Medicine (RMH) - Research Publications
Permanent URI for this collection
Search Results
1 - 4 of 4
-
ItemVaccination is Australia's most important COVID-19 public health action, even though herd immunity is unlikely( 2021)The Australian National Cabinet four-step plan to transition to post-pandemic re-opening begins with vaccination to achieve herd protection and protection of the health system against a surge in COVID-19 cases. Assuming a pre-vaccination reproduction number for the Delta variant of 5, we show that for the current Mixed program of vaccinating over 60s with AstraZeneca and 16-60s with Pfizer we would not achieve herd immunity. We would need to cover 85% of the population (including many 5-16 year-olds to achieve herd immunity). At lower reproduction number of 3 and our current Mixed strategy, we can achieve herd immunity without vaccinating 5-15 year olds. This will be achieved at a 60% coverage pursuing a strategy targetting high transmitters or 70% coverage using a strategy targetting the vulnerable first. A reproduction number of 7 precludes achieving herd immunity, however vaccination is able to prevent 75% of deaths compared with no vaccination. We also examine the impact of vaccination on death in the event that herd immunity is not achieved. Direct effects of vaccination on reducing death are very good for both Pfizer and AstraZeneca vaccines. However we estimate that the Mixed or Pfizer program performs better than the AstraZeneca program. Furthermore, vaccination levels below the herd immunity threshold can lead to substantial (albeit incomplete) indirect protection for both vaccinated and unvaccinated populations. Given the potential for not reaching herd immunity, we need to consider what level of severe disease and death is acceptable, balanced against the consequences of ongoing aggressive control strategies.
-
ItemIs Nigeria really on top of COVID-19? Message from effective reproduction number( 2020)Following the importation of Covid-19 into Nigeria on the 27 February 2020 and then the outbreak, the question is: how do we anticipate the progression of the ongoing epidemics following all the intervention measures put in place? This kind of question is appropriate for public health responses and it will depend on the early estimates of the key epidemiological parameters of the virus in a defined population. In this study, we combined a likelihood-based method using a Bayesian framework and compartmental model of the epidemic of Covid-19 in Nigeria to estimate the effective reproduction number (R(t)) and basic reproduction number (R_0). This also enables us to estimate the daily transmission rate (β) that determines the effect of social distancing. We further estimate the reported fraction of symptomatic cases. The models are applied to the NCDC data on Covid-19 symptomatic and death cases from 27 February 2020 and 7 May 2020. In this period, the effective reproduction number is estimated with a minimum value of 0.18 and a maximum value of 1.78. Most importantly, the R(t) is strictly greater than one from April 13 till 7 May 2020. The R_0 is estimated to be 2.42 with credible interval: (2.37, 2.47). Comparing this with the R(t) shows that control measures are working but not effective enough to keep R(t) below one. Also, the estimated fractional reported symptomatic cases are between 10 to 50%. Our analysis has shown evidence that the existing control measures are not enough to end the epidemic and more stringent measures are needed.
-
ItemFlattening the curve is not enough, we need to squash it: An explainer using a simple model( 2020)
Background
Around the world there are examples of both effective control (e.g., South Korea, Japan) and less successful control (e.g., Italy, Spain, United States) of COVID-19 with dramatic differences in the consequent epidemic curves. Models agree that flattening the curve without controlling the epidemic completely is insufficient and will lead to an overwhelmed health service. A recent model, calibrated for the UK and US, demonstrated this starkly.Methods
We used a simple compartmental deterministic model of COVID-19 transmission in Australia, to illustrate the dynamics resulting from shifting or flattening the curve versus completely squashing it.Results
We find that when the reproduction number is close to one, a small decrease in transmission leads to a large reduction in burden (i.e., cases, deaths and hospitalisations), but achieving this early in the epidemic through social distancing interventions also implies that the community will not reach herd immunity.Conclusions
Australia needs not just to shift and flatten the curve, but to squash it by getting the reproduction number below one. This will require Australia to achieve transmission rates at least two thirds lower than those seen in the most severely affected countries.The known
COVID-19 has been diagnosed in over 4,000 Australians. Up until mid-March, most were from international travel, but now we are seeing a rise in locally acquired cases.The new
This study uses a simple transmission dynamic model to demonstrate the difference between moderate changes to the reproduction number and forcing the reproduction number below one.The implications
Lowering local transmission is becoming important in reducing the transmission of COVID-19. To maintain control of the epidemic, the focus should be on those in the community who do not regard themselves as at risk. -
ItemStepping out of lockdown should start with school re-openings while maintaining distancing measures. Insights from mixing matrices and mathematical models.(Cold Spring Harbor Laboratory, 2020)Australia is one of a few countries which has managed to control COVID-19 epidemic before a major epidemic took place. Currently with just under 7000 cases and 100 deaths, Australia is seeing less than 20 new cases per day. This is a positive outcome, but makes estimation of current effective reproduction numbers difficult to estimate. Australia, like much of the world is poised to step out of lockdown and looking at which measures to relax first. We use age-based contact matrices, calibrated to Chinese data on reproduction numbers and difference in infectiousness and susceptibility of children to generate next generation matrices (NGMs) for Australia. These matrices have a spectral radius of 2.49, which is hence our estimated basic reproduction number for Australia. The effective reproduction number (Reff) for Australia during the April/May lockdown period is estimated by other means to be around 0.8. We simulate the impact of lockdown on the NGM by first applying observations through Google Mobility Report for Australia at 3 locations: home (increased contacts by 18%), work (reduced contacts by 34%) and other (reduced contacts by 40%), and we reduce schools to 3% reflecting attendance rates during lockdown. Applying macro-distancing to the NGM leads to a spectral radius of 1.76. We estimate that the further reduction of the reproduction number to current levels of Reff = 0.8 is achieved by a micro-distancing factor of 0.26. That is, in a given location, people are 26% as likely as usual to have an effective contact with another person. We apply both macro and micro-distancing to the NGMs to examine the impact of different exit strategies. We find that reopening schools is estimated to reduce Reff from 0.8 to 0.78. This is because increase in school contact is offset by decrease in home contact. The NGMs all estimate that adults aged 30-50 are responsible for the majority of transmission. We also find that micro-distancing is critically important to maintain Reff <1. There is considerable uncertainty in these estimates and a sensitivity and uncertainty analysis is presented.