Doherty Modelling Interim Report to National Cabinet 17th September 2021

Citation

The modelling process was led by the Doherty Institute’s Director of Epidemiology, University of Melbourne Professor Jodie McVernon, and University of Melbourne Professor James McCaw and involves collaborative work conducted by multiple teams with specific expertise in infectious diseases modelling across six institutes and universities from around Australia.

Link to the modelling hub for a full list of collaborators

Doherty Modelling Interim Report to National Cabinet 17th September 2021. Access here.

Executive Summary

Sensitivity analyses were undertaken in response to queries raised at National Cabinet about the scenarios represented in the Doherty Modelling Technical Report and Addendum (10th August 2021);

  • We have provided assurance that delays to full efficacy of two dose completion are incorporated in our dynamic assumptions, and clarified that coverage achieved at threshold cut points is further augmented by single dose completions to that date;
  • Robustness of recommendations regarding 70% and 80% coverage scenarios to the number of infections seeding epidemics was assessed. We considered three levels of introduction in the order of low (tens, approximately 10-100), medium (hundreds, approximately 300-1,000) or high (thousands, approximately 1,000-4,500) initial cases;
    • At the 70% coverage threshold with baseline PHSMs and partial TTIQ, an increase from tens to hundreds of seeded infections results in a leftward shift of timing of the epidemic meaning that it completes within the reporting window of 180 days but does not differ in overall impact;
    • At the 70% coverage threshold with baseline PHSMs and partial TTIQ, seeding of thousands of infections shifts both the timing and peak of the epidemic significantly. Overall size is notably increased. This is because the window in time between 70 and 80% coverage is sufficient to allow early epidemic growth from high numbers, resulting in ‘overshoot’ (Figure ES1);
    • Much less impact on the overall size of epidemics is observed when these seeding scenarios (tens, hundreds, thousands) are introduced from the 80% coverage timepoint, with baseline PHSMs and partial TTIQ (Figure ES1);
    • For all of the above scenarios, infections and corresponding harms are markedly reduced by application of either 1) optimal TTIQ (Figure ES2); or 2) ‘low’ PHSMs and partial TTIQ (Figure ES3), (Table ES1).
  • Given the observed sensitivity to ‘high’ seeding infections at 70%, ongoing application of ‘medium’ PSHMs at the time of transition to Phase B is deemed prudent in such cases, at least until the 80% coverage threshold is achieved (Figure ES3)(Table ES2);
  • At high caseloads, maintenance of optimal TTIQ is unlikely to be possible. In such instances, flexibility to strengthen PHSMs generally or locally will be needed (as envisaged in the National Plan) to regain epidemic control. The required intensity and duration of measures should be informed by ongoing situational assessment of transmission and its related health impacts.

The scenarios in this report representing a single national COVID-19 epidemic are clearly (and deliberately) artificial and serve to inform high level policy strategy. Their key message is to highlight the importance of a combination of timely public health responses (TTIQ) and ongoing social and behavioural measures (PHSMs) to constrain transmission, even in highly immunised populations;

  • In reality, the national COVID-19 epidemic has been and will continue to be a ‘fire’ fought on multiple fronts. Bridging of this high-level strategy to implementation requires attention to localised risk determinants, differential impact of PHSMs, small area reporting of vaccine coverage and optimisation of TTIQ and public health responses to address focal outbreaks;
    • The next phase of modelling work will focus on addressing these issues in consultation with jurisdictions and relevant committees to define evidence based and sustainable approaches;
  • Previous findings of the relatively small contribution of the 12-15 years age cohort to infection transmission have not been reinterrogated. Current model outputs do not incorporate direct protective effects of immunising this age group, or anticipated related indirect protection of children <12 years;
    • Given the recent shift in the national immunisation strategy regarding this cohort our next work phase will include attention to immunisation coverage in school settings. We are consulting with Operation COVID-Shield to identify vaccine implementation approaches that will influence likely future coverage in schools, within their population context.

Download and read the full report

Link to the modelling hub

Related Research Areas

  • Public health research