Training and Incorporating Students in SARS-CoV-2 Case Investigations and Contact Tracing

Abstract

Objectives

In June 2020, Arizona had the fastest-growing number of cases of coronavirus disease 2019 (COVID-19) worldwide. As part of the growing public health response, the University of Arizona Student Aid for Field Epidemiology Response (SAFER) team was able to modify and increase case investigation efforts to assist local health departments. We outline the recommended logistical and management steps to include students in a public health response of this scope.

Methods

From April 1 through September 1, 2020, the SAFER team identified key components of a successful student team response: volunteer training, management that allows more senior students to manage newer students, adoption of case-management software, and use of an online survey platform for students to conduct interviews consistently and allow for data quality control and management.

Results

From April 1 through September 1, 2020, SAFER worked with 3 local health departments to complete 1910 COVID-19 case investigations through a virtual call center. A total of 233 volunteers and 46 hourly student workers and staff members were involved. As of September 2020, students were completing >150 interviews per week, including contact-tracing efforts.

Practice Implications

Developing relationships between applied public health and academic programs can relieve the burden of low-risk, high-volume case investigations at local and state health departments. Furthermore, by establishing a virtual call center, health sciences faculty and students can volunteer remotely during a pandemic with no additional risk of infection.

Using public health and other health science students in response efforts and as surge capacity for state and local health departments provides an excellent training opportunity for students before they enter the workforce.^1-3 Students are also a resource for an overworked and overwhelmed public health workforce, providing a reprieve during the coronavirus disease 2019 (COVID-19) pandemic. One challenge is how to best use these students in existing health department structures or by allowing them to operate in parallel with current efforts while merging their data into reporting systems.

The Student Aid for Field Epidemiology Response (SAFER) program, which has been in continuous operation since 2005 at the University of Arizona in Tucson, Arizona, is one of the oldest student outbreak response teams in the United States.¹ This program uses a graduate-level course to train public health students to conduct case investigations and contact tracing for various infectious disease outbreaks. The SAFER team was able to quickly add severe acute respiratory syndrome coronavirus 2 cases to our workload in mid-March 2020 to assist the local health department. Since that time, we have been asked to help several other county health departments in Arizona with their COVID-19 response. These county health departments serve communities that encompass about 85% of the state’s population. We have also simultaneously taken the lead on contact-tracing efforts for the University of Arizona as one of the key components of the campus reopening plan. From April 1 through September 1, 2020, we identified key components to structure a student team for a pandemic response (n = 202 342 cases in Arizona as of September 1, 2020).⁴ This article provides guidance for (1) other universities and colleges interested in offering assistance to health departments, (2) health departments looking for ways to diversify their workforce, and (3) existing academic and applied public health teams considering ways to expand their capacity, based on our experience with the SAFER program.

Materials and Methods

Volunteer Management

Students first complete privacy and confidentiality agreements for local health departments. Then they are directed to online training materials and are given access to trained supervisors who can answer any questions that arise during training (Figure 1). All required documents and training resources are posted on a shared Google Drive folder for easy access for potential volunteers, which allows for editing by managers and supervisors when protocols change and continuous access to up-to-date resources. Although some students simply volunteer, other students participate through a newly created dual undergraduate and graduate public health course that allows students to learn about case investigation and contact tracing and requires students to spend their 10 hours of out-of-class time per week in the virtual call center. The virtual call center is an adaptation of SAFER’s physical call center to allow for supervised remote completion of case investigations and contact tracing by volunteers; more detail is provided hereinafter. More than 70 students were virtually engaged with the program through the course as of September 1, 2020.

Figure 1.

The University of Arizona Student Aid for Field Epidemiology Response volunteer onboarding process (ie, completing training materials and gaining access to systems) for the coronavirus disease 2019 (COVID-19) pandemic response in Arizona. Abbreviation: HIPAA, Health Insurance Portability and Accountability Act. Information about HIPAA is available through most university institutional review board websites. Colorado Food Safety for Excellence is a good option for interviewing. There are a variety of options for contact tracing, including Johns Hopkins University, the Association of State and Territorial Health Officials, and the Centers for Disease Control and Prevention.^5-7

After students have completed all necessary onboarding (ie, completing training materials and gaining access to our online programs), they are required to attend a training and orientation session held using Zoom, an online video conferencing software. Volunteers must attend a 30-to-60–minute orientation session before their first shift, where they can ask questions about the system and go through our “What if . . .” document. This document provides answers to common scenarios such as “What if you are telling the person their test results for the first time?,” “What if there was a death in the family from COVID-19?,” or “What if the person is concerned about sharing personal information about their contacts?”

Finally, students are asked to sign up for a minimum 2-hour shift using ScheduleSource, an online scheduling platform. A calendar on Google Docs where students can sign up for shifts also worked well when our team only had 20-40 people. The calendar shows when student supervisors are working in the virtual call center and lists available slots. A perpetual video conferencing session that is password protected and allows various users to claim the host function is used for daily shifts. If it is the student’s first shift, the supervisor will listen via Zoom Health and provide feedback either during or after their telephone calls to case patients using the chat function or verbally.

Virtual Call Center

Before the COVID-19 pandemic, our team operated out of a small laboratory space that allowed in-person interaction between students and student supervisors. At the beginning of the pandemic, we were faced with the choice of shutting down entirely or moving to an online structure. We went through a series of information technology (IT) hurdles (eg, access to case-tracking systems) and failed experiments (eg, conference calls, web-based texting) before deciding to use Zoom Health as a virtual call center platform. The software provides both password protection for security and the opportunity to use breakout rooms for smaller supervisor/volunteer groups to address topics or questions. Students generally stay muted to reduce noise during interviews and stay in breakout rooms while they are conducting interviews.

Students use an online calling platform, Amazon Connect, that can mask the caller’s identification, display “public health” on the recipient’s telephone, and direct callbacks to a central line and voicemail. Various services have similar functions. Our University IT team repurposed an IT system, Cherwell, as a call management software for case investigations for our team. Cherwell tracks call attempts, case notes, and all the information needed for our case interviewers to complete their case investigations. We repurposed the software into a basic, first-iteration product for our team; it has since been upgraded to help refine our processes further and expand our capacity to aid multiple health departments across the state. We also worked with our university Data Security and Privacy Office to achieve Business Associate Agreement/Health Insurance Portability and Accountability Act (HIPAA) licenses for all of our software to ensure data privacy. Other systems such as Trellis, Twillio Flex, and CommCare are available for purchase; cost is based on the extent to which the software needs to be modified for each site.

Case Investigations and Contact Tracing

Case investigations are the heart of the public health response, and students can be trained to conduct these interviews effectively.^1,2,8 We use training materials from the SAFER course and online resources, such as training videos from the Colorado Integrated Food Safety Center for Excellence interviewing modules.⁹ Our interview focus may differ from that of the Colorado Integrated Food Safety Center for Excellence, but the interviewing techniques are the same, and these materials provide students with real-world scenarios to follow.

Contact tracing is the next key public health step for mitigating transmission within a community. These steps include exposure notifications, up to 14 days of daily follow-up to track symptoms, and continual public health messaging on quarantine and isolation if the person becomes ill. Some health departments may have students help with case investigations, contact tracing, or both. Several quality training modules available through Johns Hopkins University,⁵ the Association of State and Territorial Health Officials (ASTHO),⁶ and the Centers for Disease Control and Prevention (CDC)⁷ also provide completion certificates for contact-tracing activities.

Many universities, including our own, have made contact tracing one of the key elements of their COVID-19 reentry plan. Our team, in conjunction with our local health department, receives results for students, faculty, and staff members who use our on-campus testing program, which uses both polymerase chain reaction (PCR) and antigen testing for severe acute respiratory syndrome coronavirus 2. This collaboration is part of a broad university effort to test students using on-campus PCR and antigen testing (thus decreasing time for results to be returned) and contact tracing. This simultaneous notification allows contact tracing to initiate several days before it would if only routine reporting was used.

Data Management

Data management is one of the biggest hurdles to overcome when managing a large team with sensitive data. Before the COVID-19 pandemic, our team manually entered data from the state’s electronic reporting system, Medical Electronic Disease Surveillance Intelligence System (MEDSIS), into a Microsoft Access database for call management. With Cherwell, our team can simultaneously upload case information for multiple cases into our system, export data for reporting to local health departments, and assign cases to investigators as needed (eg, Spanish speakers).

Cases assigned to the SAFER team are exported from MEDSIS using SAS (SAS Institute Inc) code, which runs automatically at 5 AM every day. The data are then shared with the team through a secure messaging portal owned by the health department. The SAFER team supervisor then uploads the case information into Cherwell for follow-up that day. All information from case investigations is entered into a Qualtrics (Provo, UT) survey owned by the state health department and used to collect data during the interviews. Data from the day’s interviews are then exported and uploaded to MEDSIS using SAS version 9.4 (SAS Institute Inc).

Results

From April 1 through September 1, 2020, our team completed 1910 case investigations from 6052 assigned cases. We recruited 233 volunteers and had 46 hourly student workers and staff members. During the same period, we contributed approximately 1619 volunteer hours and more than 1947 hours of paid staff and student worker time. With a minimum of 3 telephone calls attempted per case, we made more than 18 000 outbound calls and left thousands of voicemails to reach case patients. Altogether, we have contributed more than 3500 hours to local and state health department efforts, with an average of 220 combined hours per week, as we continue working through the COVID-19 pandemic. We continue to receive technical support from university staff members in addition to the estimated 100 hours they dedicated to building the Cherwell systems and integration.

Until July 24, 2020, this work was completed pro bono for the participating health departments. Our funds are a mix of state, local, and university resources. From local and state health departments, we have received contracts either directly or indirectly tied to funding from the Coronavirus Aid, Relief, and Economic Security (CARES) Act. University funds were those earmarked for campus reentry projects that also include activities such as wastewater surveillance, development of surveillance applications, and widespread testing. We currently operate from 9 AM to 7 PM 7 days per week to allow for volunteers to choose from numerous shift times and to maximize the variability in time for case patients to be called. Our team operates as an arm of the county and state health departments. We have intergovernmental agreements or contracts in place between the university and each health department. Several agreements and operational plans were already in place for our regular surveillance and outbreak activities when the pandemic started. Details on our structure are available elsewhere.¹⁰ All parties worked quickly to put these agreements in place so that students could be brought on as quickly as possible. For SAFER, the contracts outline our deliverables (either number of cases assigned and interviewed or number of hours devoted to a county each week) and data protections that have been put in place. For the county health departments, the contract outlines the direction that is provided, but all staff member and student management is handled by the university. This structure is opposed to having students volunteer directly for the health department, which is also a model that can be successful but requires training and onboarding (described hereinafter). These efforts can be scaled up or down as needed. We originally assisted with approximately 20 cases per day. With additional paid student supervisors and volunteers on board, we now take 65-150 cases per day and are ramping up for an additional 100 cases.

Discussion

Our team went through multiple troubleshooting steps in survey development, logistics, and communication. Our existing infrastructure made it easier for us to operationalize, but we learned several lessons and protocols that we believe may be beneficial for other academic or applied public health teams considering such a collaboration. For example, automating steps for our supervisors allowed more time for interviews. Before automating these processes, supervisors spent most of their shifts entering data into Cherwell, MEDSIS, and Qualtrics to ensure data quality and communication between our team and the health departments. Automating these steps came later in our process of building these systems and is the most recent major step in our system improvements. Whenever possible, health departments and student teams should consider ways to automate and streamline processes, particularly data uploads and communication.

From April 1 through September 1, 2020, it became apparent that within a health science center such as ours, faculty and staff members who manage these teams will likely not have a volunteer recruitment problem so much as a volunteer management problem. In particular, organizing, training, and briefing large and diverse groups of interested volunteers with little or no background knowledge of contact tracing becomes increasingly difficult as volunteer volume grows. We have found that it is helpful to put a single person in charge of the onboarding process to communicate training requirements with new volunteers and track their documents. We have also found the limits of using a Google Doc calendar when more than 20 volunteers sign up each day. ScheduleSource is a viable option for groups looking to scale up, because it allows for various user levels (eg, supervisors vs volunteers) and user views so that volunteers can sign up for available shifts more easily than with a Google Doc calendar. ScheduleSource also tracks efforts each week (eg, number of hours worked, individual volunteer reports) so that students can track their efforts if needed for a class and the team can add shifts for the next week. However, this feature incurs an added cost that many teams may not be able to cover.

With additional funding from health departments, our team hired additional staff members, including graduate and undergraduate students, and recruited volunteer groups with classes in preparation for the fall 2020 semester. New hires are drawn from the existing workforce when possible; recent graduate-level students are hired for staff member positions, current volunteers are hired for student worker positions, and most recruitment for new volunteers is through word-of-mouth or enrollment in classes.

Some health departments will have external surveys available that students can use. In Arizona, investigators enter their interview information directly into MEDSIS. Giving potentially hundreds of student volunteers access to that system is not a good idea from a data privacy perspective; as such, we opted to re-create their data entry form in a separate survey using Qualtrics, an online survey tool (REDCap would also work well). Using this system has 4 benefits:

1. The survey can be designed with standardized scripts and prompts for students to follow, which is beneficial for students who are new to case investigations.

2. Data entered into the system can undergo built-in quality control checks (eg, limiting an age range of 0-105 years) to reduce error in data entry.

3. Quality control checks can also be conducted by student supervisors during each shift using the reporting or analytics functions to check for missing information and give real-time feedback to volunteers to improve their data collection techniques.

4. Hundreds of students can access the survey simultaneously without fear of overloading the system.

Practice Implications

As our efforts for the state have increased during the COVID-19 pandemic, our management span of control has also increased. Although 1 faculty member (K.P.B.) oversees the entire program, it is not possible to have all student volunteers report to 1 faculty or staff member. Our organizational chart (Figure 2) includes 2-5 managers who are all full- or part-time staff members. Each has a master’s of public health degree and has previously worked with the team. These managers are responsible for all troubleshooting with our internal IT team and work on issues including data management and tracking, assistance with hiring, volunteer training and routine data reporting on the team’s productivity, and internal metrics. Each staff member oversees 2-5 hourly student workers. These students are all graduate public health students or undergraduate seniors who have excelled as volunteers. Each of these students manages up to 10 volunteers in any given shift in our virtual call center. Student managers compile daily reports for health departments, are available for students via Zoom Health to ask questions during and between interviews, and conduct interviews themselves when few volunteers are available.

Figure 2.

Overall staffing structure of the University of Arizona Student Aid for Field Epidemiology Response (SAFER) team in response to the coronavirus disease 2019 (COVID-19) pandemic. Interviews are conducted by hourly student workers when few volunteers are available. Abbreviation: IT, information technology.

For many student teams, including our own, contact tracing is a key metric for a university’s plans to reopen for in-person learning. Depending on the capacity of the local health jurisdiction to conduct timely contact tracing, the inclusion of student volunteers has been an invaluable addition to the University of Arizona’s surge plan and should be evaluated after the pandemic for efficacy and use at other universities. Given the nature of university housing and social interactions among college students (eg, sharing living/studying spaces, having prolonged close contact at social gatherings), the potential for widespread transmission in dormitories, fraternities, sororities, and other social settings is high.¹¹ Having student teams that can focus exclusively on cases and contact tracing for campus can help speed up case investigations and exposure notifications for these populations and, ideally, reduce spread within and beyond the campus community.

These efforts require funding from health departments, and that funding cannot be absorbed by a university unless those efforts directly relate to reentry planning. The return on investment for continued funding for student outbreak response teams and surge capacity funding is 2-fold. First, these teams provide a valuable resource for health departments that need to scale their response without hiring internally. Second, they provide a valuable learning experience and contribute to capacity building for the future public health workforce.^1,2

Conclusions

The scale of this student response may not be feasible for every university or college to provide to their local health department. Our hope is that this overview of our program’s efforts will allow academic programs and public health agencies to consider solutions for logistical challenges that we are all facing during these unprecedented times in modern public health. Trained students are also a great conduit for public health messaging through social media to their peers, which, in many places, represent one of the highest case counts.^4,12 These students will also be entering the job market soon, likely before this pandemic is over, and having them pretrained in these areas will allow them to be efficient and effective in their roles as staff members at federal, state, and local public health agencies that will need their expertise.

ORCID iD

Erika Austhof, MPH https://orcid.org/0000-0002-1217-8733