Covid-19 reader: 17 September

17 September 2021 Steve Brown
A projection is not a prediction

Scientific Pandemic Influenza Group on Modelling Operational sub-group, consensus statement

COVID-19 landscape web banner_293x178The government’s autumn and winter plan for dealing with Covid-19 is built around an expanded vaccination programme and booster jabs for priority groups. But it has also indicated that a plan B will kick in if needed to ‘prevent unsustainable pressure on the NHS’. This could involve mandatory face masks, working from home and vaccine passports. No specific number has been put on daily infections or hospital admissions that would trigger the contingency plan being activated.

However, we do know a little about how the scientific experts think the pandemic could pan out over the coming months. According to the latest data from the various models overseen by the Scientific Pandemic Influenza Group on Modelling Operational subgroup (Spi-Mo), and reported to the Scientific Advisory Group for Emergencies (Sage), hospitalisations could be between around 2,000 and 7,000.

The 2,000 figure assumes a reproduction number (R) of 1.1, while the 7,000 figure is based on an R of 1.5. The current estimate for R in England is between 0.9 and 1.1, although R number estimates tend to reflect the position a couple of weeks earlier.hospitalisations.graph L

Given earlier dramatic projections from the expert committee, which did not come to fruition, some might be inclined to downplay the latest gloomy figures. However, Spi-Mo stressed that its models did not make predictions, but only provided projections of the trajectory of the epidemic if no further changes in behaviour or policy take place.

In its latest consensus statement, released this week, Spi-Mo acknowledged that the modelling produced ahead of July’s easing of restrictions had been conducted at a time when cases were increasing rapidly. But this was followed by a sharp and unanticipated fall in daily cases.

Possible reasons for this rapid change include the closure of schools for the summer, changes in behaviour around the Euro 2020 football matches, good weather and a large proportion of people isolating after being identified as a contact of a positive case.

However, it said that outturn data for hospital admissions over August did fall within expectations for some scenarios, although hospital occupancy and deaths were lower.

The statement said that even at 2,000 admissions a day, over a ‘potentially protracted period of time’, the health and care sector could face a difficult few months if combined with other winter pressures or seasonal effects.

It pointed out that current hospital occupancy of around 6,200 was only two and a half doubling times away from the peak seen in January 2021 of 34,000. Given the end of the school holidays and recent events in Scotland, where occupancy doubled in the 13 days to 7 September, Spi-Mo said it had considered how and when steps may be taken if there is sustained growth in hospitalisations.

‘A basket of measures, light enough to keep the epidemic flat, would be sufficient if enacted when hospitalisations were at a manageable level,’ the statement said. ‘If the epidemic were allowed to continue to grow until hospitalisations were at a level that needed to be rapidly reduced, much more stringent (and therefore more disruptive) measures would be needed to bring prevalence down quickly.’

Its ‘light measures’ – many of which are reflected in the government’s reserve plan B – included encouraging home working, clear messaging, more widespread testing, more mask wearing and a return to requiring all contacts of cases to isolate. However, it said there was no consensus on the measures that could be needed if the growth rate is high and hospitalisations had to be reduced rapidly.

 

Moves to vaccine equity must speed up

World Health Organization, press conference

Global health leaders this week issued an urgent call for vaccine equity globally – with Africa singled out as particularly in need of the protective jabs. The World Health Organization has set targets for 10% of the population of every country to be vaccinated by September, with 40% by the end of the year and 70% globally by the middle of 2022. 

Almost 90% of high-income countries have now reached the 10% target, and more than 70% have reached the 40% target. But not a single low-income country has reached either target. And 80% of the 5.5 billion vaccine doses administered to date have been in high- and upper-middle-income countries. And while high income countries have now administered almost 100 doses for every 100 people, low-income countries have only delivered 1.5 doses for every 100 people.Tedros L

Tedros Adhanom Ghebreyesus (pictured), WHO director general, said that just two countries in Africa had reached the 40% target and only 2% of vaccine doses delivered globally had been administered in Africa. ‘This doesn’t only hurt the people of Africa, it hurts all of us,’ he said. ‘The longer vaccine inequity persists, the more the virus will keep circulating and changing, the longer the social and economic disruption will continue, and the higher the chances that more variants will emerge that render vaccines less effective.’

And Strive Masiyima, the African Union’s special envoy for Covid-19 warned against over-reliance on donated vaccines, calling for better access to markets. While high-income countries have promised to donate more than one billion doses, less than 15% of those doses have materialised. ‘Vaccine sharing is good, but we shouldn't have to be relying on vaccine sharing. Particularly when we can come to the table, put structures in place and say, “we also want to buy”,’ he said.

He also urged vaccine manufacturers and their host countries to waive intellectual property rights to enable an increase in manufacturing – one of the proposals by the WHO to remove barriers alongside freeing up supply chains and technology transfer.

There have been continued calls for action on licensing, although such moves are far from straightforward with vaccines often underpinned by a web of intellectual property claims. An article in Nature earlier in the year explained how ‘the foundational technology needed to develop a vaccine could have been invented in an academic lab setting or start-up research firm, protected through patents, and subsequently licensed out to a larger entity for further development and commercialisation.’ Others have pointed out that patents are only part of the story, with a detailed production process also needing to be passed on and sites validated.

Astra-Zeneca, whose vaccine is being sold at cost for the duration of the pandemic, has successfully partnered with the Serum Institute of India, primarily to supply vaccine to the Indian government and low- and middle-income countries.

The calls for faster sharing of vaccines comes as the UK has announced an extension of its vaccine programme to 12-15-year-olds and to offer a booster dose to priority groups. Speaking at a technical briefing by the Joint Committee on Vaccination and Immunisation this week, where the committee announced its backing for the booster programme, deputy chief medical officer Jonathan Van-Tam said there was a ‘very strong view’ of the importance of the whole world having access to vaccines. But he said that the job given to the JCVI in terms of booster jabs was to decide what was best for the UK.

‘Nine countries, by the end of August, have already announced that they are actively starting some form of booster campaign and there is firm intelligence that 18 others are also considering it,’ he said. ‘So, the UK is not alone in thinking it will need to do this to give maximum protection to its population this winter.’

 

What’s in a denominator?

More or less, radio programme/podcast

ITV political editor Robert Peston initiated something of a twitter storm last weekend when he highlighted Public Health England statistics that showed a higher infection rate among double vaccinated 40-79-year-olds than the same age group of unvaccinated people. He was initially jumped upon by cross tweeters thinking he was comparing absolute numbers (and with the vast majority of this group now vaccinated and an imperfect vaccine, this would not be surprising). But in fact the PHE numbers are looking at the rates within vaccinated and unvaccinated groups, using the size of each specific group as the relevant denominator in each calculation.

Clearly a case for statistical detectives and Radio 4’s More or less duly obliged. The rather surprising answer appears to be that we don’t actually know how many unvaccinated people there are in England, according to presenter Tim Harford – not even close.

The size of the vaccinated community is pretty straightforward. We know the number of vaccinated people from NHS records and, as everyone will be aware, the latest vaccination total is reported on a daily basis. To calculate the size of the unvaccinated population, we then take this vaccinated number away from the total population. Simple if you have a total population figure.

However, we don’t have the latest 2021 census data yet and older figures won’t allow for the numbers who have left the country because of Brexit of Covid or those who have arrived. The Office for National Statistics does provide an estimate, but this is not the figure that PHE uses. Instead, it uses the number of people registered with GPs in England – the Nims database. This is perfectly understandable as this is the database being used to underpin the vaccine rollout. But it doesn’t necessarily provide a meaningful comparison of infections in vaccinated and unvaccinated populations.

According to More or less there are six million more people in the Nims database than the ONS think actually live in England. Mathematician James Ward told the programme that according to Nims, the unvaccinated population between the ages of 40-79 was 3.5 million in August. But this dropped to 1.5 million using ONS estimates. Using the ONS data gives a case rate among the unvaccinated of about double the rate among double jabbers. Case solved apparently.

The benefits of vaccination in reducing hospitalisations and deaths are so pronounced that choosing the right denominator is much less of a factor. As a postscript to the story, the subsequent report from Public Health England has added a footnote under the relevant table. ‘Interpretation of the case rates in vaccinated and unvaccinated population is particularly susceptible to changes in denominators and should be interpreted with extra caution,’ it says. The perils of statistics…