# Imperial’s Pessimistic Model

Boris has clearly been spooked by a new gloomy modelling study by the Imperial Group which puts his road map to normality in jeopardy. How is this possible when the UK’s case rate is falling and ~ 50% of the population are vaccinated ?

Well you can trace everything back to one crucial  parameter Imperial are using in their model.

We assume the vaccines prevents to a certain extent, an infected person who is vaccinated from transmitting the virus (optimistic, although only assumed in 1 sensitivity analysis). 30% after 2 jabs and 0% after 1 dose.

So according to Imperial, vaccination only has a minor effect on transmissibility. You can still get infected even if you have been vaccinated and then infect other people without ever noticing. If this were actually true then the pandemic would never end and we would remain in permanent semi-lockdown. If instead vaccinated people can’t catch covid and don’t transmit infection to others then the pandemic will soon be over, once 75% of the population are vaccinated. That is because they assume the Kent variant is now dominant with an R0 = 4. So once  3 out of 4 people are immune R falls below 1 permanently without any social distancing restrictions. This is Herd Immunity !

So where does their low value of 30% reduction in transmission come from? It is all based on this study of  care home workers and their family members:

Effect of vaccination on transmission of COVID-19: an observational study in healthcare workers and their households

The samples from that study were as follows :

194,362 Household members : 3,123 COVID cases
144,525 Healthcare workers : 4,343 COVID cases

The healthcare workers fell into two groups A) those vaccinated = 113,253  and B) those unvaccinated = 31,272

The study found: Household members of vaccinated healthcare workers had a lower risk of COVID-19 case compared to household members of unvaccinated healthcare worker (rate per 100 person-years 9·40 versus 5·93; HR 0·70, 95% confidence interval [CI] 0·63 to 0·78).   I make that a 37% reduction in infections in households with a vaccinated healthcare worker.  Imperial use this result as direct input to their model i.e. assuming that vaccination reduces transmission by just 30% but only after the 2nd dose, and 0% after the first dose of the vaccine ! There are 12 weeks between the 1st and second doses in the UK so this makes a huge difference on Reff.

The basic assumption Imperial makes is that health care workers alone can bring Covid home and then infect others within their household. Yet this is obvious nonsense because their partners and other household members also have jobs, go shopping, and attend school etc. Perhaps some of them even work as care home staff. Imperial ignores the fact that they can catch COVID from hundreds of other people outside the household. In early February 1 in 60 people had COVID in the community and infection rates were high everywhere. So in that context a 30% net reduction in household infection really is an impressively large effect.

We can also check this. The paper was published on March 21 so the data must have been collected over a period ending a couple of weeks earlier. ONS data shows that infection rates fell from  1 in 50 to about 1 in 200 by the end of February so a fair average over January to February would be ~1.0 %.

Therefore you would expect up to 2000 cases from community infection alone in Household members. So a 30% total reduction now translates into a  78% reduction in direct  healthcare worker transmission.

If you remove this baseline of community infections then the correct figure should be more like a 60% reduction in transmission by healthcare workers after the first dose and an 80%  reduction after the second dose. If they use these figures then their gloomy predictions disappear !