Post Segments
Introduction
What will be the effect of discontinuing isolation of Covid-19 cases? Quarantine of an infected person is a standard public health response and is usually the easiest way to reduce transmission. Will the absence of isolation increase infections and deaths, perhaps overwhelm the NHS but increase economic activity?
Method
The analysis below uses Dynamic Causal Modelling developed by Professor Karl Friston FRS of the Wellcome Centre for Human Neuroimaging, University College London. His model of the pandemic, called the Covid-19 DCM, is designed to accommodate changes in population dynamics and virus behaviour. It does so by continually assimilating 39 separate series of data such as deaths, vaccine coverage and mobility as the epidemic unfolds. The model modifies the 54 parameters used in the model such as transmissibility of the virus and vaccine efficacy as new evidence emerges.
For each day since the epidemic started the model works out the probability of where each person is – at home, going out, in hospital or isolating. At the same time, it works out for each of us the probability of being infected, infectious, immune, vaccinated or still susceptible. As some people are infected but have no symptoms it works out if we are feeling well, have symptoms, have severe covid or are dead. Finally, it works out how many people are tested and their results. Our location and vaccine status affects how many people we meet who might infect us; being infectious or isolating affects whether we pass on the infection; our clinical state determines our chance of survival; our diagnostic status determines if we seek help or isolate. The model works out how each state interacts with each other. The model is complex but so is the way we respond to the epidemic.
Is the model any good? It is different from the unidimensional standard SEIR (Susceptible Exposed Infectious Removed) models used by the research groups providing predictions for SAGE. For them the most likely predictions of mitigated responses — i.e., what is likely to happen — are more optimistic than worst-case projections of unmitigated responses — i.e., what could happen. The DCM results lie somewhere between the two types of responses. DCM allows an interplay between the various effects of behaviour, epidemiology and seasonality that are key to the control of the epidemic. For instance, the non-mandatory response to an increase in Covid-19 prevalence is one of the factors used in the model. This provides insight into how individuals will respond to surges in prevalence—based upon responses to previous fluctuations. The model is good at predicting what will happen if things carry on as before (for details on predictive accuracy see here).
The use of a dynamic causal model (DCM) is ideal for modelling the evolution of the Covid-19 epidemic because it captures how our population reacts to different levels of perceived risk of catching and then going on to developing severe disease [1]. The model also considers the levels of acquired and vaccine induced immunity, how we test and isolate if infectious and what the effect is on the economy [2]. In this analysis of two scenarios the isolation period has been reduced by half and by 80% on 1st March 2022. Some people will be feeling ill and stay at home – say 20% so an 80% reduction in isolation is plausible.
Results
As would be expected new infections rise to levels not seen before with isolation reduced by 80%. This is notwithstanding vaccinations (Figure 1- red line).
Figure 1 – new cases of Covid-19 infection daily in the UK from February 2020 to 2023
Even deaths increase despite Omicron being less virulent ( Figure 2 – red line).
Figure 2 – deaths within 28 days if Covid-19 infection in the UK from February 2020 to 2023
The NHS will not be overwhelmed but will have to cope with more intensive care patients and move from low to moderate stress in the next quarter (Figure 3 red line).
Figure 3 – acute respiratory distress syndrome requiring critical care in UK February 2020 to 2023
The effect on the economy as measured by GDP is a fall of 0.9% in economic activity with no bounce back (Figure 4 red line).
Figure 4 – the effect of reducing isolation on GDP in the UK from 1st March 2022
Discussion
The early discontinuation of quarantine of Covid-19 cases will see a huge rise in cases, a modest rise in deaths and a moderately stressed NHS. These health effects are not compensated for by a rise in economic activity but instead by a reduction of 0.9% in GDP. The findings allow an assessment of the value of the proposed measure. There seems to be no virtue in abandoning such a simple and non-intrusive public health response.
References
- Karl Friston. Dynamic Causal Modelling of COVID-19; dashboard. [cited 24 May 2021]. Available: https://www.fil.ion.ucl.ac.uk/spm/covid-19/
- Karl Friston. Dynamic causal modelling of COVID-19 | Wellcome Open Research. [cited 12 Apr 2021]. Available: (https://doi.org/10.12688/wellcomeopenres.15881.2)
