The current pandemic has proved to be challenging to control without stringent and comprehensive non-pharmacological interventions (NPIs), relying mostly on isolation and contact tracing of cases, along with social distancing. However, strategies like these have a huge human cost in terms of restriction of social interactions, resulting in depressive tendencies, disruption of economic activity, and loss of educational resources.
Now, a new study by researchers at the University of Georgia and published on the preprint server medRxiv* in August 2020 compares different methods of suppressing viral spread without general social distancing, to allow for personal and business activity to resume. The researchers classify their methods as generalized and targeted methods.
The researchers define these two different intervention types: “Targeted interventions are interventions that are applied to specifically identified individuals in a population, typically based on infection or exposure status. Generalized interventions are behavioral or environmental interventions that are adopted broadly within a population.”
They evaluate four targeted and one generalized interventions for their effectiveness, cost, and robustness. The targeted approaches are divided into those which target infected persons vs. those which are directed towards uninfected persons.
Targeted Interventions 1: Targeting Infected People
These interventions aim to limit the risk of transmission, with the intensity of intervention increasing from first to last. These may remove the incentive to seek care since being diagnosed as COVID-positive removes the right to work, temporarily, in favor of the duty to protect.
Active Case Finding
Here, the onus is on the health authorities to seek out all possible cases by testing as widely as possible, including those at high risk such as healthcare workers, known contacts, and also by using the lowest test criteria to include as many people as required.
This strategy depends on identifying, communicating with, and following up possible exposed individuals from known cases. This not only serves to educate a high-risk segment of the population of their need to watch for symptoms but reduces the chances of spread from traced contacts by advising isolation. In addition, it contributes to active case finding.
Singapore - June 20, 2020: A woman scans a contact tracing QR code before entering a mall on the second day shops reopen during Phase 2 after Singapore's COVID-19 "circuit breaker". Image Credit: EA Alim / Shutterstock
This is the highest level of intervention related to stopping transmission from known cases and involves keeping cases and contacts out of circulation until the infective period is over. This means that secondary cases are isolated even before they develop symptoms, being in the traced contact group, and also dramatically reduces or suppresses the spread from these cases completely. It also reduces the rate of contact or exposure within the whole population.
Targeted Interventions 2: Targeting Uninfected People
This strategy provides an incentive to seek care as this certificate is required to return to work, and priority is given to the protection of other people.
The approach described here consists of certifying healthy people to limit exposure to the virus. Here, social distancing is reduced progressively by allowing certified individuals to enter workplaces, schools, and shops for the period of certified freedom from infection.
This may be due to durable immunity, as when the person tests positive for neutralizing antibodies. It may also be temporary, meaning the viral test has recently come back negative. However, it is essential to note that the researchers say, “Durable certification doesn’t lead to a reduction in transmission.”
The importance is rather in using immune people to provide essential goods and services while still slowing the speed of viral transmission.
These include the population-wide adoption of measures like wearing face masks, hand hygiene, cleaning and disinfection of potentially contaminated surfaces, providing adequate sick leave, and improving measures to ensure sick workers stay home.
Other approaches included here are contact-free transactions, infection barriers in public places where people gather, providing hand sanitizer in public places, avoiding physical contact with hands and faces outside the family, ensuring space between people in public spaces, and restricting gatherings beyond a certain limit.
Analysis of Effectiveness
Using these models for a population of 10 million people, they estimated the costs and benefits of each. They parametrized case finding as identifying 50% of cases, with five contacts traced per secondary infection, a 3-day delay in diagnostic testing, of a valid diagnostic certificate for a period of 7 days, and a 10-day delay in antibody testing.
The researchers assumed an incubation period of 4 days, with preclinical infections being 30% as infective as symptomatic ones. They also assumed that isolation or quarantine reduces it by 90%, and certification produces somewhat the same effect. Generalized interventions are supposed to reduce transmission by 30%. In contrast, social distancing is assumed to reduce viral spread by double that, at 60%.
They conclude that without any social distancing or targeted interventions, most of the population would be infected, with generalized interventions reducing the final size of the epidemic (FSE) by only a small amount. This means that without social distancing, intensive targeted interventions are the only way to prevent infection in most of the population.
The use of active case finding alone would require that 95% of cases be identified – which is an impossible requirement when about 20% of cases are asymptomatic (according to the current researchers – though other studies have put the figure much higher).
Adding in generalized interventions to the first approach pushes down the target for case finding to 80%, which is still much higher than the 1% to 10% of cases thought to have been detected in the first wave. However, the current study assumes that this strategy will detect 50% of cases.
The addition of contact tracing is beneficial, but the imposition of quarantine excludes transmission from general contacts while allowing for 10% of transmission within family members.
They note, “Eliminating this last 10% of transmission increases the total number of cases averted approximately tenfold from 250,000 to almost 2,500,000.”
With the exact presymptomatic transmissibility being unknown as yet, they looked at different levels, finding that for lower levels, the test can be valid for a longer duration, up to two weeks, and the period of waiting for the test can be increased, safely. However, this becomes more and more unsafe in terms of risking a major outbreak as the level of spread outside the symptomatic period increases to near the symptomatic spread level.
The findings indicate that while the transmission may be suppressed by any of these methods, a certain level of efficacy is required to achieve suppression of viral spread. The level will vary between strategies. Thus, targeted interventions for infected people will typically produce satisfying results only if a high proportion of cases are detected.
On the other hand, certification of uninfected persons will only be useful if widespread testing is possible. Either way, therefore, the ability to test widely and frequently is fundamental to the success of either of these strategies, in the absence of social distancing. Moreover, generalized interventions would be essential for their multiplier effect, adding that crucial push that achieves suppression of the epidemic.
The researchers call for a more complex model testing the optimal combination of these two types of targeted strategies for better efficacy.
They say, “Developing such a model could be a useful future step toward developing a complete understanding of transmission reduction via non-pharmaceutical interventions for acute infectious diseases.”
medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.