provider from 6/19/19 to 10/23/19. Patients were selected for VVs based on the clinical judgment of the provider. Variables collected included age, race, insurance type, reason for visit, distance and time of round‐trip commute at time of last in‐person visit as per Google Maps, cost of round‐trip commute and 2‐hour parking at location of last in‐person visit, modality of next visit, plan for procedure, and medication modification.

Method: Observational

Classification of Research: Clinical Effectiveness

Result: Fifty‐eight patients with 68 unique VVs were scheduled: 88% of VVs proceeded as planned, 3% rescheduled, 3% switched to in‐person visit, and 6% cancelled for undocumented reasons. All patients were English‐speaking males, 93% had private insurance, with median age 40 years (IQR 33–48). Median distance, time, and cost saved were 18.5 miles (9.8–44.7), 88.5 minutes (62.5–115.0), and $45 ($27‐$60), respectively. Of 62 performed VVs, 48% were for multiple complaints, most commonly erectile dysfunction.

Conclusions: This report supports the feasibility of VVs in a men's sexual and reproductive health practice. There was a high rate of patient compliance with VVs, leading to significant patient savings in time and money, and facilitating discussion of lab and imaging, medication changes, and procedure planning. A prospective, randomized study comparing patient satisfaction with VVs versus traditional office visits is underway at our institution.

40. Telehealth ‐ A Step to Transform Gap by Engaging Home‐Based Female Doctors

Syed Ali Hussain

ChildLife Foundation

The World Economic Forum's (WEF) Gender Equality Report4 released in December 2018, demonstrated Pakistan to be the second worst performer in the Gender Equality Index. Pakistan is facing severe shortage of doctors in utilizing their profession particularly women those have obligations of family or children. Research5 from Pakistan demonstrates that many female doctors prefer teaching and administrative positions within healthcare organizations due to day time fixed working hours that also not obliged from family to carry out their profession in different duty timings. It is suggested that holistically evolve technological approach by engaging home‐based doctors to participate in serving patients from home that could create an impact and help in reduce the doctor's gap. The failure of women to practice medicine is widely understood as the major cause of overall shortage of physicians in Pakistan with a ratio of only 0.83 physicians per 1,000 population6 and the major cause of shortfall reason is due to only 25% female graduates are practicing professionals.

Method – Other: Organization Experience & Survey

Classification of Research: Access to Care/Information Technology

Result: The failure of women to practice medicine is widely understood as the major cause of overall shortage of physicians in Pakistan with a ratio of only 0.83 physicians per 1,000 population6 and the major cause of shortfall reason is due to only 25% female graduates are practicing professionals. The research showed that interest of several families of female professionals are in serving from home in specified time period. The best part of the survey that comes out was the acknowledgement on home‐based work scenario by husbands. So how can we coup up doctors shortfall and reduce the gap to serve patients? The solution lies in a telehealth system which is a revolutionary step towards involving in particular female doctors from home and utilize their education by engaging them to work from home and serve patients.

Sustainable mechanism is required to put this dream comes true prior to take the home‐based doctors on‐board. High‐level measures those need to take includes but not limited to technical infrastructure, doctor's evaluation by senior consultants, one‐month refresher course that includes one month on ground services in ER and on‐going online trainings schedules, etc.

The technical mechanism magic to transform doctor's gap in particular female doctors starts by connecting remote ERs to control room where senior consultants providing their best advises over the video conferencing equipment. The video conferencing (VC) connection established on H.264 protocol with high‐definition video resolution of 1080p that transmit video over 30 frames per second (FPS). The video stream process encodes video from remote ER site and decode at aggregation site. The VC codec placed in control room has the capacity of 8 remote sites connections, however, the remote site controller has the capacity of 2 connections to be facilitated at a time. Each consultant doctor in control room is assigned 8 ER Resus room remote views i.e. one video conferencing codec for each doctor that allows to address one ER resus room patient at a time., therefore, the control room manager forward the queued call to home‐based doctor. The connection medium shall be on optic fiber with encrypted network of layer2 to be deployed between the control room and remote ER sites. The speed of 2mbps CIR (committed information rate) bandwidth is allocated for each remote ER connection for 1080p video broadcasting.

The VOIP setup as hotline also has been deployed for voice communication that help doctors of remote and control room to communicate hassle free with each other. Each doctor at aggregation site and remote ER resus room has assigned identical extension number over which communication are made. Call forwarding function to control manager's extension number has been configured that forward queued call from ER to control room doctor's extension when doctor's numbers are engaged in advising any ER. As soon as the control room manager gets informed on required consultation needed at any ER, he/she takes the remote session of that respective codec through Polycom real‐presence on a separate computer system and share the screen immediately with doctor sitting at home over Microsoft Team. The doctor sitting at home with laptop or android / Mac mobile handset with enabled Microsoft Team services shall then access to view the shared video screen from aggregation site with pan, tilt, zoom (PTZ) functions of remote ER camera. The home based doctor after examining patient through Microsft Team shared screen, give advises of consultation, General treatment orders, CPR cycle, etc to on‐ground doctor over VOIP and maintain records in web based HMIS against identical MR (Medical Record) number of patient.

We are optimistic that evolving cost‐effective revolutionize telehealth solution, a life saving virtual‐eye, could transform the doctors gap by engaging home‐based female doctors and will become a successful step towards minimize the doctor's short fall, saving children life in Pakistan and serving humanity at max.

Conclusions: By engaging home‐based female doctors through telehealth solution, we will fill the gap of doctors shortage for serving humanity at max.

41. Examining Patient and Caregiver Telehealth Satisfaction in the Veterans Health Administration

Gail Castaneda, Amanda Olney, Marla Kaufman, Mi Jung Lee, Consuelo Kreider, Jennifer Hale‐Gallardo, Kimberly Findley, Zaccheus Ahonle, and Sergio Romero

Department of Veterans Affairs

The U.S. Department of Veterans Affairs' (VA) Office of Rural Health (ORH) supports the health and well‐being of rural Veterans via the dissemination of Enterprise‐Wide Initiatives. In 2017, the Telerehabilitation Enterprise Wide Initiative (TREWI) implemented a hub and spoke model aiming to expand Veteran access to rehabilitation services. Four VA centers located in Minneapolis, Richmond, San Antonio, and Seattle, were recruited to serve as TR‐EWI Hub sites providing care to spoke sites in rural areas. This work examines the Seattle hub site's patient and caregiver telehealth satisfaction for fiscal year (FY) 2019, quarters 2 through 4. A total of 337 telehealth satisfaction questionnaires were administered over the telephone to Veterans or their caregivers for telerehabilitation care received. Descriptive statistics were obtained on clinic, session modality, respondent sex, and respondent role (patient or caregiver). Nine telehealth satisfaction questions were rated on a 5‐point Likert scale. As the data were positively skewed, it was necessary to dichotomize based on Strong Agreement compared with those who Did Not Strongly Agree. Independent samples t‐tests were conducted to determine whether differences were present by sex and between satisfaction groups for the overall sample and by respondent role.

Method: Observational

Classification of Research: Patient Experience

Result: Strong Agreement with overall telehealth visit satisfaction was reported by 81.2% of respondents. Respondents indicating Strong Agreement with their overall visit satisfaction displayed significantly higher mean differences than counterparts on average telehealth satisfaction and across each satisfaction question (p < 0.001). Significantly higher mean differences (p < 0.001) were detected for females across the “comfortable”, “listen”, and “scheduling” satisfaction questions.

Conclusions: Telehealth satisfaction data indicate more than three‐quarters of respondents report Strong Agreement with their overall clinic visit satisfaction. While a much smaller percentage of respondents still indicate Agreement with their overall clinic visit satisfaction, this work underlines distinct differences between respondents who indicate Strong Agreement with overall clinic visit satisfaction and those who do not across average telehealth satisfaction and specific telehealth satisfaction questions. Telehealth satisfaction item‐level differences detected for caregivers (indicating Strong Agreement with overall clinic visit satisfaction) and female respondents underscore telehealth service delivery implications for tailored programming.

42. Innovative Post‐Operative Physical Therapy: Using Telehealth to Enhance Patient Experience & Access

Christina Crawford

Department of Veteran Affairs

Following a total‐knee arthroscopy (TKA) at the Phoenix Veteran Affairs (VA) Medical Center, a Veteran receives home physical therapy (PT) with a community provider. Once home, a Veteran has historically had no scheduled communication with the Phoenix VA PT team until they start out‐patient services. There is also a frequent delay in starting in‐home PT, along with other challenges and barriers, that postpone access to a physical therapist in both the home and out‐patient settings. Knowing the use of telehealth in PT is a growing practice with the VA, a process change was implemented. The telehealth PT team at the Phoenix VA Medical Center collaborated with the orthopedic team to address the gap in access to the VA PT after a TKA. Through a pilot program, Veterans are now connecting with VA PT before surgery to establish a post‐operative telehealth PT plan of care to complement standard PT.

Method: Mixed methods: Descriptive and Qualitative

Classification of Research: Access to Care

Result: In a retrospective data analysis, the implementation of telehealth services and a process change has improved access to VA PT. More than 80% of post‐operative TKA Veterans have communication with a VA PT in seven days or less. Over two quarters, 35% more Veterans started out‐patient PT in the recommended time (R2 = 0.67). Veterans report satisfaction and multiple benefits of post‐operative telehealth. Perceived Veteran benefits include decreased resource burden and feeling connected with the health care team. Providers report benefits to include essential home exercise program feedback, home safety considerations, and pain management strategies.

Conclusions: Access to out‐patient PT can be improved with pre and post‐operative communication for TKA Veterans. Veterans and providers have identified a number of patient‐centered benefits with the utilization of telehealth physical therapy after a TKA. Implications for the future include expanding post‐operative telehealth PT services to other areas. Future research may focus on functional outcomes and the cost‐effectiveness of telehealth PT.

43. Requiring Video Calls for Telemedicine May Contribute to Health Care Inequality

Lauren Broffman

Ro

Increasing access to health care via telemedicine is a promising solution to well‐documented health inequities. Prior research demonstrates that lack of broadband availability is both associated with health inequities and is a hindrance to telemedicine usage and therefore undermines telemedicine's potential. However, earlier research on the relationship between broadband availability and telemedicine use fails to distinguish between modalities with differential levels of broadband dependency; care delivered via video calls requires higher internet speeds than care delivered asynchronously. As some states require video visits while others allow asynchronous or phone‐based telemedicine treatment, the relationship between broadband availability and telemedicine usage might vary in conjunction with state‐level policy, with implications for the ability of telemedicine to reduce health inequities.

Method: Observational

Classification of Research: Regulatory & Policy Research

Result: Poisson regression predicted the number of telemedicine patients in a United States county depending on whether broadband availability in that county was low, medium, or high and whether the county was in a state requiring video visits. In video visit states, there was a clear linear relationship between broadband availability and model‐predicted number of patients (e.g. 0.4 patients (CI = 0.5, 1.4) in low availability counties vs. 2.5 patients (CI = 2.2, 2.6) in high broadband availability counties).

Conclusions: Results suggest that state laws requiring video visits perpetuate health inequality by limiting telemedicine usage as a function of broadband access, but broadband availability does not limit telemedicine usage in states where less broadband‐dependent modalities, like asynchronous and phone calls, are allowed.

44. Implementing a Digital Navigation Program to Improve Outcomes for Total Joint Replacement Patients

Farah Fasihuddin, Jason Rogers, Morgan Black, Jonathan McLaughlin, Shashank Garg, and Ashish Atreja

Icahn School of Medicine at Mount Sinai

Evidence‐based patient education and consistent, timely communication is key to ensuring good outcomes among joint replacement patients. Mount Sinai Hospital (MSH) participates in the mandatory CMS bundle for comprehensive joint replacement (CJR). MSH's bundled payment strategy focuses on the development of a standardized model of care, built around evidencebased best practices to achieve the triple aims of strengthening population health while controlling cost and improving the quality of care. MSH launched a comprehensive softwaredriven digital navigation program (DNP) to guide joint replacement patients and their caregivers through pre‐surgical preparation and recovery. The objective was to to improve the quality of care for joint replacement patients through creation of a digital navigation program specifically tailored to Medicare patients (age 65+) across the continuum of care. Mount Sinai App Lab, in collaboration with the Department of Orthopedics, developed three digital therapeutic modules that were delivered through the RxUniverse Digital Medicine platform. These automated messages, programmed to send at specific times, included exercise instructions, medication reminders, and suggestions for how to prepare the home for optimal recovery. Messages targeted key patient outcomes: length of stay, readmissions, ambulation on postoperative day 0, and discharge disposition. Staff “prescribed” each module to patients.

Method: Implementation Science

Classification of Research: Quality Improvement

Result: Clinicians, patients, and their caregivers were receptive to the DNP. Patients showed a high rate of engagement, clicking links to educational content 873 times, and patients called their surgeon's office when prompted.

After 9 months, clinical outcomes for the DNP were compared to other Medicare patients who had not received it. DNP patients had significantly shorter length of stay (2.81 vs. 4.31 days). They also had a lower readmission rate (1.9% vs. 2.9%), as well as a higher rate of discharge to home (87.8% vs. 64.3%) and were more likely to ambulate on the day of surgery (47.9% vs. 33.3%).

Conclusions: The TJR DNP provided a direct, automated channel to educate and support patients at each stage of care. It demonstrated that digital navigation technology can be used even among nondigital native populations. It resulted in a significantly reduced length of stay and hospital readmissions among participating patients. Next steps include scaling the program across the health system and adding Spanish‐language support. Similar programs are now being implemented for other surgical and disease use cases.

45. Interactive Care Plans – Implementation of a novel digital solution for Systolic Heart Failure

Lukas Manka, Laura Christopherson, Julie Brown, and Megan Strole

Mayo Clinic

Interactive Care Plans are a digitally enabled, actionable record of guidance that is an extension of a patient's care at Mayo Clinic. In October of this 2019, a Systolic Heart Failure care plan was implemented in the medical practice. This plan enables the care team to manage patients who use a mobile application that is integrated with their electronic health record. The care plan was implemented to enhance patient engagement, improve patient outcomes and limit the number of face‐to‐face visits and hospital readmissions. Through the mobile application, education is delivered to the patient, the patient is prompted to report symptoms and vitals, and patients are able to send a secure message to their provider. By using the care plan, patients are empowered to care for themselves and understand care expectations as well as make informed decisions about when to seek help.

Method: Observational

Classification of Research: Patient Experience

Result: Results will include patient and care team satisfaction levels, enrollment volumes and utilization metrics. Patient utilization data and patient and staff reported feedback from using the Systolic Heart Failure care plan will be shared.

Conclusions: Best practice and lessons learned will be shared, covering the full cycle of development, including design, implementation and quality outcomes. Recommendations for what quality measures to use to monitor care plan effectiveness.

46. Reducing Hospital Admissions Using Interreality Care on High Risk Patients at Managed Care Setting

Maria Camila Patino, Irene Kouz, Nicholas Sanfilippo, Kaelin E. Demuth, and Michael Shen

Duxlink Health

Reducing unnecessary hospital admission/ER visits on HMO/MSO/ACO pts is very challenging beyond traditional 30/90‐day readmission threshold. Our study focused on expanding an Interreality (On‐Site & On‐Line) Care (IC) model (EHJ 2018;39:S225) in high‐risk/cost, CHF/COPD pts in HMO setting.

Method: Implementation Science

Classification of Research: Cost Analyses

Result: IC created by integrating: On‐Site Care: ARNP testing at pts residency; On‐Line Care: 24/7 monitoring & specialty cardiology/pulmonary intervention; and new high‐risk care system with telemedicine, device, monitoring, protocols & management. Group of 29 Medicare Advantage pts with multiple admission/ER visits (mean:6/patient/year) were enrolled by MSO based on CHF/COPD & costs. Mean age was 76 (range 55–96) with 35% HFrEF. Admission/ER visits normalized by total days baseline and compared between IC and conventional care. IC showed not only lowered proBNP and a 96% total CHF/COPD Admission/ER, but also a 94% decrease for ALL admission/ER visits with savings of $1.3M.

Conclusions: To our knowledge, this is the 1st study to test IC on reducing admission/ER visits in high‐risk/cost, CHF/COPD pts in HMO setting. Preliminary results show IC can improve both disease‐specific (CHF/COPD) and overall quality of care along with costs. Compared to conventional analytics and administrative/financial models for unnecessary admissions, integrating new technologies to treat high‐risk pts may deliver significant clinical and financial benefits to healthcare.

47. Partnering with National Society Supporting Episodes of Care with Digital Navigation Program

Natalie Bishop, Ashish Atreja, Sravya Kurra, Brian Wasielewski, Shashank Garg, Sarthak Kakkar, Haydee Garcia, Usman Baber, Samin Sharma, and Annapoorna Kini

Rx.Health

Existing pre‐procedure care models that inform and prepare patients for cardiac catheterization are labor and resource intensive, resulting in procedural cancellation, loss of revenue, and poor patient satisfaction. Further, post‐procedure adherence to antiplatelet agents and engagement with lifestyle modifications remain significant targets for improvement in cath lab “episodes of care” designated by CMS. Our overall goal is to support cath lab “episodes of care” through automated digital navigation programs (DNPs), optimizing the patient experience and improve outcomes. As part of the American College of Cardiology (ACC) Transformation Network, Mount Sinai Health System (MSH) Cath Lab led an inaugural cohort of partnering sites to share the approach, process, and results from digital transformation of cardiac care. We implemented RxUniverse (Rx.Health, inc), a cloud based platform that allows prescription of digital health assets from EHRs into DNPs which collate apps, education, and reminders for patients undergoing percutaneous coronary interventions. The DNPs are built from cath lab evidence best practices to support pre and post procedure care and are approved by the ACC innovation advisory group. The DNP consists of automated messages, programmed to send at specific times to the patient's smartphone, including pre and post op instructions, medication reminders, and much more.

Method: Survey/Qualitative

Classification of Research: Quality Improvement

Result: Patients were asked how likely they were to recommend the DNP received to those who needed them on a scale of 0–10, allowing for calculation of a Net Promoter Score (NPS). 82% of respondents answered “extremely likely” to recommend the program with an NPS of 75 and 73 for the pre‐and post‐catheterization pathways, respectively. Electronic patient reported outcomes (ePROs) reflecting adverse events, adherence, and quality of life were collected and summarized over 30 days. Among patients who completed the ePRO (n = 118) 1.7% reported a heart attack, 2.5% reported being hospitalized due to bleeding, and 3.4% reported non‐compliance with medications post‐discharge.

Conclusions: Prescription of automated pre‐ and post‐procedure digital navigation program for cardiac catheterization have been shown to be effective in delivering timely information to patients, achieving a high level of patient engagement and satisfaction as measured by NPS. We have also demonstrated the feasibility of collection of ePRO data over 30 days, which is linked to immediate interventions. As part of the ACC Transformation Network, the digital navigation program is now expanded to include other procedures and conditions relevant to cardiology and will be reported on a national scale with long term multi‐site follow‐up outcomes.

48. Impact of engagement w/ digital navigation program on reduction in no‐shows & incomplete colonoscopy

Natalie Bishop, Farah Fasihuddin, Ashish Atreja, Sravya Kurra, Shashank Garg, Jason Rogers, Sarthak Kakkar, Gaurav Narang, Julian Maximilian, and David Greenwald

Rx.Health

About 20–50% of patients scheduled for colonoscopy procedures do not show up for procedures (no‐shows) or reschedule their procedures at a later time. This has a significant impact on efficiency in the endoscopy suite, as well as an overall increase in the cost of care. We have previously reported on the impact of digital navigation on improvement in bowel preparation through the automation of pre‐procedure guidance. Our objective is to evaluate the impact of engagement with digital navigation program (DNP) on overall incomplete colonoscopy rate at two endoscopy centers enrolled in the American Gastroenterological Association (AGA) endoscopy transformation initiative. A quasi‐experimental design was used to compare cohorts of patients scheduled for a colonoscopy in two hospital‐based endoscopy centers in an integrated delivery network. All patients received usual care in addition to time‐ and text‐based clinical rules and multimedia education modules that were built on, and delivered through, RxHealth's DNP. Patients received notification messages and educational content from the day of appointment scheduling to one‐day post‐procedure completion.

Method: Survey/Qualitative

Classification of Research: Quality Improvement

Result: Of the 291 patients receiving the DNP in site A, 136 clicked ≥1 message received from the DNP (actively engaged). Patients actively engaged had a 14.8% (n = 136) incomplete rate vs. 40% (n = 155) for patients not actively engaged (P < 0.01). In site B, 81/192 patients were actively engaged and those had a 20.9% incomplete rate (n = 81) vs. 46% (n = 111or those not actively engaged (P < 0.01). Combined data showed the active engagement with the DNP led to a 59.7% related decrease in incomplete procedure rate. About 92% of patients who filled an online survey “strongly agreed” and “agreed” that the DNP was helpful.

Conclusions: The two‐center study showed that patients actively engaged with DNP have about half the rate of incomplete procedures than those not actively engaged. Furthermore, non‐engagement with DNP for colonoscopy is a strong predictor of overall incompletion rate and can be used to identify patients at the highest risk for non‐completion. This high‐risk group can now be included in the priority call list for endoscopy center staff so as to leverage multimodal interventions (including calls, letters, digital messages through a patient portal) that can further decrease no‐shows and incomplete rate for endoscopy procedures.

49. Modeling HealthCARE: Capacity, Access and Resource Evaluation

Deborah Ercolini, Matthew Henchey, and Mary Lowe Mayhugh

The MITRE Corporation

As virtual models of care continue to expand, organizations need tools to determine how to best implement these new models of care, such as telehealth, that support patient centered healthcare. Furthermore, organizations are driven to maximize access to patient care across a multitude of demographics and to incorporate the growing demand for the limited availability of specialty care by weaving telehealth into the traditional healthcare journey. Modeling and simulation (M&S) provide a method for organizations to use data driven decision making tools to evaluate new or expanded care models and identify the impacts to patients, providers and other resources across the health care system. We propose a framework, Modeling HealthCARE: Capacity, Access, and Resource Evaluation, to aid resource planning and inform decision makers of healthcare systems on the impacts of new healthcare delivery services, including telehealth. The model focuses on the integration of services, such as telehealth, to help healthcare systems evaluate supply and demand requirements and incorporates multiple data sources into a dashboard and visualization tool for impactful decision making. Our framework uses both agent based and discrete‐event modeling to replace assumptions about behavior with observed societal behavior, to include social determinants which influence adoption of new healthcare services.

Method: Implementation Science

Classification of Research: Measurement Frameworks & Tools

Result: We applied the framework to a regional Veterans Health Administration (VHA) system to evaluate treating Veterans diagnosed with Post Traumatic Stress Disorder using telehealth. This research was conducted independently from the VHA using publicly available data. Using a hybrid M&S approach, we model the veteran as an agent adopting telehealth, following a discrete‐event process through the existing and new healthcare journey. Scenarios were evaluated around resource capacity and telehealth adoption, showing impacts to appointment wait times and travel distances for the patients. Dashboards visualizations were designed to help identify gaps in personnel, resources, technology, and policies to inform decision makers.

Conclusions: This research provides a framework for evaluating new healthcare delivery options in a simulated environment, allowing organizations to make informed decisions around future delivery of telehealth services without having to go through costly physical pilots. We have shown how integration of new telehealth services, patient behaviors, and changes in capacity impact patient centric performance measures. As research continues, these models may be extended to provide a platform for assessing other new delivery services, which should help maximize access to care and quality of care. In addition, access to real‐world data could provide more detailed insights into capacity, access and adoption.

50. Diabetes Education through Shared Medical Appointment utilizing Digital Health

Leighanne Hustak, Matthew Faiman, and Marianne Sumego

Cleveland Clinic

Diabetes along with obesity hypertension and hyperlipidemia are chronic diseases currently increasing the already overburdened Unites States Healthcare System. These chronic diseases account for 81% of hospitalizations, 76% of medical provider visits and 91% of all prescriptions filled (Smith, A., 2016). Health Care organizations are seeking novel approaches to chronic disease management and adapting digital health platforms and smart devices. These digital health tools include applications, mobile technology, wearable devices, e‐coaching, telemedicine and personalization of medications. Often the barriers to care in chronic disease include limited access, inadequately trained caregivers, travel costs, lack of real‐time data and the limitations of office time for education. The novel use of digital health seeks to optimize care and address many of the barriers to quality care. The new requirements of health care regulation in the United States at the same time challenges health care organizations to improve population management of these chronic diseases or face reduced reimbursement for services.

Method: Descriptive

Classification of Research: Clinical Outcomes

Result: The Distance Health team and Shared Medical Appointment (SMA) team at the Cleveland Clinic in an effort to positively affect population management of Type 2 Diabetics across practices, developed a work effort for the delivery of group education and medication management currently used in clinics as an SMA into a virtual format. The results are in process and the goal is to implement real‐time referral in the office visit with the primary care provider to a group education virtual meeting translating traditional SMA into virtual delivery and enhancing quality chronic disease management with improved AIC and satisfaction scores.

Conclusions: Currently participants are being identified through care gap registries, internal medicine provider referrals for DM education and patient self‐referral. A family nurse practitioner with CDE certification follows and engages the patients monthly in SMA education sessions through a digital platforms over a 4 month period using change in A1C and qualitative patient satisfaction data to review SMA engagement, improved disease state and platform use ability. The process from patient identification, digital consent, appointment invitation, patient engagement, provider satisfaction and A1C improvement will be used to describe the usability and improved chronic disease management outcomes.

51. Effect of Virtual Wait Time upon Patient No Show Rates in Direct‐To‐Consumer Acute Care Telehealth

Sean Britton and Anthony Consolazio

UHS Hospitals and Medical Group

An integrated healthcare system opened a direct‐to‐consumer acute care telehealth service (Virtual Walk‐In) staffed exclusively by its own provider group. The Virtual Walk‐In is open daily from 08:00 am to 7:00 pm and patients have the option of completing the intake interview during off‐hours to be seen immediately after the clinic opens the following morning. Some patients complete the intake interview and then no show for their video visits despite contact efforts by the provider. The intent of this analysis is to determine the influence wait times may have upon the likelihood of a patient to present for his/her/hir video visit. This is an observational analysis of patients who completed an intake interview for service at the UHS Virtual Walk‐In beginning in August and ending in November of 2019. Wait time is defined as the time from when the patient completed the intake assessment until when the provider initiated the video visit, with the caveat that intake interviews completed during off‐hours would count the wait time as beginning at the opening of the clinic for the day. Mean wait times were calculated for patients who did and did not show and a t‐test analysis of independent means was performed.

Method: Observational

Classification of Research – Other: Patient Behavior

Result: 264 patients completed an intake interview. 198 patients (75.00%) presented for their video visit and were either definitively diagnosed or referred for in‐person follow‐up care by the provider. 66 patients (25.00%) did not show for their video visit. Mean wait time for patients who presented for video visits was 23.78 minutes and mean wait time for patients who were a no show for their video visits was 70.25 minutes. A t‐test analysis showed no significance between the group of patients who presented for their video visit and the group of patients that was categorized as no show (p‐value = 0.125).

Conclusions: Although the t‐test showed insufficient statistical significance to reject the null hypothesis and make a conclusion about patient behavior based upon wait times, we view the difference of means between the two groups (46.47 minutes) to be significant from a patient experience and clinic operations perspective. Completion of the intake assessment during open or closed hours of the clinic was not factored and is a limitation within this analysis. Further analysis of the effect of off‐hours intake completion should be performed.

52. It's not millienals, it's moms. Optimizing direct to consumer telehealth for pediatric populations

Kimberly Cronsell and Jennifer Ruschman

Children's Wisconsin

A survey by the employee benefits research institute in 2017 found that 40% of millennials state that telehealth is an “extremely” or “very important” option when it comes to their healthcare. An estimate of 64% of parents “have used” or “plan to use” telemedicine within the next year for their child according to a 2017 survey. Despite this evidence, adoption of new care solutions nationally has been low and pediatric solutions criticized for quality, specifically over prescribing of antibiotics3 However, when CNBC spoke with executives of four leading telehealth companies, all said they targeted moms and relied on them to utilize service with families. Pediatric organizations already know what these executives were sharing, it's not just millennials that desire telehealth solutions, but specifically moms, and especially working moms, are seeking out telehealth solutions for addressing their families' healthcare needs. Learn how two pediatric organizations are implementing telehealth solutions designed for and targeting the healthcare decision maker‐ the moms.

Method: Survey/Qualitative

Classification of Research: Patient Experience

Result: Children's Wisconsin focused on reducing low acuity ED visits. Through detailed patient journey mapping unique pain points in seeking acute care were identified, and tailored solutions such as online scheduling, patient portal expansion, and direct to consumer online video visits were developed and deployed. Cincinnati Children's focused on developing a new care model and building a scalable platform when they evaluated direct to consumer telehealth. Cincinnati will focus on the teams and resources involved in driving forward a controversial strategy to deliver pediatric direct to consumer telehealth services as well as needed resources for implementation.

Conclusions: Children's Wisconsin online urgent care is an integrated on demand platform with a hybrid staffing model that helped align clinical guideline adherence, decrease variation across the system, improve provider efficiency, and convert hesitant clinicians to digital champions. Cincinnati's direct to consumer program demonstrates that strategic planning and team based implementation can lead to high family satisfaction in new models of care. Both teams have demonstrated the need for successful partnership in a complex tertiary children's hospital organization, and expanded the ability to connect with patient families when and how they need us.

53. Introducing Virtual Check‐Ins to an Existing Virtual Care Platform

Meghan Glanville and Tasia Walsh

Medical University of South Carolina

In 2019, Medicare started reimbursing for virtual check‐ins which allows for more efficient and cost effective communication between patients and their providers. Described as a patient initiated “brief communication technology‐based service” that allows for communication through phone, video or other communication methods, the virtual check‐in allows for patients to initiate a conversation with their provider for a medical condition. The Medical University of South Carolina rolled out diabetes and hypertension virtual check‐ins in an existing virtual care platform to allow patients to use technology they are already familiar with initiate a check‐in. Patients had an additional module on the platform enabled which allowed them to select their condition and answer a questionnaire to be sent the provider. Once the provider reviewed the submission, they followed up with the patient through various modalities.

Method: Implementation Science

Classification of Research: Clinical Effectiveness

Result: This process utilizes an existing platform to allow patients to initiate a check‐in with their provider for diabetes or hypertension. This allows providers to free up clinic time from med checks that can be done virtually, reduces unnecessary travel and missed work for patients, and creates a method of accounting and reimbursing provider for their time with patients. Virtual Check‐Ins will be expanded to additional services and condition to increase patient utilization.

Conclusions: While the service is only currently reimbursable for Medicare patients, this standardized virtual check‐in process is available to all established patients as a method for providers to follow‐up with their patients outside of the clinic. This increases utilization of an existing virtual care platform to enable patients to feel more in control of their healthcare and with their providers at their fingertips.

54. Key attributes for implementing teleophthalmology to improve diabetic retinopathy surveillance

Jesica Basant, Rajeev Ramchandran, Adam Ross‐Hirsch, Reza Yousefi‐Nooraie, Ann Dozier, and Sule Yilmaz

Flaum Eye Institute, Department of Ophthalmology, University of Rochester Medical Center

Teleophthalmology substantially increases annual retinal screening rates for vision threatening diabetic retinopathy in low income primary care clinics. However, implementation of such programs is challenging. This study explores the implementation of a teleophthalmology program for diabetic retinopathy and visual acuity surveillance in three urban, low income, largely minority serving primary care clinics in Upstate New York.

Method: Implementation Science

Classification of Research: Access to Care

Result: A total of 13 staff agreed to participate in semi‐structured interviews (4 MDs, 3 NPs, 4 RNs, 1 PharmD, 1 Admin. Nursing staff focused on buy‐in and motivation, workload and effective staffing, space and time management, and camera‐interface training. Higher‐level providers concentrated on patient education and follow‐up, financial and operational sustainability, clinic champions, maintenance of inter‐departmental relationships, and opportunities for resident physician learning and engagement. All primary care staff expressed great value for having vision and camera‐based retina evaluation in primary care for patients with diabetes.

Conclusions: This study provides further understanding of the barriers and facilitators of implementing teleophthalmology within primary care clinics for diabetic retinal examinations in underserved populations. Ensuring standardization of processes, workflows, and staff knowledge; having adequate staff, space, and time; consistently well‐functioning technology with robust customer support; financial viability and understanding financial impacts on patients; and continuous coordination between primary care and ophthalmology to improve follow‐up are needed for ideal implementation. Knowledge gained from the clinics currently utilizing these programs can help inform the development of implementation strategy toolkits to evaluate and improve the uptake of similar programs across the United States.

55. When Telehealth Can't Get There: Patients Willing to Access Telehealth But Not the Clinic

Sean Britton and Anthony Consolazio

UHS Hospitals and Medical Group

An integrated healthcare system opened a direct‐to‐consumer acute care telehealth service staffed exclusively by its own provider group. The Virtual Walk‐In is open eleven hours daily and patients may complete the intake interview during off‐hours to be seen immediately after the clinic opens the following morning. The asynchronous intake interview will triage higher acuity patients for whom a virtual visit isn't appropriate to seek in‐person care. Some patients have a synchronous video visit during clinic hours which results in the provider referring the patient to follow‐up with in‐person care. The intent of this analysis is to determine the rate of follow‐up to in‐person care as evidenced by referral by either algorithmic triage or provider referral. This is an observational analysis of patients who selected a condition to be seen for and then were referred to in‐person care or had a video visit with a provider resulting in in‐person referral at the UHS Virtual Walk‐In beginning in August and ending in November of 2019. A proportion of patients who followed up with same day in‐person care were calculated and a z‐test to compare two proportions was performed.

Method: Observational

Classification of Research: Access to Care

Result: 68 asynchronous encounters resulted in referral to in‐person care by algorithmic triage and 27 (39.71%) patients followed up in‐person. 38 patients had video visits with a provider resulting in referral for in‐person care and seven (18.42%) patients followed up in‐person. There was an observable difference of 21.29% and the z‐test to compare 2 proportions was statistically significant (z = 2.3, p < 0.024, two‐tailed). A difference between the 2 groups was expected since they represent different levels of patient acuity. We also lack data as to whether a patient followed up in‐person at a facility outside of our healthcare system.

Conclusions: A minority of patients who initially attempted to access medical care via the Virtual Walk‐In did follow the advice for in‐person follow‐up whether given by asynchronous algorithmic triage or via synchronous video interaction with a provider. The majority of patient's directed for in‐person follow‐up did not do so. This would imply there is a significant population of individuals who want medical treatment for an acute condition and yet for unknown reasons, likely related to the social determinants of health, they do not follow‐up on medical advice to obtain such care when it is required to be in‐person.

56. Telemedicine with an Ultra‐widefield Camera for Diabetic Retinopathy Screening

Patrick Le, Michelle Nguyen, Josh Tanner, Janet Yan, Thomas Miller, and Seema Garg

University of North Carolina School of Medicine

Diabetic retinopathy (DR) is one of the leading causes of blindness in the world, and the incidence continues to increase. Screening and early treatment are more important than ever. In this study, we investigate and report the utility of implementing a telemedicine model using an ultra‐widefield device (UWFD) for DR screening.

Method: Observational

Classification of Research: Clinical Effectiveness

Result: A total of 625 patients (1,247 eyes) with type II diabetes completed screening. Twelve patients were excluded from analysis due to poor image quality, resulting in a 1.9% failure rate. The study model added an average of 4 minutes and 44.7 seconds to the overall visit, expanded the DR screening coverage by a linear diameter of 6 miles, and improved the primary care clinic's DR screening compliance rate from 53.0% to 72.8%. The incidence of DR was 18.71% and peripheral findings were found in 50.9% of patients with DR.

Conclusions: The proposed model of telemedicine, in conjunction with an UWFD, had low failure rates, increased the coverage of diabetic screening, improved clinic compliance of screenings, and mildly increased visit times.

57. Deploying Telemedicine for Emergency Treatment of Opioid Use Disorder

Tiffany Champagne‐Langabeer and James Langabeer

University of Texas Health Science Center, School of Biomedical Informatics

Telemedicine technology has had significant diffusion in recent years for both inpatient and outpatient care, but in addiction medicine it has not been widely adopted. As mortality rates involving opioids continue to rise, as well as addiction rates overall, it is essential to identify novel ways to initiate and maintain treatment. Technology solutions are being proposed, including both mobile health and telemedicine; however, there have been few published manuscripts demonstrating feasibility of the conceptual framework. For individuals with opioid use disorder (OUD) and those who have experienced non‐fatal overdoses, it is critical to initiate rapid treatment while individuals are still experiencing withdrawal symptoms and before they re‐engage in prior behaviors. Barriers to entering recovery programs exist for patients, including lack of familiarity with where and how to enter treatment, a scarcity of addiction medicine providers‐ particularly in some geographic areas, cost of care, and limited transportation. Telemedicine could offer one solution. In this study, we present the framework for a telemedicine solution developed in Houston Texas.

Method – Other: Interventional, Single‐arm study

Classification of Research: Clinical Effectiveness

Result: The objective of this research was to demonstrate a pilot telemedicine program for initiating medical treatment of patients with opioid use disorder (OUD). From December 2018‐February 2019, ten patients were seen through synchronous video conferencing. Physicians assessed patients for opioid dependence, completed a brief history and medical exam, and ensured patients were in at least moderate withdrawal. Patients were followed through daily and weekly telephonic follow‐ups to measure retention in treatment 30‐days post‐telemedicine encounter. After 30 days, nine patients (90%) were still actively involved in treatment.

Conclusions: Feasibility of a model for utilizing telemedicine for the emergency initiation of opioid agonist treatment was demonstrated in this pilot project. We found that telemedicine can be used to connect board‐certified physicians with individuals at a distance, reducing geographic, transportation, and cost barriers to care. Initial outcomes, 30 days after telemedicine encounter, are promising and suggest that immediate treatment of individuals remotely through telemedicine can positively initiate and retain individuals into treatment for the short‐term. There are policy implications for telemedicine opioid treatment which need to be explored, including prescribing barriers and financial reimbursement for these services.

58. Self‐Assessment Tool to Improve Neo TeleHealth Resuscitation

Stephen Minton, Shaun Odell, and Taunya Cook

Intermountain Healthcare

Intermountain Healthcare has been using Neo TeleHealth Resuscitation for 5 years connecting neonatologists with clinicians/staff who infrequently perform neonatal resuscitation. This includes using TeleHealth for training, case reviews, bimonthly simulations including neonatologists, resuscitations, post resuscitation huddles, staff testing, and most recently for a Post Code Review Assessment Tool (PCRAT) looking at 5 areas of focus.

Each Item scores 1 point if done correctly for a maximum score of 10.

PCRAT is self‐completed immediately post resuscitation and submitted with the Neo Resuscitation Record Sheets (NRRS). An Administrative Resuscitation Review Team (Neonatologists, APPs, RTs, and clinical staff from both the referring and referral hospitals) reviews each NRRS line by line and PCRAT then completes their own PCRAT. Feedback is given to staff not only on the actual resuscitation record but on both PCRATs.

Method: Implementation Science

Classification of Research: Clinical Effectiveness

Result: The PCRAT has been used since August 2018 at American Fork Hospital (AFH), a 10‐bed, level II, Special Care Nursery with approximately 3,000 births per year. During that time, 101 resuscitations received simultaneous PCRATs. The staff self‐PCRAT scores were almost always higher with 89.7% either 8, 9 or 10 while administrative PCRAT scores were only 48.5% for 8, 9, and 10. The major discrepancy between the self and administrative PCRAT scores were APGAR discrepancies from written documentation, undocumented cord gases, initial oxygen use, and pulse oximeter not placed within 30 seconds. Since we have introduced PCRAT, errors have been decreasing.

Conclusions: Having staff score themselves (PCRAT) immediately after a resuscitation, has increased awareness and correct use of major NRP points resulting in improved neonatal resuscitation outcomes. Using telehealth at all steps ‐ teaching, simulation with neonatology, actual neonatal resuscitation participation, post resuscitation huddles, Neo Resuscitation Record Sheets review (NRRS), PCRAT reviews, and specific case reviews has improved quality of care, reduced transfers, reduced costs and markedly improved physician/staff satisfaction.

59. Avera eCARE: Medical Student Education in Telemedicine

Kelly Rhone, Jenny Lindgren, Luke Mack, Lindsay Spencer, and Susan Anderson

Avera eCARE

With the past and current disparities in rural health, telemedicine has evolved into the forefront of filling the gaps in provider coverage. Medical students have reported feeling unprepared to use telemedicine and uninformed about laws regarding telemedicine usage following graduation. However, they also reported that telemedicine training is relevant and important for their future work (Waseh, Dicker, 2019). With the next generation of physicians unprepared to utilize the growing field of telemedicine, the disparities observed in rural America will continue to grow, and steps toward educating upcoming providers will play a vital role in preventing this growth.

Method: Survey/Qualitative

Classification of Research: Clinician Experience

Result: A Chi‐square analysis was used to look for associations' pre/post by year. Both years show an increase in favorable responses for questions to telemedicine training and education. For analyses by topic area, we created clusters of questions to build scores. T‐tests were used to look for associations' pre‐/post‐ by year. The analysis resulted in 3 topic areas to build scores. Both years show a significant increase in the Rating of Overall Knowledge and Interest in Curriculum and Utilization. There is no significant difference in Willingness to Practice.

Conclusions: Results show notable differences in how students perceive and understand telemedicine after structured exposure to telemedicine services. Further, this study demonstrates students' need for and interest in more telemedicine training opportunities in their curriculum. Results demonstrate the willingness to practice in rural settings in having no significant difference, thus this may be an area to focus attention on in future studies. Specifically how telemedicine training is perceived by those more willing to work in rural communities.

60. But What about the Medical Students? Developing a Medical School Telemedicine Curriculum

Peter Greenwald, Mary Mulcare, Rahul Sharma, Neel Naik, Yoon Kang, Kriti Gogia, Kaitlin Schullstrom, and Sunday Clark

Weill Cornell Medicine

Telemedicine and virtual care have become part of mainstream medical practice. It is increasingly clear that the ability to provide high quality care via video is a skill that will be required of most physicians. The skills required for effective evaluation and communication during a video encounter differ from skills required at the bedside, yet to our knowledge, few educational modules with content focusing on training medical students in telemedicine have been developed. Medical students, regardless of anticipated specialty, stand to benefit from early exposure to and education in this new modality of clinical care delivery. Our objective was to develop, implement, and assess a training module designed to teach medical students techniques to deliver professional, effective and compassionate care during a telemedicine encounter. We created a simulation‐based curriculum using advocacy/inquiry methodology debriefing with video‐based encounters focused on “web‐side manner” as a critical corollary to traditional bedside manner. We recorded simulated cases for each student with standardized patients, guided debriefs using advocacy/inquiry methodology, and incorporated table‐top exercises to teach advanced communication and examination skills in telemedicine.

Method: Survey/Qualitative

Classification of Research: Survey/Qualitative, Clinician Experience

Result: Thirty of 31 students, taking the course, consented to have their data and responses analyzed for research purposes. Of these 30, 29 (97%) reported being enthusiastic about the curriculum. After participation in the curriculum, 60% believed that telemedicine had the potential to become part of their practice in the future and 93% reported improvement in their own comfort and ability to conduct video‐based patient encounters. 100% of students reported that simulation was an effective mechanism for delivery of the educational material.

Conclusions: Teaching telemedicine using simulation and advocacy/inquiry debriefing was well received by students and provided early exposure to aspects of medical practice that will be pertinent to their future practice. Objective assessment tools of students completing telemedicine encounters need to be created, as well as research showing student improvement in managing clinical encounters via telemedicine after taking this course, potentially as measured in objective structured clinical examination format.

61. Kiosk use in Direct to Consumer Telemedicine: assessment using National Quality Forum Guidelines

Peter Greenwald, Mary Mulcare, Rahul Sharma, Kriti Gogia, Sunday Clark, Hanson Hsu, and Sapir Nachum

Weill Cornell Medicine

Evaluation of direct‐to‐consumer telemedicine programs has focused on care delivery via patient's personal electronic devices (App). Telemedicine kiosks for the delivery of virtual urgent care services have not been systematically described. In order to better understand how kiosks are being used by our patients, we compared patients who accessed telemedicine urgent care from kiosks to those who used an App to access the same telemedicine service.

We conducted a retrospective review of adult patients using either a pharmacy‐based kiosk or tan App for direct‐to‐consumer telemedicine urgent care evaluation by our Emergency Medicine doctors. Automated reports were reviewed to assess patient and visit characteristics. Medical records were reviewed to determine diagnosis codes, follow‐up recommendations, and whether the patient was traveling. Results were interpreted using the National Quality Forum framework for telemedicine service evaluation in the domains of access, experience, and effectiveness. Comparisons were made using Chi‐square test, Student's t‐test, and Wilcoxon rank‐sum tests, as appropriate.

Method: Observational

Classification of Research: Access to Care

Result: During 1 year 1,996 patients were seen via DTC urgent care; 238 (12%) initiated from kiosks (0.10 visits per kiosk day). Compared to App patients, kiosk patients were older (age 38 ± 13 vs 35 ± 11; P < 0.001), and more likely male (52% vs 39%; P < 0.001). The most common diagnoses among both groups were respiratory (35% vs 21%; P < 0.001), genitourinary (14% vs 13%; P = 0.69), and other symptomatology (13% vs 17%; P = 0.18). Kiosk patients were more likely travelers (25% vs 3%; P < 0.001), and had less technical difficulty (10% vs 19%; P = 0.003). Referral for urgent evaluation was low in both groups (10% vs 16% P = 0.02).

Conclusions: Kiosks may increase access to care for the traveling public and may provide a better technical experience than a patient's own device. Diagnosis related groups were generally similar between groups. The majority of patients were not referred for urgent in‐person follow up, suggesting that effective care was provided for both types of visits. Despite their potential advantages, kiosk visits accounted for a minority of overall visits for our direct‐to‐consumer telemedicine service line, and daily use of each kiosk was low.

62. Patients' Perspective of Specialty Telemedicine Consults at a Federally Qualified Health Center

Neal Sikka, Waala Alsufyani, Jeff Jacob, Nicole Ehrhardt, Susie Lew, Guenevere Burke, and Lisa Martin George

Washington University

Telemedicine is becoming more common in both urban and rural settings especially for mental health and primary care. However, there is limited data on specialty care and acceptability of specialist care through telemedicine. We conducted a study which evaluated the use of telemedicine for cardiology, nephrology and endocrinology consultation in a community clinic and evaluated patient's perception of specialty telemedicine care. After Institutional Review Board approval, patients were enrolled at a local community Federally Qualified Health Center (FQHC) in the District of Columbia. Nephrology, cardiology, and endocrinology specialty care were provided via telemedicine (real‐time secure audio‐video consultation facilitated by a technician at the distance site) for a designated clinical condition. Patients' perception of telehealth for specialty care was assessed by an 11 question survey that was completed by each patient immediately following their visit.

Method: Survey/Qualitative

Classification of Research: Patient Experience

Result: Satisfaction surveys were completed from 112 patients. Results show 92.8% were satisfied with the telemedicine system; 60.7% felt their health improved after using telemedicine; 77.4% followed the doctor's advice better since working with the telemedicine system; 86.5% were more likely to use this technology in the future; 93.7% were able to explain their medical problems well enough during the video visit.

Conclusions: Patients' perception of telemedicine for specialty care in the community was positive, saved them time, and an overwhelming majority of patients strongly agreed that the telemedicine experience is convenient for them and they would choose to receive their care via telemedicine even after the project concludes. Increased access to specialty care via telemedicine is needed and further research needs to be done in order to champion availability to the community given high patient satisfaction with care.

63. Utilizing Patient Geographic Data and Geomapping to Plan Outreach Locations for Pediatric Genetic Services

Elizabeth Null and Omar Abdul‐Rahman

Department of Pediatrics, University of Nebraska Medical Center

Distance to pediatric subspecialty care can be a significant barrier for families. Telemedicine and outreach clinics are means of providing additional access to care, but methods used to select sites are often arbitrary. Geocoding and geospatial analytical software are novel tools for providing a data‐driven method in determining sites for establishing clinics. Community Needs Index (CNI) and average travel distance (ATD) are objective measures that quantify a community's needs based on zip code and the distance a patient must travel to access healthcare, respectively. CNI is calculated from aggregated data from the following domains: income, culture/language, education, housing status, and insurance coverage for a given zip code (scored 1 through 5, with 5 being highest need). Using geospatial mapping techniques, CNI, and ATD, patient access to one pediatric subspecialty, clinical genetics, can be assessed that can aid in the strategic allocation of healthcare resources. A retrospective chart review was conducted over an 18‐month period for all patients seen at all five of Nebraska's genetics outreach clinics (excluding Omaha), both in‐person and telemedicine to calculate ATD, CNI, and trends in heat maps of patient locations.

Method: Secondary Data Analysis

Classification of Research: Access to Care

Result: 465 encounters were reviewed and home addresses originated from 5 states (NE, SD, WY, KS, TX), 104 towns and 118 unique ZIP codes. The ATD (in miles) for patients were: Gordon 60; North Platte 47; Kearney 34; Scottsbluff 21; and Lincoln 36. A majority of patients came from Lincoln (52/465, 11.2%) and the zip code of 69361 in Scottsbluff (46/465, 9.9%) the latter having a CNI score of 4. The average ATD for all patients attending all clinics was 34 miles. The average CNI score was 3.56 for the top 10 zip codes represented.

Conclusions: Pediatric patients traveled 34 miles (on average) to access in‐person or telemedicine genetics outreach clinics in Nebraska from Jan 2018 – June 2019. Based on the aggregated heat map, patients from northeastern Nebraska travel to attend the Lincoln clinic. Norfolk is the most population dense city in that quadrant. Plans are to establish a telehealth clinic, assess genetic referral patterns of pediatricians in Norfolk, and conduct ongoing assessments for additional sites. Retrospective chart review and geospatial analysis of geocoded patient addresses is a methodology that can be applied for the strategic implementation of healthcare resources in genetics and other fields.

64. Improving the Telehealth Educational Curriculum ‐ A Delphi

Kristi Sidel

Avera eCARE

As national interest in telehealth grows, there is an opportunity to support advancement in the field among practitioners, administrators and future providers. Telehealth education programs focus on positively impacting the lives and health of individuals and communities by training providers and administrators to deliver the gold standards of care when utilizing telehealth. A telehealth education program for professionals can help inform providers and administrators in understanding and addressing barriers as well as ensure that they are set up to provide high‐quality, compliant telehealth services. Setting standards for practice will also support a more consistent regulatory landscape, allowing practitioners to practice across states, insurance companies, and facilities without navigating complex reimbursement and regulatory differences. To date, there has been no gold standard for telehealth education.

Method: Survey/Qualitative

Classification of Research: Clinician Experience

Result: Final editing of curriculum is still underway, and expected to be completed in March 2020.

Conclusions: Using a modified Delphi method, a gold standard, evidence based curriculum in telehealth was created taking into account the experience of a panel of experts in the field. Use of this curriculum can aid in the standardization and improvement of the telehealth practice.

65. Capturing the Attitude Delta: Using the “Before and After Approach for Patients Choosing Telehealth

Debbi Lindgren‐Clendenen, Sharaz Mohammed, Marc Newell, and Ross Garberich

Minneapolis Heart Institute/Allina Health

The Mpls Heart Institute (MHI) determined the need to provide telehealth visits for our cardiology population in outstate Minnesota. Patients endured an average of four‐week wait time to be seen for a cardiology consultation. Our plan was to offer these visits via real‐time, face to face interactive video visits using the local clinic as a partner. Because most of our cardiology patients are older, there was concern by leadership whether or not these patients would be willing to use telehealth technology to see the cardiologist for their consultation. In order to support our belief that our cardiology patients would be willing to be seen via telehealth technology, we determined that we would survey the patients prior to the telehealth visit and then after the visit to see if there was any change in attitude towards using telehealth services by our cardiology patients. A 2014 literature review validated leadership concern regarding the use and acceptance of technology in providing services to older cardiology patients. We also noted minimal literature that measured a baseline attitude towards telehealth visits and then use of a post visit survey to capture patient attitude changes.

Method: Survey/Qualitative

Classification of Research: Patient Experience

Result: We analyzed the first 366 respondents (septuagenarians) of the initial Teleheart Program patient survey. Willingness to refer to our program was 98%, with 99% stating using telehealth made it easier for them to see the cardiologist. A majority (96%) of respondents believed that the visit was as least as good as an in‐person visit.

Conclusions: Electronic surveys are a feasible and reliable tool for assessing Telehealth patient satisfaction, even in elderly patients. Adaptation of this survey from paper to electronic means did not impact its effectiveness. Average time to administer it is 3 min per patient for each survey. Five years later, patient satisfaction metrics still result in the 90^th percentile despite double digit growth each year.

66. Emergency Telemedicine Systems for Disability Group Homes: Moving the Emergency Department to the Be

Renoj Varughese, David Ellis, Bonnie Sloma, John Carnevale, Anthony Billittier, James Collins

University at Buffalo, State University of New York

Individuals residing in Disability Group Homes represent a special medical needs population that can benefit significantly from the application of a multi‐faceted emergency telemedicine care system. Many of the components that make emergency departments able to deal with both life‐threatening and less serious urgent problems can be applied with this system. Key components of the program include a Nurse Call Center which functions similar to the Charge/Triage Nurse in the ER managing initial assessments and flow of patients in partnership with the Emergency Telemedicine Provider (ETP). A Certified Home Health Agency Nurse with a skillset of an ER nurse/technician is available for home response to facilitate tele‐presentation, patient nursing assessment, medication administration, and simple procedures including G‐tube re‐insertion and unblocking, splinting of injured extremities and simple wound care. A Rapid Response Protocol allows the emergency telemedicine team to make a timely decision on whether to call 911 or allow a full telemedicine evaluation. ETPs consisted of advance practice providers (PA/NP) with Emergency Medicine Attending Physician back‐up utilizing voice and video. The end result is to essentially move the emergency department functionality to the patient bedside obviating patient plus staff transport to the emergency department for evaluation of patients.

Method: Observational

Classification of Research: Access to Care

Result: 722 calls for service were received over 8 months of system operation from over 200 group homes. Each group home contained an average of 5 residents per group home each staffed with a caretaker. Technologies used included telemedicine software (MobileTelemed 2.0™) running on tablets with remote EKG capability. Mobile radiology services included portable x‐rays and doppler ultrasound. Weekly meetings provided an opportunity for patient care review and the ongoing refinement of emergency patient care systems and protocols. A majority of the patients were successfully managed at the facility with the remaining patients requiring transport to ERs and Urgent Centers.

Conclusions: Emergency medicine systems with established roles personnel and patient care guidelines are essential to the provision of high quality emergency medical care. Patients residing in Disability Group Homes can be effectively managed using digital network videoconferencing, bedside EKG and mobile radiology services in an emergency telemedicine system staffed with Call Center RNs, Home Response Nurses and Emergency Telemedicine APP Providers with Emergency Physician backup. This emergency medicine patient care model mirrors that of an emergency department system and effectively moves the system of care provided in the ER to the bedside in a Disability Group Home.

67. Telehealth in Community Health Centers

William England

Office for the Advancement of Telehealth, Federal Office of Rural Health Policy, Health Resources and Services Administration

Although telehealth has increasingly been used by federally qualified health centers (FQHCs) since they were first designated as eligible rural telehealth originating sites for Medicare in the Benefits Improvement and Protection Act of 2000 (BIPA), the increase has been slow. The most commonly cited reason for slow growth has been limited reimbursement, but there are also other issues. In 2016, an effort to assess the use of telehealth in health centers was launched by the Health Resources and Services Administration (HRSA). Telehealth questions were incorporated into the Uniform Data System (UDS) annual report for health centers. Similar questions were asked in 2017 and more questions were added in 2018.

Method: Survey/Qualitative

Classification of Research: Regulatory & Policy Research

Result: In 2018, the nation's nearly 1,400 health centers with 12,000 service delivery sites and 235,000 health care professionals served nearly 28 million people. Over 43% of those health centers reported using telehealth and over 10% reported they were considering or in the process of implementing telehealth services. In addition to reporting on how they used telehealth, health centers not using telehealth reported why they were not using telehealth.

Conclusions: This presentation will analyze HRSA's 3 years of data on telehealth in health centers, both how it being used and why it is not being used. The analysis will discuss reimbursement, health center and population demographics, broadband and technology issues as well as other observations concerning the use of telehealth services available from the public UDS dataset. Additional telehealth information expected from future UDS data reports will be discussed.

68. Everything We Needed to Learn In School ‐ Based Telehealth

William England

Office for the Advancement of Telehealth, Federal Office of Rural Health Policy, Health Resources and Services Administration

Over 100 years ago, we learned that nurses in schools could improve education simply by applying public health concepts of hygiene and health education to the student population, to control the spread of disease and reduce absenteeism. As the role of school nurses expanded from health education to health screening and immunizations, referrals, and primary care, we learned the value of more advanced clinical services in schools. However, many rural schools cannot not afford the cost of such school‐based health care staff or services. Telehealth is an efficient mechanism to provide such services and in recent years, there has been a significant increase in the adoption of telehealth by schools. In an effort to evaluate the impact of telehealth in schools, the Office for the Advancement of Telehealth (OAT) in the Office of Rural Health Policy (FORHP) in the Health Resources and Services Administration (HRSA) launched a study of school telehealth, looking at key student health issues including asthma, behavioral health, diabetes, healthy weight, and oral health.

Method: Survey/Qualitative

Classification of Research: Measurement Frameworks & Tools

Result: To perform this study, applicants in the School‐Based Telehealth Network Grant Program collected information on patient specific measures developed by OAT's Rural Telehealth Research Center specifically for school‐based telehealth. The data collected to date in this ongoing study covers 12,000 telehealth encounters with 3,300 students in 179 schools. Qualitative interviews have been used to assess the varied approaches taken by schools and their telehealth service providers.

Conclusions: This presentation will share details of the 29 measures and 40 data elements collected and will summarize quantitative and qualitative results to date from the project. Results will include assessment of travel miles saved by parents or students in visiting off campus providers along with impacts on student absenteeism. Challenges of the project and best practices for sustainability of school‐based telehealth projects will be discussed.

69. Reimbursement of Specialty Telehealth Claims for Medicaid and Private Insurance

Neal Sikka and Guenevere Burke

George Washington University

Background: One barrier to establishing new telehealth programs is concern about reimbursement. We established a specialty teleconsult program providing live audio video visits at a federally qualified health center in Washington, DC. The program was funded through a grant from the CareFirst Foundation designed to reimburse for consultations not paid for by the patient's insurer or for self‐pay patients.

Methods: We examined practice claim history for 54 unique patients enrolled in the telehealth program for which we had complete claims history between March 2017 and Sept 2019. A total of 150 visits were identified in which the claim specified the location of service as ‘Telehealth’ or included the modifier code GT. We then excluded claims that were noted as self‐pay, Medicare, or Medicare Advantage. We calculated percent of claims with reimbursement from the payer.

Method: Cost Analysis

Classification of Research: Cost Analyses

Result: Results: Of 150 total claims, 93 met inclusion criteria. These included 15 private insurance claims, 27 Medicaid and 51 Medicaid Managed Care Organization (MCO) claims. Claims were paid most often for patients enrolled in Medicaid (100% claims paid, n = 27), followed by Medicaid MCOs (94% claims paid, n = 48), then private insurance (73% claims paid, n = 11).

Conclusions: Washington DC has parity laws in place for Medicaid and private insurance, requiring payment for video telehealth services if the same services would be covered in‐person. Thus, DC Medicaid covers telemedicine services with limitations under a published guidance document. Medicaid MCOs can determine their own coverage for telehealth beyond the Medicaid guidance. Based on the high rate of successful claims reimbursement (73–100%), clinicians in Washington DC and in areas with similar parity legislation should not consider reimbursement a significant barrier to telehealth program development for patients with private insurance or Medicaid as long as originating site restrictions are considered.

70. Cloud‐Based Improvement of Patient‐Provider Communications Regarding Free‐Clinic Retinal Screening

Andrés Eduardo Guerrero Criado, Gnanashree Dharmarpandi, and Kelly Mitchell

Texas Tech University Health Sciences Center

TTUHSC LIFC offers healthcare to uninsured patients in rural Texas, their services include retinal screening for diabetic retinopathy and glaucoma. Storage and transfer of screenings were originally conducted through the exchange of an encrypted USB between students and physicians. While safe, this model proved inefficient, with issues in timely communication and chain of custody concerns. Lapses in communication led to patients being lost to follow‐up, which is already a ubiquitous problem in free clinics (Buys et al.). Without a centralized system summarizing patient encounters, retention to resolution (R2R) was uncommon. To remedy this, we analyzed the positive and negative predictors of patient retention in our previous system to create an all‐encompassing Cloud‐Based Electronic Medical Record (CB‐EMR) system through the TTUHSC BOX platform. Our Route Cause Analysis addressed the following faults in the previous system of care to be addressed in our novel CB‐EMR.

Method: Implementation Science

Classification of Research: Clinical Effectiveness

Result: Analysis of 530 patients encounters one year after application of CB‐EMR:

Conclusions: Transitioning to CB‐EMR has streamlined access to information, allowing consistent care for the uninsured ophthalmologic patient. It is now over 100% more likely to reach R2R, and retention across the board has sky‐rocketed. This shift has allowed us to consolidate patient information and follow their conditions in real‐time with our attendings. Thus determining what additional interventions are necessary to improve patient outcomes safely and efficiently across specialties. This system allows us to triage our patients based on diagnostic needs to be seen by an attending Ophthalmologist rather than order of arrival. Our future directions point towards personalizing patient educational encounters.

71. Co‐Developing a Remote Monitoring Platform for Heart Failure Management: Factors to Consider for Effective Clinical Integration

Ankit Bhatia¹, Brett Ramsey¹, Thomas Maddox¹, Andre Dias², Sonia Koesterer²

¹BJC/Washington University School of Medicine Healthcare Innovation Lab and ²Myia Health

Heart failure (HF) remains one of the leading causes of morbidity and mortality in the US, and a significant driver of hospital readmissions. Conventional approaches to outpatient HF management have generally been shown to be ineffective in mitigating readmissions. Accordingly, health systems are increasingly adopting digital noninvasive remote patient monitoring (RPM) platform solutions for outpatient HF management. While evidence regarding noninvasive RPM in HF management is conflicting, institution‐specific data have revealed more promising outcomes when RPM is paired with robust clinical integration that is informed by end‐user clinicians. The BJC/Washington University School of Medicine Healthcare Innovation Lab partnered with Myia Health, an external RPM vendor, to codevelop an institution‐specific RPM platform for HF management. The initial goal of the partnership was to identify institutional factors vital to the successful clinical integration of HF RPM.

Method: Implementation Science

Methods: We conducted a systematic review of existing RPM technologies, alongside an institutional needs‐assessment via interviews with HF stakeholders. Stakeholders included HF cardiologists, nurses, HF patients, information technology personnel, and clinical administrators. Stakeholder insights were collected using structured interviews and contextualized inquiries. Findings were jointly synthesized to identify important factors for institutional RPM design and clinical integration model.

Classification of Research: Clinician Experience

Result: 50+ HF clinical stakeholder interviews were conducted over 3 months. Identified priorities vital for successful RPM integration included: (1) maximizing patient engagement and adherence through passive approaches for patient data acquisition and integration, (2) optimizing platform interface for ease‐of‐use by clinicians‐ including integration into existing clinical workflows and electronic health record, and (3) ensuring the platform's clinical care requirements match care team capabilities (e.g. triaging clinical alerts such that existing clinical resources could have capacity to respond accordingly).

Conclusions: Institutional clinical stakeholder interviews uncovered novel and important aspects needed for the successful clinical implementation of RPM for HF management. Incorporation of these insights into RPM programs may help to improve efficacy in HF management and optimize clinical workflow. We accordingly codesigned a HF RPM platform and clinical interface based on these insights and conducting a pilot to evaluate its clinical utility and efficacy.

72. Designing An Effective Clinical Interface for Remote Patient Monitoring for Heart Failure Management

Ankit Bhatia¹, Brett Ramsey¹, Thomas Maddox¹, Andre Dias², Sonia Koesterer²

¹BJC/Washington University School of Medicine Healthcare Innovation Lab and ²Myia Health

Heart failure (HF) remains a leading driver of care utilization, and health systems are increasingly embracing noninvasive remote patient monitoring (RPM) as an approach for high‐risk outpatient HF management. While commercial offerings exist to deliver RPM data, most remain primarily “one‐size fits all” data‐reporting mechanisms, and lack the tools that enable clinicians to directly and efficiently manage patients based on these new data streams. The BJC/Washington University School of Medicine Healthcare Innovation Lab partnered with Myia Health, an external RPM company, to codesign an institution‐specific RPM platform and clinical interface for HF management, with the goal of promoting clinician and patient engagement.

Method: Implementation Science

Method: To identify RPM platform features required for enhanced clinical utility, we conducted a series of interviews with HF clinicians, including contextual inquiries and co‐design workshops. Multiple rounds of clinician feedback were then used to inform an iterative design and agile development process. Using prototypes which successively increased in fidelity, from paper to click‐through digital renderings to usability testing, a primary user interface was designed; this was then leveraged to refine features into a launch‐ready clinical interface.

Classification of Research: Clinician Experience

Result: Notable features of the codesigned RPM platform included a bias to ambient passive sensors for data collection to minimize patient burden and maximize adherence‐ which included the use of a ballistocardiogram, an under‐bed sensor to assess ∼100 raw and derived data features including heart rate and respiratory rate. The clinical interface centered around patient lists, with customizable prioritization based on patient‐specific thresholds (e.g. vital signs outside of pre‐defined range) that trigger actions for clinicians. Alerts could also be “opted‐in” to allow for rapid HF medication up‐titration. The interface also included data visualization workspace that juxtaposes longitudinal health data (vital signs, symptoms) with event data to aid in clinical management. Preliminary testing indicated high fidelity of patient engagement/data transmission (>80% after 270 days). The resultant RPM clinical interface well‐addressed the needs voiced by HF clinicians, with subsequent demonstrated clinician engagement and perceived utility.

Conclusions: Working through an iterative codesign process, we codesigned an RPM platform and clinical interface that addressed the needs and eased integration into existing workflow for HF clinicians. Clinical teams may consider similar approach when establishing RPM programs.

73. Telehealth Research Dissemination Forum: Update on ATA's PCORI Sponsored Research Forum

Elizabeth Krupinski¹, Sabrina Smith², David McSwain³, Curtis Lowery⁴, and Thomas Wilson⁵

¹Emory University, ²American Telemedicine Association, ³Medical University of South Carolina, ⁴University of Arkansas Medical Sciences, and ⁵Trajectory^® Healthcare

Method: Survey/Qualitative

Method – Other: Change Management

Classification of Research: Measurement Frameworks & Tools

Classification of Research – Other: Research

Result: The perceptions about the status of telehealth research changed prior to vs after the forum and being exposed to the different types of PCORI projects and evidence from the research results. Scores below 2 (average) indicate areas where telemedicine can improve and those above indicate positive findings. Although none of the questions resulted in statistically significant changes, there were a number of topics that did show changes in the 10–20% range (in both positive and negative directions). Of interest was the fact that the results of the survey distributed to the more general ATA membership showed very similar results. The brainstorming groups focused on 4 PCORI project themes: Greater Engagement of American Indian Populations, Internet‐Based Tools to Improve Physical Therapy to Captivate and Engage Patients, Diabetic Neuropathy and Tele‐Home Monitoring for Underserved Communities. There were a number of common themes that emerged from the interactive discussions: the need to identify, gather and engage stakeholders; the need to educate, train and engage all participants in the effort; the need to make tools and platforms easy to use and cost effective; the need to incentivize users; the need to empower team members; the need to identify and create the value proposition for telehealth; and the need for metrics.

Conclusions: If we can maintain momentum, through revolutionary thinking and taking lessons learned back to our home institution, to make telehealth mainstream, deploy (universal) healthcare equitably if we concentrate on some key points: recognize the telehealth is healthcare not telecare; leverage each other's talents and collaborate on research projects; have the right people in place; embrace the past and accomplishments to date (even if imperfect); and remove regulatory barriers.