Based on a new modelling study published in the Lancet Infectious Diseases journal, a combined approach of physical distancing including school closure, workplace distancing and quarantine of infected individuals and their families has been found to be the most effective approach to lower the number of SARS-CoV-2 cases in comparison to the other intervention scenarios conducted in the study.
The study, which was done in a simulated Singapore setting, is the first of its kind to probe into early intervention options in the country via simulation. The pandemic risk still persists as the number of cases continues to rise in the country. This is even though suspected and infected individuals have been isolated and surveillance efforts increased. As of 23 March, workplace distancing has been suggested and schools have been reopened, though this is not a national policy.
The second best approach, according to the study, is quarantine done together with workplace measures. The next best approach is quarantine combined with school closure, followed by the approach of only quarantine. Compared to no intervention, all these intervention scenarios do help with lowering the spread.
A national outbreak could be prevented through the combined approach, the study reported. This prevention occurs at a relatively low levels of infectivity (basic reproductivity value, R0 = 1.5). However, at higher infectivity scenarios (moderate and likely reproductivity with R0 = 2.0) and R0= 2.5, it is more challenging to prevent outbreak because transmission still happens despite infections being reduced effectively.
“Should local containment measures, such as preventing disease spread through contact tracing efforts and, more recently, not permitting short-term visitors, be unsuccessful, the results of this study provide policy makers in Singapore and other countries with evidence to begin the implementation of enhanced outbreak control measures that could mitigate or reduce local transmission rates if deployed effectively and in a timely manner,” according to Dr Alex R Cook from the National University of Singapore.
The authors developed an individual-based influenza epidemic simulation model to assess the potential impact of interventions on outbreak magnitude if local containment fails. The simulation model estimates the likelihood of human-to-human transmission of SARS-CoV-2 by accounting for social contact rates in workplaces, schools, and homes as well as demography and individual movement.
Among the model parameters in the simulation are (1) the duration of hospital stay after symptom onset (3.5 days), (2) the cumulative distribution function for the mean incubation period (both SARS and COVID-19 viruses have the same mean incubation period of 5.3 days), (3) the proportion of the population assumed to show no symptoms (7.5 per cent) and how infectious an individual is over time.
The cumulative number of SARS-CoV-2 infections at 80 days was estimated by the model, after detecting 100 cases of community transmission. The authors chose three values for R0, which are (1) relatively low (R0= 1.5), (2) moderate and likely (R0= 2) and high transmissibility (R0= 2.5). R0 values were chosen based on the analyses of figures from infected individuals in Wuhan, China.
Aside from the baseline scenario with no interventions, another four additional intervention scenarios were simulated after local containment failed. The additional intervention scenarios are (1) isolation of infected individuals and quarantine of their family members (quarantine), (2) quarantine plus immediate school closure for 2 weeks, (3) quarantine plus immediate workplace distancing, in which 50 per cent of the workforce is encouraged to work from home for 2 weeks and (4) a combination of quarantine, immediate school closure, and workplace distancing. These interventions imitate some of the measures implemented by the Singaporean Ministry of Health such as workforce distancing and quarantine.
In the baseline simulation with R0= 1.5, the median cumulative number of infections at day 80 was 279,000 or 7.4 per cent of the resident population of the country. As infectivity increases, the median number of infections also rose: R0= 2.0 equates to 727,000 cases with 19.3 per cent of the population whereas R0= 2.5 equates to 1,207,000 cases with 32 per cent of the population.
The combined intervention protocol was the most effective in comparison to the baseline scenario, with the estimated median number of infections lowered by 99.3 per cent when R0= 1.5, with 1,800 estimated cases. Even so, outbreak prevention becomes substantially more challenging at higher infectivity. For instance, at R0= 2.0, the estimated median cases were 50,000, which is 93 per cent lower than baseline. On the other hand, at R0= 2.5, the cases were 258,000 which is 78.2 per cent lower than baseline.
The study also investigated the potential impact that arises if the proportion of asymptomatic cases in the population was more than 7.5 per cent. There are challenges posed with the combination of high asymptomatic proportion and low infectivity (R0 being 1.5 or less). If asymptomatic proportions rises up to 50 per cent, at day 80, there would be 277,000 estimated infections with combined intervention in place. This is much higher than the 1,800 infections when R0= 1.5 at baseline.
“If the preventive effect of these interventions reduces considerably due to higher asymptomatic proportions, more pressure will be placed on the quarantining and treatment of infected individuals, which could become unfeasible when the number of infected individuals exceeds the capacity of health-care facilities. At higher asymptomatic rates, public education and case management become increasingly important, with a need to develop vaccines and existing drug therapies,” Dr Cook further explained.
Some of the limitations highlighted by the authors are (1) the dynamics of contact patterns between individuals, (2) the impact of seeding of imported cases, (3) impact of migrant movement , (4) dated census population data and (5) other unforeseen factors. In addition, the epidemiological features of COVID-19 are still unclear in terms of the infectivity and transmission profile of the virus. Due to this, the authors adopted the baseline from SARS-CoV when modelling (1) how infectious a person is across time, (2) time between symptom onset and hospital admission and (3) asymptomatic rate.
“Although the scientific basis for these interventions might be robust, ethical considerations are multifaceted. Importantly, political leaders must enact quarantine and social-distancing policies that do not bias against any population group. The legacies of social and economic injustices perpetrated in the name of public health have lasting repercussions. Interventions might pose risks of reduced income and even job loss, disproportionately affecting the most disadvantaged populations: policies to lessen such risks are urgently needed. Special attention should be given to protections for vulnerable populations, such as homeless, incarcerated, older, or disabled individuals, and undocumented migrants. Similarly, exceptions might be necessary for certain groups, including people who are reliant on ongoing medical treatment,” as noted by Dr Nathan C Lo of University of California, San Francisco and Dr Joseph A Lewnard of University of California, Berkeley.