Five approaches to the suppression of SARS-CoV-2 without intensive social distancing

Abstract

Initial efforts to mitigate transmission of SARS-CoV-2 relied on intensive social distancing measures such as school and workplace closures, shelter-in-place orders and prohibitions on the gathering of people. Other non-pharmaceutical interventions for suppressing transmission include active case finding, contact tracing, quarantine, immunity or health certification, and a wide range of personal protective measures. Here we investigate the potential effectiveness of these alternative approaches to suppression. We introduce a conceptual framework represented by two mathematical models that differ in strategy. We find both strategies may be effective, although both require extensive testing and work within a relatively narrow range of conditions. Generalized protective measures such as wearing face masks, improved hygiene and local reductions in density are found to significantly increase the effectiveness of targeted interventions.

Keywords: COVID-19; SARS-CoV-2; coronavirus; non-pharmaceutical interventions.

Figure 1.

Compartmental model for Strategy 1 interventions. (Online version in colour.)

Figure 2.

Compartmental model for certifying infection status. (Online version in colour.)

Figure 3.

Two baseline scenarios which do not include targeted interventions. The top plot assumes that transmissibility, β, is at 22% of its original value due to social distancing. The bottom plot assumes that transmissibility, β, is at its natural value (βN = 0.8). Both plots assume that generalized interventions reduce transmissibility by a further 50%. Here, q = 0, κ = 0, δ = 1/10, ξ = 0 and all other parameters are as given in table 1. Initially, there are 500 latently infected individuals, 500 recovered individuals, with the remainder susceptible. (Online version in colour.)

Figure 4.

Strategy 1 approaches to suppressing COVID-19 transmission as a function of case ascertainment (q). The red dashed line shows the critical value q* at which suppression is achieved (R_c = 1) with generalized interventions. Generalized interventions reduce transmissibility by a further 50%. Without generalized interventions, suppression by active case finding is not possible. Approaches with contact tracing take α = 10 × β and κ = 1/3. The quarantine scenario assumes that $b_{I_t}=0$ and $b_{L_t}=0$. Other parameters are as in table 1. (Online version in colour.)

Figure 5.

Strategy 1 approaches to suppressing COVID-19 transmission with perfect isolation ($b_{L_t}=0$ and $b_{I_t}=0$) as a function of case ascertainment (q). The red dashed line shows the critical value q* at which suppression is achieved (R_c = 1) with generalized interventions. Generalized interventions reduce transmissibility by a further 50%. Without generalized interventions, suppression by active case finding is not possible. Approaches with contact tracing take α = 10 × β and κ = 1/3. The quarantine scenario assumes that $b_{I_t}=0$ and $b_{L_t}=0$. Other parameters are as in table 1. (Online version in colour.)

Figure 6.

Cases averted by reducing transmission from isolated patients from 10% to zero ($b_{L_t}=0$, $b_{I_t}=0$) as a function of case ascertainment (q). Other parameters are as described in figure 4. (Online version in colour.)

Figure 7.

Final outbreak size as a function of viral test validity duration (1/ξ) and test waiting time (1/κ) without generalized interventions. At the assumed value of presymptomatic transmission (b_L = 0.44), there is only a very small region (dark blue) within which certification can prevent a major epidemic. Outbreak size is given on a log₁₀ scale. Other parameters are as in table 1. (Online version in colour.)

Figure 8.

Final outbreak size as a function of viral test validity duration (1/ξ) and test waiting time (1/κ) with generalized interventions. At the assumed value of presymptomatic transmission (b_L = 0.44), there is only a very small region (dark blue) within which certification can prevent a major epidemic. Outbreak size is given on a log₁₀ scale. Other parameters are as in table 1. (Online version in colour.)

Figure 9.

Effect of presymptomatic infectivity on outbreak size for $0<b_{L_u}<1$. The vertical dashed line shows the default value of $b_{L_u}=0.44$ for comparison with other figures. Generalized interventions are assumed to reduce β by 50%. Other parameters are as in table 1 and q = 0.5. (Online version in colour.)

Figure 10.

Certification with generalized interventions for presymptomatic infectivity (b_L) assumed to be 40%, 44%, 60%, 80% and 100% of the baseline value of β × N = 0.8 for transmission with generalized interventions. Combinations of test validity and days to obtain a test that are below the contour have total outbreak sizes less than 10 000 and may be considered to be ‘contained’. The default value of b_L = 0.44 is plotted in grey for comparison with figure 8. Other parameters are m = 0.1, γ = 1/6, σ = 1/4 and δ = 0.1. (Online version in colour.)