More vaccines to Oslo could save more lives – VG

On Tuesday, the answer came from the assessment that FHI has been working on for a long time: They are in favor of adjusting the vaccine distribution and administering a few more doses in certain heavily infected areas.

There is an answer that many have been waiting for: the public debate on vaccine distribution has stalled since January. The Oslo region has demanded multiple doses for areas with high infection, while others disagree. At the same time, no one has known exactly what FHI has calculated: it has been ad-free.

For several weeks, FHI has been working on running mathematical vaccine models again. The last time they did it was before Christmas. They have analyzed how different strategies can affect the goals of limiting hospitalizations and deaths at the national level.

One of the results The models What has been published now shows that we could have prevented 80,000 infected and saved 108 lives in Norway in 2021 with a more contagious variant of the virus, if Oslo had received 20 percent more vaccines.

All models work with uncertainties. Try it for yourself in the vaccine calculator below in the case!

But the strategy that the government and FHI have followed is not one of those that are counted in the models. When FHI has finished, they have performed a new calculation. According to the report, it can save 1 to 5 lives between March and June.

These are the differences

• FHI has advocated for what they describe as a “modest and skewed distribution” of vaccines: they want to increase vaccine doses by 20 percent in some districts and municipalities of Oslo with a large proportion of hospital admissions over time, and take up to three percent of the doses from municipalities that have had few hospitalizations.
• The calculation is made by calculating the number of admissions that can be avoided among those who have been vaccinated. Hospital admissions in the last six months were chosen as a measure of disease burden over time. This strategy is not taken into account in the models and the calculation does not take into account the indirect effect, that is, the risk of becoming infected decreases when more people have been vaccinated around it.
• What is raining on The models It is more complete. For example, 20 percent of vaccines are distributed in Oslo or Oslo and Viken. Numbers analysts have looked at several possible developments in the infection and how new, more infectious variants are spreading.
• Models are developed at the county level, not at the municipal level. The indirect effects of vaccination have been taken into account.

The model concludes that it has the effect of reprioritizing doses, but that the effect is not much greater if you reprioritize more than 30 percent.

If you look at the results of the models, a redistribution to Oslo can provide several benefits, but if you re-prioritize too much, the result can be negative:

In all model scenarios, deaths and hospitalizations are reduced nationally when more vaccines are administered in Oslo until all people over 65 and healthcare workers have received a vaccine.

– The model shows the consequence of a choice on the vaccine distribution, so FHI has chosen to recommend a less drastic redistribution choice than the one we have modeled, says Birgitte de Blasio, who leads the FHI modeling team.

She emphasizes that the modeling group’s task is not to give advice. It is to provide reports that are interpreted and included as part of a summary report where NIPH provides recommendations to the government.

– One way of thinking is that when models show positive health effects by prioritizing vaccines for high infection areas, it is an indication that they will likely respond regardless of the chosen strategy.

-A risk

The main FHI report states that to some extent so far “the emphasis has been placed on individual ‘demand’ for individual protection rather than societal demand for the most possible ‘economical’ use of available vaccine doses”, and that this is also reflected in ethical and legal considerations. But they believe there are several arguments in favor of a geographic targeting of vaccines over a period.

When asked why FHI does not choose a strategy with more redistribution, FHI Director Camilla Stoltenberg responds that it is because there is a risk if other counties receive significantly less and vaccinate their elderly more slowly.

– Therefore, the inhabitants of these municipalities are more vulnerable if the spread of the epidemic changes significantly. We choose a model with a more moderate change in the distribution key so that other municipalities get almost the same as with the original distribution key.

She says county-level models are coarse mesh, but there can be big differences within a county. Therefore, FHI has also performed the calculations at the city and municipality level.

– Both these simple calculations and the modeling are subject to uncertainty, but point in the same direction, namely that it appears that the skewed distribution to areas with high infection pressure / disease burden over time can result in a number reduced income nationwide. So it remains true that the situation in the country is generally unstable, he says.

He also notes that the models operate with a “constant” increase or decrease in infection in counties and therefore do not take into account that outbreaks can occur in large municipalities.

– The risk of such outbreaks was emphasized and considered highly probable, especially since the epidemic was characterized by more contagious virus variants. This would represent a significant risk that the model has not taken into account. Therefore, after a general assessment, we wanted to carry out only a moderate redistribution of vaccines between municipalities.

She says that FHI believes that the model alone did not provide a sufficient basis for changing the distribution key so extensively.

– We have chosen a solution that is in line with the results of the model, but after a general evaluation we have chosen to reduce the amount of vaccine doses that we redistribute to 2-4 percent.

– It is not certain that this is the optimal model. Much will depend on how the epidemic plays out in the weeks and months to come, where we have outbreaks in the future, and how quickly we now get larger amounts of vaccines so that we can offer vaccines to everyone. The latter is the most important.

I got results two weeks ago

According to the models, a prioritization of Oslo can result in a “sharp decrease” in deaths in Oslo, but at the same time an increase in other counties. Overall, this means fewer deaths in Norway.

– The models show a global effect of vaccination, both directly and indirectly. That is why you have mathematical models because they simulate the spread of the infection. Your risk of becoming infected depends on how many infected people you have around you and whether you have been vaccinated, explains Birgitte de Blasio.

The results, which show a positive effect of skewed distribution up to a certain percentage, have not changed during this new model, he says.

– When did you get these results for the first time?

– It was probably the week before we delivered the report. Before that, we spent time developing the models and adapting them, says de Blasio.

The last time the FHI group of models presented a report on skewed distribution was in December. The report, which served as the basis for the strategy that has been followed thus far, contained the results of their calculations and other FHI evaluations.

The figures then showed that a skewed distribution could yield positive health outcomes in the form of fewer deaths and hospitalizations, if the spread of infection (R) nationwide is low.

– So what you have found now corresponds to what you have looked at previously?

– Corresponds yes. But before Christmas, it was a completely different and more drastic approach. Then we pretend that all vaccines should go to some areas. And in the simulation we did before Christmas, we had little information about the effect of vaccines; now we understand a little better how vaccines work.

What was defined as a regional priority at the time was to vaccinate in up to three counties each month, depending on where the infection is most spread until the R number is below one. It was weighed against prioritizing all counties equally.