Meredith Matone, DrPH, MHS, is the scientific director of PolicyLab at Children’s Hospital of Philadelphia (CHOP). Deanna Marshall, MPH, is a clinical research coordinator at PolicyLab at CHOP.
As communities reopen across the country, strategies for preventing another wave of widespread COVID-19 transmission are top of mind. Arguably, contact tracing is the approach receiving the most media attention and around which many states are mobilizing. The premise of contact tracing is to create a bubble around a potential outbreak within a social network that prevents further spread, and we’ve seen it utilized in successful approaches in countries such as Vietnam.
Given the momentum for large-scale adoption of this strategy in the United States, we suggest the need for a more nuanced discussion of contact tracing’s likely effectiveness as the primary transmission mitigation strategy for the summer and fall – and propose a precision public health approach that considers strengthened surveillance paired with targeted contact tracing.
Contact tracing is a cornerstone of public health infectious disease mitigation and has proven effective, for example, in stemming transmission of tuberculosis and HIV within Southeast Asian and African countries. Contact tracing is also regularly deployed by U.S. states and local health departments for monitoring potential outbreaks of influenza and vaccine-preventable diseases such as measles. A version of the method is even used to identify the source of food-borne illness outbreaks. These scenarios are not overly generalizable to our current SARS-CoV-2 reality.
Receiving less attention, but as important, is the public health practice of surveillance. Operationalizing surveillance with increased intensity as we reopen is critical. Though neither surveillance nor contact tracing is a perfect fit for the many pathogen-specific and societal complexities of the COVID-19 pandemic in the U.S., together, we could deploy them in a pragmatic public health response that plays to each method’s strengths and considers current resource realities in order to improve early detection of potential outbreaks, protect the most vulnerable populations and use testing resources smartly.
This approach, we argue, would deploy syndromic and participatory surveillance as a primary state or county-level strategy for monitoring disease activity in the general population while prioritizing testing resources and corresponding contact tracing protocols for front line essential workers, those living in congregate settings (e.g., nursing homes, shelters, prisons) and medically high-risk individuals.
A precision public health approach of this nature is responsive to four realities:
- We lack testing resources and coordinated state-level testing procedures to render contact tracing a viable method of population-level transmission mitigation.
- The evidence to support contact tracing as a transmission mitigation strategy strongly suggests its effectiveness is in situations of low circulating cases. As we enter summer, this assumption is likely not to hold true based on PolicyLab at Children’s Hospital of Philadelphia’s modeling projections.
- SARS-CoV-2 has both a long period of infectiousness (estimates range from 5 to 14 days) and presymptomatic (or possibly asymptomatic) spread. This presents challenges in contact tracing because its success relies on an individual’s ability to recall contacts with while contagious, including prior to feeling ill. During this time, they may have been in public places or at work. Thankfully, the expert opinion so far on SARS-CoV-2 suggests that brief interactions confer less risk and masking may limit person-to-person spread. Still, as we ease social distancing protocols, the ability to recall will lessen—so will our ability to play catch up on contact tracing. Public transit, large workplaces and dense urban areas pose particular challenges to contact tracing. Location-based contact tracking apps, as Australia and Singapore have launched, can address recall but have yet to be proven and are limited to individuals with cellphones, WiFi access and technological literacy—disproportionately leaving behind elderly, low-income and rural residents. In addition, to date, voluntary programs have had challenges with uptake and such apps have also raised ethical and privacy concerns.
- While evidence suggests contact tracing, and its associated quarantine policies, does not require 100% compliance to be effective, we know there are populations for whom compliance will present an inequitable burden—namely, individuals in under-resourced communities. To date, it is unclear how state-led contact tracing efforts will help mitigate the challenges of being quarantined, which will require community-level coordination of employment security and pay, food and prescription delivery, and navigation to other needed community social safety net programs. Without effective case management to trusted community-level resources and workforce protections, exposed individuals may be unable to comply with quarantine notifications following a tracing call.
These realities aside, contact tracing must be done. But the scope and focus of these efforts are likely best served among medically high-risk populations and those in employment or residential settings with high-exposure risks, and not as part of an effort to stand-up mass test-trace initiatives on the general population.
We propose a contrasting approach that includes a number of synchronous activities.
- An enhanced surveillance model featuring syndromic and participatory methods: We can adapt and leverage current state and local public health syndromic surveillance for influenza, combined with standardized reporting of hospitalization data, for SARS-CoV-2. Syndromic surveillance is a proven intervention for which the opportunity for utility has nowhere to go but up as health systems transition to highly sophisticated electronic medical record platforms with increased interoperability (e.g., Health Information ) Supplemental information feeds to increase sensitivity could include the addition of telephone triage data for symptom complaints into doctor’s offices (an approach used by the VA for influenza) and monitoring of all-cause mortality data.
Participatory disease surveillance involves individuals actively reporting illness symptoms through a website, text-messaging survey or engagement with community health workers. There are promising examples of web-based reporting platforms with demonstrated feasibility and validity for influenza surveillance in Europe and Canada. Participatory surveillance systems have detected outbreaks as earlier than traditional surveillance and can be successful without even close to universal participation. Researchers in Europe and China have already begun to validate SARS-CoV-2-case definitions for public reporting in an app-based platform. Though symptom-based reporting approaches for this disease aren’t perfect because individuals’ symptoms tend to vary in combination and intensity, these methods still represent a low-resource, large-scale intervention with a growing evidentiary foundation. Deployment of participatory surveillance can occur through health departments, large employers, schools and community institutions (e.g., libraries). In this way, though technology-reliant, this approach can overcome accessibility challenges. Moreover, participatory surveillance methods can be anonymous to some extent, allowing local or state public health officials to identify specific communities or workplaces of focus for targeted testing protocols.
Larger employers and schools may also opt to implement other surveillance metrics including absence monitoring or smart thermometer use, and this information should be coordinated with health departments.
Additionally, direct outreach to hard-to-reach communities (e.g., refugee or homeless populations) can help expand these efforts to individuals who may be unlikely to seek health care, those with limited English proficiency, and those with limited internet or phone connectivity. A community-based approach, used in Ebola response, employs community health workers to identify symptoms through door-to-door visits and leverages trusted community leaders or institutions to collect data. This direct form of engagement has been effective with as little as 40% participation.
An investment in a well-coordinated multi-method surveillance effort is significantly less resource-intensive than scaled testing and contact tracing and may achieve similar results for the general population.
- Prioritized testing for high-risk populations: Focusing testing on high-risk groups will maximize the benefits of testing and consequent contact tracing. Further, there are efficient testing approaches, like block testing, that preserve reagent and minimize the number of tests needed to identify positive cases within groups. In this approach, testing is prioritized to clear high-risk employees for workforce participation and routinely monitor high-risk residential settings and medically at-risk populations.
- Community-level quarantine protocols: It is essential that communities be nimble in moving into brief periods of local mitigation procedures (e.g., closure of non-essential businesses) during outbreaks. State and local policymakers should plan to leverage surveillance and testing data to make real-time decisions about whether or not to enact these protocols in hotspotted areas. More expedient and targeted entry into community-level mitigation may produce fewer and shorter periods of closure. Communities entering closures due to an increase in circulating cases may pursue targeted testing approaches (stratified community sampling) coupled with syndromic surveillance data inclusive of hospitalization data to determine when it is safe to reopen.
An approach that relies on the strategic use of foundational public health operations is likely to meet the challenge of resource constraints without compromising quality or perpetuating inequity. It’s not too late for states to enact pragmatic approaches grounded in these foundational public health tools to protect against another devastating wave of COVID-19. The time is ripe to invest in a robust surveillance infrastructure.