Directors' Blog

Prime Minister recognises transformative science

Prime Minister recognises transformative science

The 2020 Prime Minister’s Science Prize has been awarded to Te Pūnaha Matatini for our contribution to Aotearoa New Zealand’s COVID-19 response.

The Prime Minister’s Science Prize is awarded for transformative science which has had a significant economic, health, social or environmental impact.

Te Pūnaha Matatini are being recognised for our work that developed a series of mathematical models, analysed data and communicated the results to inform the New Zealand Government’s world-leading response to the global pandemic.

Te Pūnaha Matatini is a Centre of Research Excellence funded by the Tertiary Education Commission and hosted by the University of Auckland. Over the past six years, Te Pūnaha Matatini has grown from the kernel of an idea into a diverse national network of over a hundred investigators and students who are tackling the interconnected and deeply interdisciplinary challenges of our time. Our values, expertise and focus on communication made us uniquely positioned to grapple with the COVID-19 pandemic in Aotearoa New Zealand.

Te Pūnaha Matatini’s modelling was key in helping the government make good decisions about lockdowns, particularly in April and May when the need to relax Alert Levels arrived, and in August, when a tailored lockdown was used in Auckland to eliminate a large outbreak. These public health interventions have had an immense impact on New Zealanders’ lives, not the least of which was preventing a considerable number of deaths due to COVID-19 if the virus had been allowed to spread unimpeded.

“Even I underestimated the centrality of [science] advice for me, in this time in office, and just how important it would become to us as a government.” – Jacinda Ardern, Prime Minister of New Zealand

The team made sure their models served the health system by working with Orion Health data scientists to ensure information got to where it was needed. Orion Health works with healthcare sector clients to deploy and manage machine learning models, which meant they were able to offer their technology and processes to support the Te Pūnaha Matatini team.

Te Pūnaha Matatini’s work and related research from around the globe was actively communicated to the public throughout 2020, and several of Te Pūnaha Matatini’s researchers were the most prominent science communicators during the crisis.

“I want to thank the many, many, many people in this room who were a part in your own ways in either helping us generate the information we needed to make those decisions, or who helped us communicate those decisions when it mattered most.” – Jacinda Ardern, Prime Minister of New Zealand

The transdisciplinary team working on COVID-19 that received this award brought together researchers from the University of Auckland, University of Canterbury, Victoria University of Wellington, Manaaki Whenua Landcare Research, Market Economics, and Orion Health.

The COVID-19 programme at Te Pūnaha Matatini continues into 2021 with projects focusing on branching process models, complex network models, phylodynamics, and the spread of disinformation and misinformation.

Vaccination and testing of the border workforce for COVID-19 and risk of community outbreaks: a modelling study

Vaccination and testing of the border workforce for COVID-19 and risk of community outbreaks: a modelling study

NEW RESEARCH — LINK TO FULL PDF

22 March 2021

 

Vaccination and testing of the border workforce for COVID-19 and risk of community outbreaks: a modelling study

 


Executive summary:

  • Vaccination of New Zealand’s frontline border workforce is a priority in order to protect this high-exposure group from the health impacts of COVID-19.
  • Although vaccines are highly effective in preventing disease, their effectiveness in preventing transmission of COVID-19 is less certain.
  • There is a danger that vaccination could prevent or reduce symptoms of COVID-19 but not prevent transmission. Counterintuitively, this means that vaccinating frontline border workers could increase the risk of a community outbreak.
  • In a scenario where the vaccine reduces transmission by 50%, vaccinating border workers could increase the risk of a significant community outbreak from around 7% per seed case to around 9% per seed case.
  • Until more is known about the effect of the vaccine on transmission, we recommend increasing the routine testing of vaccinated border workers to mitigate this risk. Regular saliva testing may be a good way to achieve this.
  • Careful attention should be paid to any groups, such as frontline workers’ family members, who may be vaccinated but who are not undergoing routine testing to ensure they do not become asymptomatic spreaders.

Abstract:

Australia and New Zealand have a strategy to eliminate community transmission of COVID-19 and require overseas arrivals to quarantine in government-managed facilities at the border. In both countries, community outbreaks of COVID-19 have been sparked following infection of a border worker. This workforce is rightly being prioritised for vaccination. However, although vaccines are highly effective in preventing disease, their effectiveness in preventing transmission of COVID-19 is less certain. There is a danger that vaccination could prevent symptoms of COVID-19 but not prevent transmission. Here, we use a stochastic model of COVID-19 transmission and testing to investigate the effect that vaccination of border workers has on the risk of an outbreak in an unvaccinated community. We simulate the model starting with a single infected border worker and measure the number of people who are infected before the first case is detected by testing. We show that if a vaccine reduces transmission by 50%, vaccination of border workers increases the risk of a major outbreak from around 7% per seed case to around 9% per seed case. The lower the vaccine effectiveness against transmission, the higher the risk. The increase in risk as a result of vaccination can be mitigated by increasing the frequency of routine testing for high-exposure vaccinated groups.


 

Estimated inequities in Covid-19 infection fatality rates by ethnicity for Aotearoa New Zealand

Estimated inequities in Covid-19 infection fatality rates by ethnicity for Aotearoa New Zealand

NEW RESEARCH — LINK TO FULL PDF

4 September 2020

 

Estimated inequities in Covid-19 infection fatality rates by ethnicity for Aotearoa New Zealand

 


Abstract:

There is limited evidence as to how clinical outcomes of COVID-19 including fatality rates may vary by ethnicity. We aim to estimate inequities in infection fatality rates (IFR) in New Zealand by ethnicity. We combine existing demographic and health data for ethnic groups in New Zealand with international data on COVID-19 IFR for different age groups. We adjust age-specific IFRs for differences in unmet healthcare need, and comorbidities by ethnicity. We also adjust for life expectancy reflecting evidence that COVID-19 amplifies the existing mortality risk of different groups.

The IFR for Māori is estimated to be 50% higher than that of non-Māori, and could be even higher depending on the relative contributions of age and underlying health conditions to mortality risk. There are likely to be significant inequities in the health burden from COVID-19 in New Zealand by ethnicity. These will be exacerbated by racism within the healthcare system and other inequities not reflected in official data. Highest risk communities include those with elderly populations, and Māori and Pacific communities. These factors should be included in future disease incidence and impact modelling.


 

 


 

Mathematical modelling to inform New Zealand’s Covid-19 response

Mathematical modelling to inform New Zealand’s Covid-19 response

NEW RESEARCH — LINK TO FULL PDF

22 February 2021

 

Mathematical modelling to inform New Zealand’s COVID-19 response

 


Abstract:

Between February and May 2020, New Zealand recorded 1504 cases of Covid-19 before eliminating community transmission of the virus in June 2020. During this period, a series of control measures were used including population-wide interventions implemented via a four-level alert system, border restrictions, and a test, trace, and isolate system.

Mathematical modelling played a key role in informing the government response and guiding policy development. In this paper, we describe the development of a stochastic mathematical model for the transmission and control of Covid-19 in New Zealand. This includes features such as superspreading, case under-ascertainment, testing and reporting delays, and population-wide and case-targeted control measures.

We show how the model was calibrated to New Zealand and international data. We describe how the model was used to compare the effects of various interventions in reducing spread of the virus and to estimate the probability of elimination. We conclude with a discussion of the policy-modelling interface and preparedness for future epidemic outbreaks.


 

Modelling support for the continued elimination strategy

Modelling support for the continued elimination strategy

NEW RESEARCH — LINK TO FULL PDF 

8 December 2020

 

Modelling support for the continued elimination strategy

 


Executive Summary

  • We model the effects on the risk of COVID-19 border reincursions of a wide variety of different border policies, including changes in managed isolation requirements for travellers as well as different testing regimes for frontline border workers.
  • A more detailed modelling study and risk analysis of a specific policy change would be recommended before any implementation.
  • One potential change in policy that could be considered is to replace the current requirement for 14 days in MIQ with 7 days in MIQ followed by 7 days in home isolation (including a second PCR test) for arrivals from countries with low prevalence of COVID-19 such as Australia.
  • However, any increase in the number of arrivals from high-prevalence countries, for example due to an increase in MIQ capacity or repurposing of existing MIQ capacity, will lead to an increase in the risk of border reincursions.
  • Weekly PCR testing of frontline border workers helps to ensure most border reincursions are detected before they grow too large. Supplementing this with an additional weekly rapid test would be an extra safeguard that decreases the risk of a large outbreak.

 


 

Early intervention is the key to success in COVID-19 control

Early intervention is the key to success in COVID-19 control

NEW RESEARCH — LINK TO FULL PDF 

9 November 2020

 

Early intervention is the key to success in COVID-19 control

 


Executive Summary

  • Evaluating the effectiveness of New Zealand’s COVID-19 response, relative to counterfactual (alternative ‘what-if’) scenarios, is important for guiding future response strategies. We assess the importance of early implementation of interventions for controlling COVID-19.
  • We model counterfactual scenarios in which the timings of three policy interventions are varied: border restrictions requiring 14-day quarantine of all international arrivals, border closure except to returning residents and citizens, and Alert Level 4 restrictions. We compare these to a modelled factual scenario in which intervention timings are the same as occurred in reality.
  • Key measures describing the dynamics of a COVID-19 outbreak (notably peak load on the contact tracing system, the total number of reported COVID-19 cases and deaths, and the probability of elimination within a specified time frame), are used to compare outcomes between scenarios.
  • Key measures were more sensitive to the timing of Alert Level 4, than to timing of border restrictions and border closure. Of the counterfactual scenarios, an earlier start to Alert Level 4 would have resulted in the greatest reduction in numbers of cases and deaths.
  • Delaying the start of Alert Level 4 by 20 days could have led to over 11,500 cases and 200 deaths, and would have substantially reduced the probability of eliminating community transmission of COVID-19, requring a longer period at Alert Level 4 to achieve control.

 


 

Economic comparison of the use of Alert Levels 3 and 4 for Auckland’s August outbreak

Economic comparison of the use of Alert Levels 3 and 4 for Auckland’s August outbreak

NEW RESEARCH — LINK TO FULL PDF 

21 October 2020

 

Economic comparison of the use of Alert Levels 3 and 4 in eliminating the Auckland August outbreak: a cost-effectiveness analysis

 


Executive Summary

  • We compare the economic costs of containing the Auckland August outbreak of COVID-19 using Alert Level 3 to those that might have been incurred from the use of Alert Level 4.
  • We estimate the effectiveness of Alert Level 3 using data from the actual August outbreak. The effectiveness of a putative regional Alert Level 4 is less certain, but we consider an optimistic estimate based on what was achieved in the March-April outbreak, as well as a more pessimistic estimate, which reflects the higher transmission rates observed in August.
  • We use a decision-making model for de-escalation of alert levels based on observations of weekly case numbers, which is a simpler decision-making criterion to that used in New Zealand and likely underestimates the duration of Alert Level 4 periods that would be used in practise.
  • To achieve the same likelihood of elimination, we find that both the optimistic and pessimistic Alert Level 4 period has a shorter duration than the period needed at Level 3.
  • To achieve the same likelihood of elimination, the optimistic Alert Level 4 controls have a lower economic cost than the Alert Level 3 controls.
  • To achieve the same likelihood of elimination, the pessimistic Alert Level 4 controls come at a comparable economic cost to the Alert Level 3 controls.
  • This analysis does not take into account the longer term economic costs of these measures, nor does it consider social, or health impacts that might differ between strategies.

 


 

Impact of COVID-19 pandemic on New Zealand research students

Impact of COVID-19 pandemic on New Zealand research students

NEW REPORT — LINK TO FULL PDF 

31 October 2020

 

Impact of the COVID-19 pandemic on research students in Aotearoa New Zealand

 

Early career researchers (ECRs) provide a valuable contribution to the productivity and connectivity of the research ecosystem in Aotearoa New Zealand, but are particularly vulnerable to the societal fallout from the COVID-19 pandemic.

This report – a letter prepared on behalf of Te Pūnaha Matatini Whānau, a national network of emerging researchers associated with Te Pūnaha Matatini – calls for everyone, at all levels, to find ways to support our new generation of academics at this critical time.

 


 

Simulations of re-emergence and spread of COVID-19 in Aotearoa

Simulations of re-emergence and spread of COVID-19 in Aotearoa

NEW RESEARCH — LINK TO FULL PDF 

19 October 2020

 

Network-based simulations of re-emergence and spread of COVID-19 in Aotearoa New Zealand

 


Executive Summary

  • We simulate the late July/early August re-emergence and spread of COVID-19 in Aotearoa New Zealand.
  • We use a stochastic, individual-based network model of all ~5 million individuals in Aotearoa, and run simulations for a period of 30 days.
  • Based on these simulations, we calculate: the expected time to detection of the first case after initial seed cases; the number of cases at the time of detection; the time until detection of a first case outside of Auckland; and how the overall number of cases increases without intervention.
  • Our model includes interaction pathways, referred to as ‘contexts’ in the network, broken down into network ‘layers’ representing home, work, school, and community structure.
  • Each simulation starts from initial (seed) cases corresponding to the first detected re-emergence cases in August 2020.
  • We run 50 realisations of each simulation for 30 days – each simulation scenario corresponding to one of three different levels of transmission rate.
  • To model the behaviour of individuals in the weeks prior to the August 11th re-emergence, we assume a moderate rate of people getting tested if mildly symptomatic.
  • No contact tracing or intervention is present in this scenario, other than cases that test positive being isolated to their dwelling.

 


 

Auckland’s August 2020 COVID-19 outbreak – Cabinet advice

Auckland’s August 2020 COVID-19 outbreak – Cabinet advice

NEW REPORT — LINK TO FULL PDF 

14 October 2020

 

Summary of Advice to Cabinet on Auckland’s August 2020 COVID-19 Outbreak

 

This paper summarises the modelling advice provided to Cabinet during the Auckland August outbreak in 2020 as well as detailing the methods used to provide that advice. The actual values in this report, particularly the probability of elimination and the effective R value, varied depending on the date the advice was given. Values given here are representative of those calculated in the later part of the outbreak (October 2020).

 


Executive Summary

  • For the Auckland August outbreak, the effective reproduction number Reff was found to be between 2.1 and 2.5 before Auckland moved to Level 3 on August 12 and between 0.6 and 0.8 during Level 3.
  • This was a higher value for Reff in August pre-lockdown compared to that seen pre-lockdown in the March/April outbreak. This may be due to a combination of factors, including Level 1 conditions (no gathering size restrictions, etc.), different behaviour of cases associated with international travel in March/April, higher transmission rates in winter, and differences between the communities affected.
  • Highly effective contact tracing and case isolation played an important role in keeping below 1 in Alert Level 3 and 2.5/2.
  • We estimated that it was highly likely that the Auckland August cluster was eliminated by October 5 before Auckland returned to Alert Level 1 on October 7. However, in scenarios that did not lead to elimination, case numbers grew rapidly in the absence of Alert Level 3 restrictions.