Too soon to call Omicron a global game-changer

Early reports from South Africa and the United Kingdom have generated concerns about the high rate of breakthrough infections among those who have been fully vaccinated and the significantly increased infectivity of this new variant but led to cautious optimism that the fatality rate from Omicron could be much lower than for earlier variants. Still, not all the early experience has been positive.

Although it is still too early for most of the Omicron infection waves to have generated significant numbers of deaths, it appears that the combination of higher numbers of infections (especially among the vaccinated) combined with lower severity per infected individual may well result in excess mortality remaining at elevated levels through this next phase of the pandemic. We highlight key signals and our current observations below.

There is very clear evidence that the Omicron variant is many times more infectious than earlier variants of the SARS-CoV-2 virus. In particular, Omicron is exhibiting high breakthrough rates in vaccinated populations and those with immunity acquired through prior infection.

A study of COVID transmission in Danish households conducted in December 2021 as both Delta and Omicron were circulating in the community showed that (after risk-adjustment) Omicron was much more likely than Delta to be passed on to vaccinated members of the same household (see table below). The infectivity of Omicron for unvaccinated household members was only slightly higher than for Delta. In fact, for households with Omicron infections, there was little difference in infection rates between unvaccinated and fully vaccinated individuals, although those who had received booster vaccinations were roughly half as likely to be infected with Omicron ⁱ.

The velocity at which the Omicron variant spreads can also be seen by how quickly it has replaced Delta in several countries. Although Omicron is now dominant across Europe and North America, as recently as early December, the majority of newly diagnosed cases (outside of Southern Africa) were still Delta infections with Omicron infections overtaking Delta only towards the middle of December. It is worth bearing this in mind when looking at mortality trends, which tend to lag infection trends by 2-3 weeks. Many of the hospitalisations and deaths in early to mid-December would have been from the Delta variant, as new infections from Omicron were only starting to ramp up at that time.

On the positive side, there is also emerging evidence that the Omicron variant is less likely to result in severe disease than earlier variants. To date, there have been a number of studies from South Africa and the UK (with a handful emerging in recent weeks from other countries) that suggest that the proportion of Omicron infections that result in hospitalisation is 20% to 70% that of Delta infections ^{ii iii iv v} (after multivariate adjustment for key risk factors). The wide range in reported outcomes can be partly attributed to differences in the study populations, particularly in relation to vaccination status, prior infection status, observation period and severity of the study endpoint.

Based on a handful of studies published to date (many of them not yet subject to full peer review):

• In unvaccinated populations, there is a greater reduction in risk of hospitalisation for those infected with Omicron vs those infected with Delta;
• In vaccinated populations, while the risk of hospitalisation with Omicron is still lower than for unvaccinated populations, it appears that the reduction in risk of hospitalisation from Omicron versus Delta is smaller than in unvaccinated populations; 
• Adjusting for prior COVID infection also narrows the difference in the risk of hospitalisation with Omicron compared to Delta (a higher proportion of those infected with Omicron have had a prior infection compared to those infected with Delta);
• However, there are also promising hints that hospitalisations from Omicron tend to be shorter than for Delta and the reduction in the risk of ICU admission for Omicron versus Delta is larger than the reduction in risk of any hospitalisation, hinting at a lower probability of severe outcomes from Omicron compared to Delta ^vi. 

Our key metrics of interest are the mortality and the infection fatality rate. At this time, there is very little reliable information to date on mortality impacts, with most Omicron outbreaks being too recent for credible mortality outcomes to be measured. Studies published to date show very wide confidence intervals for mortality outcomes from Omicron. Death trends lag behind case trends by 2-3 weeks, so reports coming in the next few weeks will be instructive.

There is also some evidence of a supporting biological rationale for these observations with the Omicron variant behaving differently compared with other variants. Reports on initial lab studies on animals and tissue samples are suggesting that Omicron has much higher viral replication rates in the upper respiratory system around the nose and throat (driving higher infectivity), but lower viral replication rates in the lower respiratory system deep in the lungs (driving lower mortality). ^vii 

Early evidence from South Africa and the UK is encouraging, suggesting that mortality outcomes from Omicron might be even better than the early hospitalisation outcomes, with the increase in deaths being significantly smaller, so far, than the increase in new diagnoses. Despite apparently lower fatality rates from Omicron, the higher infectivity of Omicron is now leading to higher daily death toll in the UK than during the peak of the Delta wave, although still considerably lower than the peaks in 2020 and early 2021 in largely unvaccinated populations. 

So far, the takeaway from the South African and European data is that Omicron has caused record breaking case numbers, but without a similar rise in the rate of hospitalisations and deaths.

An array of high vaccination rates, booster programmes (especially in the elderly and at-risk groups), recovery immunity, antiviral & monoclonal antibody treatments and some mild social distancing restrictions appear to have combined with a more intrinsically mild variant to decouple the link between infections to hospitalisations to death.

The number of new infections is already falling in the UK, with deaths only recently starting to rise slowly but by much less than the increase in the number of infections. Similar patterns are being observed in other European countries (with a short lag). Questions remain as to whether this pattern will also be observed in the US. New York City, the first metropolitan region in the country to experience a spike in Omicron cases, has so far seen a notable rise in deaths and cases (mostly driven by the unvaccinated) towards the end of 2021 although some of this may be the lingering impact of Delta infections which were still common in the first half of December. The US in general has seen the death rate almost double from November 2021 to early January 2022. 

Over time, we will gather more evidence on the severity of Omicron; for now, the data is very limited and significant uncertainty remains. Accurate estimates of Omicron severity (compared to earlier variants) are confounded by various factors. These include the fact that underlying vaccination rates are higher now for the Omicron outbreak compared to earlier outbreaks, that early new infections with Omicron tend to occur in younger people, on average, than earlier outbreaks. Countries with the earliest Omicron outbreaks (South Africa and the UK especially) have relatively high numbers of people who have acquired immunity from prior infections and estimates of Omicron severity in these countries may be less applicable to countries like Australia which are now experiencing a severe wave of Omicron infections with high vaccination rates, but with very low proportions of the population having had a prior infection. 

Further uncertainty in relation to mortality outcomes is driven by emerging constraints on healthcare capacity (potentially leading to excess mortality due to both COVID-19 and non-COVID -19 causes of death). This comes as a result of record infection rates in the general population as well as high infection rates among healthcare workers, potentially limiting the ability of hospitals to provide adequate care to patients. The measurement of hospital outcomes and the ability of hospitals to provide effective care is also potentially distorted by the high numbers of patients who are admitted for non-COVID-19 reasons, but who are testing positive for COVID-19 upon admission (because of the very high underlying infection rates in the general population). Although antivirals and monoclonal antibodies are now available in principle in many countries in Europe and North America, they are not always available in practice. Additionally, certain monoclonal antibodies, such as Sotrovimab (made by GSK) are being tightly controlled as they have been shown to be particularly effective against Omicron and should therefore be reserved for patients who require them. 

Given these uncertainties, and considering the high case load, our outlook at present for insurers is that COVID-19 excess mortality is likely to remain elevated.

Further out in time, Omicron could lead to a lower infection fatality rate than seen so far. The lower severity of Omicron, combined with higher transmissibility and greater breakthrough infections, will lead to larger proportions of the populations with immunity acquired from infection with Omicron. Vaccine manufacturers are planning Omicron-specific vaccines in the coming months, which could reduce the breakthrough transmission rate currently seen.

One final feature of Omicron worth noting has been its speed. It took months for both Alpha and Delta variants to spread globally while Omicron managed to become dominant in several countries within weeks of initial detection. With Omicron, it appears that everything is happening more quickly: faster transmission, shorter incubation periods, quicker discharges from hospital, and perhaps even a shorter period of acute infectivity. This could result in Omicron's impact being compressed into a much shorter timeframe than previous variants – which augurs well for severity and societal disruption. This could also mean that the severity of Omicron is even lower than implied by early Omicron studies, since these studies typically compare hospitalisations and deaths assuming that the progression of Omicron infection (for example, time from infection to hospitalisation or from infection to death) is the same as for earlier variants.

Irrespective of how severe Omicron finally turns out to be, it remains a dangerous disease. Although likely to be milder than Delta, the much higher infectivity and greater ability to produce breakthrough vaccination makes it by no means a "mild disease" – yet.  As the global community continues to get vaccinated and take advantage of medications to reduce the severity of disease, we hope to see COVID-19 become a much milder but likely endemic disease someday soon.