Controlling the course of the New Zealand COVID-19 epidemic 

By Dr Mark Thomas, Associate Professor in Infectious Diseases, University of Auckland

14 April 2020

New Zealand is in the very early stages of an epidemic of COVID-19 disease, caused by a novel coronavirus, SARS-CoV-2. Infection with this respiratory virus causes no or minimal illness in approximately 50% of infected people, mild to moderately severe illness in approximately 48%, and potentially fatal disease in approximately 2%. Because this virus is so different from other viruses everyone who has not already been infected and recovered is susceptible. However, spread of the disease through the community would result in the development of immunity in those who became infected and survived. Once approximately 80-90% of the population had been infected the high level of herd immunity would dramatically slow the spread of the infection in the community. (1,2)

The effect of physical distancing is to reduce the reproductive number (R) - the average number of people who acquire infection from an infected person. The reproductive number of SARS-CoV-2 in most societies, in the absence of any physical distancing interventions, appears to be between 2 and 3. Experience in China, and in many other countries, shows that strong physical distancing interventions can consistently reduce the reproductive number to less than 1, so that on average each infected person transmits infection to less than one person, and the epidemic declines, and may even temporarily cease, as shown for China in Figure 1.

Figure 1. The number of daily new confirmed cases of COVID-19 in several countries, between 20th January and 1st April 2020. For all countries, to varying degrees depending on the national rates of laboratory testing, the total number of cases would have been be substantially greater than the total number of confirmed cases. The large increase in confirmed cases in China during 11-15 February was the result of a change in reporting methodology and did not reflect changes in actual disease incidence. Note the levelling off and/or decline in the reported number of daily cases in China in February, and in some other nations in March, following the widespread introduction of physical distancing. (Reference 3)

However, if physical distancing is relaxed before 80-90% of the population have developed immunity (either as the result of natural infection or as the result of vaccination) there is a risk that the epidemic will recur. This risk may result from transmission of infection, either from unrecognised cases in New Zealand, or from people newly arriving in New Zealand with the infection. Because an effective vaccine is not expected to be available until late 2021, an option considered by many nations, including New Zealand, is to allow controlled spread of infection through the community. This would lead to a gradual increase in the proportion of the population who are immune, and potentially could result in 80-90% of the population becoming immune. Allowing spread of infection through the community would inevitably result in some patients developing severe disease, but careful ongoing regulation of physical distancing could ensure that the number of patients with severe disease did not exceed the human and physical resources of our healthcare system, particularly the capabilities of hospital intensive care units. 

A possible strategy to allow continued spread of infection in the community, while consistently avoiding excessive numbers of severely unwell patients overloading the healthcare system, is illustrated in Figure 2. This strategy relies on periods of strong physical distancing interrupted by periods of less strong physical distancing. During the periods of less strong physical distancing the incidence of infection and disease would rise, while during the periods of strong physical distancing the incidence of infection and disease would fall. The duration of these alternating periods, and the strength of the physical distancing perhaps could be adjusted to ensure that the number of patients who require intensive care does not exceed the capability of the healthcare system to provide that care.

Figure 2. The effect of periods of strong physical distancing (R=0.75) shown in blue, alternating with periods of less strong physical distancing (R=1.75) shown in white, on the predicted number of cases of COVID-19 disease shown by the orange line, for a period of 1800 days from the onset of the epidemic in New Zealand in early 2020. The solid red line shows the predicted number of cases in the absence of any interventions to slow transmission of disease in the community. The predicted peak prevalence of symptomatic disease, in the absence of any interventions, is approximately one person in six. The horizontal dashed pink line represents 4,000 current infections (a prevalence of approximately 8 infections per 1000 population) and approximates the infection prevalence that might result in 500 people requiring intensive care (equating to the anticipated maximum capacity of all intensive care departments in New Zealand during the coming year). (Reference 2)

An alternative strategy is to try to consistently suppress the transmission of infection within New Zealand, by imposing very strong border controls, a period of strict isolation for all people arriving in New Zealand, intensive case finding and isolation of any new cases, and testing and isolation of their close contacts. Because this strategy would prevent widespread transmission of infection the overwhelming majority of the population would remain susceptible to infection and disease. Therefore, this strategy would need to be maintained until after an effective vaccine had been administered to approximately 80-90% of the New Zealand population. If successful this strategy would be expected to result in much less severe pressure on hospitals, and very few deaths from COVID-19, but would require sustained, intensive, public health activity to consistently prevent transmission of infection within New Zealand, and access, within a reasonable period of time, to an effective vaccine.


Click here for references.