key: cord-0048689-zx0dmnac authors: Barbash, Ian J.; Sackrowitz, Rachel E.; Gajic, Ognjen; Dempsey, Timothy M.; Bell, Sarah; Millerman, Konstantin; Weir, David C.; Caples, Sean M. title: Rapidly Deploying Critical Care Telemedicine Across States and Health Systems During the Covid-19 Pandemic date: 2020-07-22 journal: NEJM Catal Innov Care Deliv DOI: 10.1056/cat.20.0301 sha: cdeec304237d77e5d23d268daedf008099b9ae43 doc_id: 48689 cord_uid: zx0dmnac A team of clinicians from three organizations — UPMC, Mayo Clinic, and NewYork-Presbyterian —rapidly developed a teleICU model to address an immediate need by drawing on standards of implementation science while relying on nimble actions to quickly establish relationships and lines of communication and facilitate timely and effective deployment of a model that could be re-used for both urgent and long-term needs. » Fewer barriers to licensure and credentialing would reduce administrative costs and streamline telemedicine implementation across state and institutional boundaries. » Institutions and practices should consider telemedicine as an efficient tool in the team-based management of population health. The novel coronavirus disease-2019 (Covid-19) produced a surge of critically ill patients without precedent in modern health care, requiring hospitals to address multiple components of critical care capacity.1 -3 At baseline, there already existed significant disparities in the distribution of access to critical care expertise across and within regions of the United States.4 , 5 Compounded by the pandemic, the sheer number of patients far exceeded intensive care unit (ICU) beds available in some hospitals. This surge required hospitals to expand capacity and implement tiered staffing models under which physicians without formal critical care training were often caring for severely ill patients, many of whom were mechanically ventilated. 6 Even in hospitals that did not exceed physical ICU bed capacity, the overall number and acuity of patients increased due to Covid-19, which created strain in pre-existing staffing models and among teams that were not accustomed to caring for that combination of patient volume and acuity. It was impractical, if not impossible, to address these access and capacity challenges by deploying more critical care physicians on the ground, so innovative solutions were needed to expand the reach of physicians with critical care expertise. Our goal was to rapidly deploy critical care telemedicine teams (teleICU) to increase access to critical care expertise during the Covid-19 pandemic. Our plan was to integrate teams from UPMC (based in Pittsburgh) and Mayo Clinic (based in Rochester, Minn.) to assist hospitals associated with NewYork-Presbyterian in New York City, which were experiencing surge levels of Covid-19 patients in April 2020 and had reached out for assistance via national professional societies at the end of March. We also were attentive to the need to ensure continued coverage for local patients within the UPMC and Mayo health systems in need of critical care for Covid-19 or other causes. To implement the teleICU program, we addressed several issues. Due in part to the compressed implementation timeframe (about one week), we often worked in parallel rather than in a series of steps. The first and most crucial step was to identify one or more clinical leaders to serve as champions at each institution and facilitate telemedicine implementation." " Establish Implementation Champions The first and most crucial step was to identify one or more clinical leaders to serve as champions at each institution and facilitate telemedicine implementation. These champions had explicit buy-in from executive leaders to implement teleICU support. The champions communicated with the teleICU development team weekly and as needed by email, phone, and videoconference to identify the optimal care models, develop workflows, and provide opportunity to troubleshoot any challenges during implementation. The implementation champions collaboratively identified several categories of need for teleICU support among four NewYork-Presbyterian facilities: NYP-Lawrence, NYP-Weill Cornell, NYP-Lower Manhattan, and NYP-Queens; this collaboration drove our subsequent approach to workflows and staffing. First, there was a need to have teleICU staff available for around-the-clock, on-demand consultation for acute questions. Second, there was a need for dedicated rounding time during which a teleICU intensivist could support and advise New York-based bedside care teams on patient management; this need was particularly acute in New York City, where, due to high patient volume associated with the pandemic, critical care treatment was frequently provided by clinicians trained in other disciplines. After understanding these needs, we designed complementary staffing models, relying heavily on staff volunteerism due to the need for rapid deployment. To address on-demand coverage, UPMC recruited a group of intensivists to staff an operations center 24 hours per day, 7 days per week. UMPC was able to leverage a large academic faculty, ordinarily dedicated to a mix of scholarly and clinical activities, to support the teleICU. The Mayo Clinic relied on a teleICU infrastructure that had been deployed in 2013. To address the need for rounding support, intensivists were recruited from UPMC and Mayo to conduct morning and evening rounding with specific counterpart teams and units at the New York-Presbyterian hospitals. Rounding was augmented and streamlined with the use of the CERTAIN checklist.7 Physicians champions on the ground in the NewYork-Presbyterian partner hospitals were vital to develop, test, and implement workflows that would facilitate clinical care. Working together, more than 20 physicians provided continuous coverage in a hybrid rounding/on-demand model. As we developed these staffing models, we collaborated with information technology (IT) experts locally and at the New York hospitals to address several elements of the teleICU. First, we needed IT assistance to build a web-based workflow management application to record incoming calls, triage the consultations, and facilitate completion by the teleICU intensivist. Second, teleICU physicians had to become familiar with multiple new electronic medical records (EMRs) in order to review data and, in some cases, document recommendations across hospitals and health systems. Finally, we worked with IT across health systems to identify shared platforms by which to establish HIPAA-compliant video connections for rounding support. We also worked with our colleagues in operations and credentialing to obtain disaster privileges across hospitals and health systems. While operating across health systems and state borders typically presents a significant barrier to telemedicine implementation (requiring multiple state licenses and an extensive credentialing process), the Covid-19 pandemic state and federal disaster declarations significantly reduced the administrative burden.8 , 9 As a consequence, we were able to decrease to a matter of days a credentialing process that often takes months. Within each health system, the existence of a shared medical staff office streamlined the process. We encountered several challenges during our implementation of teleICU services. First, the approach during a pandemic differed dramatically from the typical, multi-phase telemedicine implementation schedule. Usually, such a process would occur over several months and involve multiple face-to-face organizational meetings, which would serve to build the trust and relationships that promote effective implementation and adoption of the telemedicine service.10 In the absence of this traditional implementation model, utilization of the services was variable across sites. Scheduling a routine point of contact for the clinical teams and maintaining regular contact among implementation champions served to promote adoption of the telemedicine services in this context. Second, the dynamic nature of the pandemic demanded flexibility in implementation and service models, which required clinicians to adapt to shifting roles. TeleICU teams were required to navigate new and unfamiliar EHR systems, and the rapidly deployed technical system and care models were streamlined versions of the proactive monitoring approach with which physicians working in a more comprehensive teleICU model are accustomed. For some hospitals, the teleICU service was most valuable in support of general critical care issues; in others, there was greater need for Covid-19-specific issues; and, in yet others, the need was greatest in intermediate care units rather than ICUs. Adapting to these shifting roles required flexibility and regular communication among the members of the TeleICU teams. Finally, implementing ICU telemedicine entails a number of implicit and explicit costs. Indeed, high-intensity ICU telemedicine models can cost in excess of $50,000 annually per covered bed.11 These costs support staff payroll, the administrative overhead required for licensure and credentialing, and telemedicine equipment and technology. During our pandemic response, we were able to minimize these costs in several ways. First, the temporary suspension of regulatory barriers substantially reduced administrative time. Second, we leveraged a variety of technologies already in place for audiovisual connection and remote access to the EHR, reducing the need for additional resources. Finally, our staffing models all relied on physicians volunteering their time, effort, and expertise in support of the various telemedicine service models. Our programs were effective in the short term on a volunteer basis, and they wound down as the surge abated and local needs changed. Sustaining similar,light-touch teleICU models over a longer timeframe would require methods to address relevant costs, which could be manageable for hospitals with more limited resources. systems to build the model, and aclinical team to deliver care." The rapid deployment of critical care telemedicine required the coordinated efforts of a development team of individuals from different professions and disciplines across hospitals and health systems to build the model, and aclinical team to deliver care. The development team comprised physicians (including intensivists ICU directors and hospitalists) to assess clinical considerations and staffing models, at least one of whom was designated as a champion at each site and provided personal interaction with colleagues during roll-out. An ICU nurse with experience as a unit director provided insights on triage considerations. From the information technology department, telemedicine project managers provided guidance on workflows, hardware, and software platforms for establishing audiovisual connections. The team also included members responsible for operations: a medical staff office representative and executive administrators at each site for credentialing issues, an associate counsel for malpractice considerations, and teleICU medical directors for onboarding of those who would joining the Critical Care Telemedicine team. Collectively, the implementation process involved well over 100 person-hours in the week leading up to go-live. The Critical Care Telemedicine clinical team itself included more than 20 practicing critical care physicians and 15 ICU nurses helping to triage the work. Between the groups from UPMC Health System and the Mayo Clinic, more than 20 physicians completed more than 400 encounters with patients in New York City hospitals over a period of four weeks from April 6 to May 1. This work included a mix of once-and twice-daily ICU rounds, support for an intermediate care unit, and on-demand consultation around the clock; these services were delivered seven days per week, with one intensivist dedicated to each location at a time. Due to the compressed implementation timeframe and lack of data integration across multiple EMRs, we were unable to quantify the impact of the teleICU on specific processes or patient outcomes. However, the stories that arose yield a complementary illustration of the impact of the teleICU program. For example, in one hospital we were emergently asked to assess a mechanically ventilated patient who was chemically paralyzed and in the prone position. An alarm on the ventilator indicated that the patient was not receiving adequate tidal volumes. The teleICU physician assessed the patient and ventilator via video, examined a chest radiograph, and rapidly " identified a dislodged endotracheal tube. The patient was urgently reintubated by the hospital's inhouse airway team; the patient was eventually discharged to rehab. Without access to the teleICU, this patient's outcome might have been different. We identified several key steps in an expedited change management approach, including 1) identifying the need, 2) developing the staffing model, and 3) developing an implementation plan (Table 1) . We learned several important lessons during the process of establishing a teleICU to meet clinical care demands in the midst of a pandemic. First, while the timeframe was necessarily compressed, many lessons from implementation science in "normal" times still applied.10 In particular, it was especially important -and difficult given the context -to rapidly establish relationships across sites, with open lines of communication to facilitate effective implementation and service delivery. Second, while technology is clearly important in establishing a telemedicine program, the optimal role of technology should be a solution that facilitates streamlined, effective connections between clinical teams, rather than the technology itself serving as a central focus of the telemedicine program. We found that in many interactions, we were able to support clinical teams on the ground using audio connection via the telephone, with or without data review in the EMR. Many teleICU programs use a comprehensive, technologically intensive approach to patient monitoring and proactive management. Our experience suggests that even a light-touch model can be useful, particularly when the demand is acute, the implementation timeframe is short, local IT and financial resources are limited, or the primary goal is to facilitate access to intensivists. It is possible to imagine future conditions that would facilitate the flexible deployment of telemedicine across states and systems to mitigate disparities in access to critical care expertise (for example, in rural or resource-limited hospitals), regardless of pandemic conditions." " Finally, our experience illustrates what we can achieve when priorities and incentives align. It is possible to imagine future conditions that would facilitate the flexible deployment of telemedicine across states and systems to mitigate disparities in access to critical care expertise (for example, in rural or resource-limited hospitals), regardless of pandemic conditions. For example, fewer barriers to licensure and credentialing would reduce administrative costs and streamline telemedicine implementation across state and institutional boundaries. Also, as we move away from fee-forservice reimbursement, institutions and practices may be more likely to approach telemedicine as an efficient tool in the team-based management of population health.12 Our experiences with this rapid telemedicine deployment illustrate the power of the sense of duty and service that call many of us to fulfill the health care mission. By capitalizing on the urgency of the clinical need, the intrinsic motivation of the individuals involved, and the complementary skills of the team members, we were able to rapidly establish a telemedicine platform at scale across multiple health systems and states. Elements of the teleICU model -catalyzed by the Covid-19 pandemic -will remain in use at UPMC for within-system support of critical care needs at some of its 40 hospitals moving forward. The lessons we learned are applicable to ongoing capacity issues related to Covid-19, future pandemics, other health care emergencies, and in efforts to mitigate disparities in access to critical care expertise. Clinical Characteristics of Coronavirus Disease 2019 in China Adaptations and Lessons in the Province of Bergamo Locally Informed Simulation to Predict Hospital Capacity Needs During the COVID-19 Pandemic Critical care delivery in the United States: distribution of services and compliance with Leapfrog recommendations United States Resource Availability for COVID-19 Innovative ICU Physician Care Models: Covid-19 Pandemic at NewYork-Presbyterian Checklist for Early Recognition and Treatment of Acute Illness (CERTAIN): evolution of a content management system for point-of-care clinical decision support Cuomo Declares State of Emergency 3548 -CARES Act Determinants of intensive care unit telemedicine effectiveness. An ethnographic study Clinical and Economic Outcomes of Telemedicine Programs in the Intensive Care Unit: A Systematic Review and Meta-Analysis What 21st Century Health Care Should Learn from 20th Century Business We would like to thank the many nurses, physicians, and operational and administrative staff at UPMC, the Mayo Clinic, and NewYork-Presbyterian, without whom we could not have established these teleICU programs. And we would like to thank the innumerable physicians, nurses, and other staff in New York City who worked long, arduous hours to deliver care to patients with Covid-19.