key: cord-0762777-hd4xpig1
authors: Ganesh, Ravindra; Salonen, Bradley R.; Bhuiyan, M Nadir; Bierle, Dennis M.; Moehnke, Darcie; Haddad, Tufia C.; Tande, Aaron J.; Wilson, John; Hurt, Ryan T.
title: Managing patients in the COVID-19 pandemic: A Virtual Multidisciplinary Approach
date: 2020-12-15
journal: Mayo Clin Proc Innov Qual Outcomes
DOI: 10.1016/j.mayocpiqo.2020.12.003
sha: c9372674f68a87bdab6151559ed8e100d1885c55
doc_id: 762777
cord_uid: hd4xpig1

Objective To study the impact of a 60-day pilot of an innovative virtual care model using General Internal Medicine physicians and nurses to rapidly respond to over 1200 COVID-19 positive nasopharyngeal PCR tests. Patients and Methods The current study was approved by the Mayo Clinic COVID-19 Research Committee and the Mayo Clinic Institutional Review Board. The data for all SARS-CoV-2 positive patients treated by our team were entered into a prospectively maintained internal REDCap database. We retrospectively searched this database for the first 60 days of our program(March 23, 2020 until May 22, 2020). The data included basic deidentified demographics, symptoms at intake into the program, date of symptom onset, risk factors, location, and outcomes including hospitalization, ICU admission, and death. Results Patients were contacted on average 6.3 hours after their results became available. There was a total of 138 ED visits. Of these, 40% were admitted to the hospital, with 36% of those admitted requiring ICU level of care. Of the 849 patients in this sample, there were only 2 deaths (0.23%) at 60 days. Conclusion Our innovative multidisciplinary COVID team provided excellent clinical care for patients with COVID with a very low mortality rate compared to the national average. While data is not available on a national scale for time to contact patient, our team was able to contact patients within the established recommendation for contact within 48 hours of testing is optimal.

In late December 2019, a cluster of viral pneumonia cases emerged in Wuhan, China, with these cases subsequently being linked to common exposure at a live animal market. Soon after, human to human transmission was reported, most notably in healthcare workers. 1, 2 Samples from respiratory epithelial cells of affected patients revealed a novel beta coronavirus which was named SARS-CoV-2 due to its similar initial presentation to the outbreak of severe acute respiratory syndrome coronavirus (SARS-CoV) in 2002, and the clinical syndrome was named coronavirus disease 2019 (COVID-19) by the World Health Organization (WHO). 3 COVID-19 rapidly spread across the globe within the next few months, resulting in the current pandemic being declared on March 11, 2019 by the WHO, and since has created unprecedented strain on the healthcare infrastructure of multiple countries. The majority of infected patients (around 80%) present with mild symptoms, including fever, cough, myalgias, and diarrhea, or are asymptomatic. [4] [5] [6] Approximately 20% of patients may have more severe presentations including dyspnea and acute hypoxic respiratory failure secondary to the insidious onset of acute respiratory distress syndrome (ARDS) and may need hospitalization. 4 This has led to critical shortages of intensive care unit (ICU) beds, ventilators, and personal protective equipment (PPE). 7 While COVID-19 is certainly capable of causing severe disease in young, healthy patients, the bulk of those adversely affected belong to high-risk categories. Early clinical data has demonstrated that older age, smoking history, and history of cardiac disease predict worse outcomes. 6, 7 While 80% of individuals do not require hospitalization, many will seek emergency department (ED) care due to a high level of global concern. Due to the risk of transmission to healthcare workers and necessary resource management, it is imperative to preserve inpatient care for those that meet criteria and to minimize ED utilization. The low-risk and stable, high-risk COVID-19 population may be managed in the outpatient setting, but the necessary infrastructure may become overwhelmed when faced with pandemic patient loads. In the setting of a critical shortage of hospital-based resources, including PPE, beds, and ventilators, a telemedicine or virtual initiative is aptly positioned to intervene and address several of these challenges. 8 A centralized physician team is preferable to having hundreds of community based physicians managing these patients in order to provide standardized, high quality care that complies with the rapidly changing best practice recommendations.

The use of physician-initiated, non-face-to-face patient encounters has the potential to effectively triage patients at increased risk of severe disease from the majority who will recover without intervention. 9 The use of telephone support and non-invasive telemonitoring devices to follow patients' symptom and vital sign trends with centralized nursing may help earlier identify and effectively triage patients in need of escalating care to the ED or inpatient settings while reassuring those who can safely continue in-home monitoring. 10 This risk stratification should ideally occur as soon as possible after the positive SARS-CoV-2 testing returns.

In addition to early risk stratification of SARS-CoV-2 positive patients and subsequent outpatient monitoring, patient quarantine and contact tracing should start as soon as possible. Virtual patient education on the importance of quarantine should be discussed with every SARS-CoV-2 positive individual and will minimize the number of secondary infections. All of these essential tasks including risk stratification, monitoring, tracing, and quarantine education can be done effectively using telemedicine. We describe our comprehensive, multidisciplinary telehealth surveillance program to address the population-based health concerns of the COVID-19 pandemic. We also report the first 60 days of this program in which we treated over 1200 SARS-CoV-2 positive patients.

The current study was approved by the Mayo Clinic COVID-19 Research Committee and the Mayo Clinic Institutional Review Board. The data for all SARS-CoV-2 positive patients treated by our team were entered into a prospectively maintained internal REDCap (Research Electronic Data Capture) database. Some patients refused being followed by the CFCT and we did not follow patients already in clinical facilities such as group homes and skilled nursing facilities. We retrospectively searched this database for 60 days starting with the first patient seen virtually by our team on March 23, 2020 until May 22, 2020. The data included basic deidentified demographics, symptoms at intake into the program, date of symptom onset, risk factors, and location.

At our institution, the process of identifying, triaging, and subsequent follow up care of SARS-CoV-2 positive patients was initially handled by our colleagues in Infectious Diseases (ID). However, given the limited ID personnel resources and expanding duties during the COVID-19 pandemic, a joint relationship with General Internal Medicine (GIM) was established. In this model, the GIM division assumed the outpatient management of patients with positive polymerase chain reaction (PCR) results for SARS-CoV-2, with ongoing collaboration with ID. In turn, ID physicians focused more on inpatient consults on COVID-19 infections, institutional COVID-19 protocol development and infection control measures. The overarching goal was to develop a centralized team that would rapidly respond to SARS-CoV-2 positive tests from our institution's various testing centers that were resulted in the electronic health record (EHR; Epic Verona, WI, USA) and coordinate the level of outpatient care.

Prior to forming the GIM COVID-19 Frontline Care Team (CFCT) the specific workflow was developed between the ID and GIM divisions, identifying a number of different clinical scenarios in which the positive cases would need to be assessed and followed. The GIM CFCT team would respond to positive SARS-CoV-2 results from all the testing sites at Mayo Clinic Rochester as well as the affiliated Mayo Clinic Health System community-based practices serving Southern Minnesota, Northern Iowa, and Western Wisconsin. After testing was performed, patients were instructed to isolate at home until contacted by a community medicine nurse (if test was negative) or CFCT physician (if test was positive). In addition, patients who had been dismissed from the hospital due to COVID-19 related illness were followed by the CFCT. Identified COVID-19 disease risk factors were used to further stratify patients into low, high, and very high-risk groups (Figure 1) . The very high-risk group was eventually combined with the high-risk group. These risk factors were based on available literature which have identified that those patients who are male, over the age of 65, have structural heart or lung disease, are immunocompromised, have a malignancy, have end stage renal disease, are current smokers, or have diabetes mellitus have a worse prognosis in COVID-19. 4, 5, [11] [12] [13] The initial physician phone call to the patient for risk stratification would occur within hours of the results being available in the EHR and nursing follow-up notification would occur within 24 hours. Patients with severe symptoms on the initial phone call were instructed to go to the Emergency Department (ED) for further evaluation.

We initially recruited a small force of GIM physicians and GIM nurses to accomplish the initial goals of the CFCT. The physicians rapidly became knowledgeable about COVID-19 clinical presentations and complications and created education modules to enable further expedient recruitment and coverage of physicians as needed. Most of the GIM physicians had recent or current hospital experience so they were familiar with triage and management of the decompensating patient. The GIM nurses were trained by our home parenteral and enteral (HPEN) and complex care nursing teams who have experience working with the virtual care of complex patients. Our nursing teams made follow up phone calls on days 2, 7 and 14 to assess for changes in symptoms. In addition, they developed and delivered education modules centered on the principles of the importance of self-quarantine and social distancing. They also staffed a nurse line for patients to call in with worsening symptoms. If symptoms warranted escalation, the nursing pool routed these concerns to the physicians on call. This number was covered after hours by an on-call physician who would provide triage for patient concerns. This ensured the patient had a single contact number which was staffed 24 hours a day to call as it pertained to COVID-19 related symptoms.

In those patients deemed high-risk on initial assessment or following a COVID-19 hospital discharge, remote monitoring systems (RMS) were delivered to the home. Patient symptoms and vital signs were reported at least twice daily and continuously monitored by a Remote Patient Monitoring (RPM) team of nursecoordinators and support staff. The COVID-19 RPM program comprised of two care pathways and RMS: 1) a complex care monitoring plan, by which a cellular-enabled tablet collected patient-reported symptom assessments and connected with a Bluetooth enabled scale, blood pressure monitor, pulse oximeter, and thermometer, and 2) an interactive care plan (ICP) by which patients were given a pulse oximeter and thermometer and utilized the Mayo mobile app (EHR portal) on their smartphone or tablet to self-report symptoms, temperature, and oxygen saturation. In both care pathways, any abnormal symptoms or deviation of vital signs beyond predefined parameters established by a multidisciplinary team (CFCT, ID, RPM) created an alert to the RPM dashboards and nursing team. The RPM nurse would then contact the patient, evaluate their symptoms, troubleshoot any technological errors, and escalate as necessary to the CFCT physician for further assessment with the provision of direct escalation to a higher acuity center in cases of emergency.

The CFCT worked with other COVID-19 stakeholders including Occupational Health, the Olmsted County Public Health Department (OCPHD) department, and Mayo Clinic Infection Prevention and Control. As the primary team responding to COVID-19 diagnoses, the CFCT is able to identify high risk contacts and emerging trends in the community. We modified our process to collect data surrounding possible contacts, recent work exposure, and home address to identify any potential clusters of cases which were then communicated to these stakeholders. We additionally established daily communication strategies

All data is presented as mean ± standard deviation for normally distributed data and median for nonparametric data. All statistical analysis and graphical figures was performed using R (version 3.6.3).

A total of 1,291 patients had a positive SARS-CoV-2 PCR test result and were referred for management by CFCT. Research authorization consent was provided by 849 patients who were included in the study. Twenty-one patients were unable to be contacted by the CFCT team physician and were thus excluded from final analysis. Table 1 shows demographic characteristics of our patients. The median age of patients was 40 years (IQR 29-54) and there was a slight female majority (n=444; 54%) with most patients residing in the state of Minnesota (n=793; 96%). The majority of patients were determined to be low-risk for COVID-19 related complications (n= 653: 79%) with 60 (7%) medium risk, and 107 (13%) high-risk (please see Figure 2 ). Table 2 shows risk factors for severe COVID infection with most patients having no risk factors (61%), and diabetes (11%), asthma (7.7%), age > 65 years (7.5%), and current smoker (6.0%) being the most commonly identified . Reported symptoms are shown in Table 3 , with the four most commonlyreported sympoms being cough (n=392; 47%), headache (n=315; 38%), myalgia (n=311; 38%), and fever (n=267; 32%).

The majority of lab results for PCR testing were available within 24 hours of sample collection. The overall time from positive test result to first contact for the cohort when data was available (n=767) was 6.3 hours with 75% being less than 12 hours. For those patients with limited English proficiency, the average time from the test positive being resulted in the EHR until first contact with the patient (positive to first contact) was significantly increased compared to those whose primary language was English (p=<0.001). Patient Outcomes 115 patients (8.9%) were sent to the ED based on the CFCT physician recommendations for a total of 138 ED visits (Table 4) . Of these, 40% were admitted to the hospital, with 36% of those admitted requiring ICU level of care (Table 5) . For those patients requiring ICU level of care, 9 (45%) were initially triaged as high/medium risk and 11 (55%) were initially triaged as low risk. Of the 849 patients in this sample, there were only 2 deaths (0.23%) at 60 days.

The main objective of the CFCT model was to create a centralized virtual multidisciplinary group that would rapidly respond to COVID-19 positive tests from a large healthcare organization caring for patients across three states. 14 This novel approach helped decrease the time the positive test result was delivered in the EHR to the time the patient was first contacted, and risk stratification occurred. This coordination was performed using an entirely virtual platform which greatly decreased potential exposure to health care workers and conserved precious PPE supplies.

The RPM team combined with the CFCT physician and nurses provided a centralized method of managing COVID-19 positive patients. This model fosters collaboration between several physician teams and utilizes nurses as physician extenders by the establishment of well-defined decision trees and treatment algorithms. Early detection of patient decompensation by using the remote monitoring technology allowed for establishment of pathways for direct admission of patients to the COVID-19 inpatient service or the ED depending on stability. Further evaluation will be needed to determine if this virtual care model and early detection of adverse trends reduced the need for hospital observation or admission, shortened length of stay, or reduced the need for ICU level care. The ability to rapidly contact and risk stratify patients and then deliver these specifics to OCPHD can decrease the lag time thereby improving the efficacy of contact tracing.

Countries such as Iceland and South Korea have avoided the large healthcare burden associated with COVID-19. 15 South Korea had cases as early as early January following which a concerted nationwide containment strategy was implemented. 15 The cornerstones of this containment strategy were aggressive widespread testing, prompt contact tracing, and quarantine of positive individuals as well as those that have been exposed to SARS-CoV-2. During the South Korea COVID-19 peak in late February they were performing over 10,000 PCR tests a day. By June 1, 2020 they had reduced reported daily cases to 35. Rapid expansion of testing is the first critical step and should be offered initially to those with high suspicion of COVID-19 and later to asymptomatic individuals as testing capacity increases. A critical component to any screening program is the actions taken immediately following the return of a positive test result. A recent article emphasized that a patient testing positive for COVID-19 needs to be immediately notified, educated, isolated and their contacts identified. 15 The importance of a rapid response to the positive test is highlighted by modeling data that suggests that if contact tracing is to be effective patients and contacts should be quarantined within 24 hours of testing. 15, 16 This may explain the successful response to COVID-19 in countries with centralized health care systems and robust public health infrastructure. The United States healthcare system is in dire need of multidisciplinary, centralized care team models that are capable of rapidly responding to positive test results.

The authors recognize limitations of the present study. This was a single health care system retrospective review of our experience with COVID-19 positive patients. Despite the good clinical outcomes we reported, the lack of a comparison group makes it difficult to quantitatively assess the magnitude of our team's impact. The centralized model may be challenging to reproduce in health systems are less tightly integrated or rely on outside laboratories for testing. Because our testing was performed in a high-volume, centralized location with rapid turnaround time, and results were made available in a unified EHR, our CFCT team was able to rapidly respond (<6 hour) thus potentially improving outcomes. There is a need for prospective cohort trials to confirm the findings of our retrospective study. 

Guide to Understanding the 2019 Novel Coronavirus

COVID-19: A 2020 update

A Novel Coronavirus from Patients with Pneumonia in China

Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study. The Lancet

Risk factors of critical & mortal COVID-19 cases: A systematic literature review and meta-analysis

Clinical Characteristics of Coronavirus Disease 2019 in China

Critical Supply Shortages -The Need for Ventilators and Personal Protective Equipment during the Covid-19 Pandemic

Covid-19: a remote assessment in primary care

Virtually Perfect? Telemedicine for Covid-19

Remote Monitoring of Patients With Heart Failure: An Overview of Systematic Reviews

Systematic Review and Meta-Analysis of Predictive Symptoms and Comorbidities for Severe COVID-19 Infection. Public and Global Health

Smoking Is Associated With COVID-19 Progression: A Meta-analysis

Does comorbidity increase the risk of patients with COVID-19: evidence from meta-analysis

Telemedicine Consultations and Follow-up of Patients With COVID-19

From Mitigation to Containment of the COVID-19 Pandemic: Putting the SARS-CoV-2 Genie Back in the Bottle

Quantifying SARS-CoV-2 transmission suggests epidemic control with digital contact tracing

J o u r n a l P r e -p r o o f