Skip to content

It's becoming a little tougher to rely on pandemic data now. Reporting from most, if not all, US jurisdictions is infrequent plus misses most of the home test results; we know individuals generally aren't going to report their home rapid test results. Furthermore, testing around the globe is likely worse, driven additionally by lack of testing resources.

So, to satisfy my craving for data I've had to turn to a bit of a jigsaw puzzle strategy to assemble data pieces into a big picture.

Friends Across the Pond, Plus Some CDC Data

Europe, though with different pandemic epidemiologic drivers, has helped to foreshadow events in the US. I turned to England's poop patrol first. The image below is one of many from the UK's excellent reporting system; focus on the blue line depicting England's viral concentration in wastewater through early March.

What you see are viral levels, mostly representing the BA.2 omicron subvariant, coming down to what was seen at the low point last October. I find this particularly encouraging because this downtrend is happening without British healthcare system overload. Furthermore, R value (reproductive number) in the UK also is heading down. (This last link is only for hardcore pandemic geeks, at the website you then need to download a spreadsheet and study the data.)

In the US, it's hard to find much about the pandemic in the lay press, probably a combination of other important news, less data, and overall pandemic fatigue. The screenshot below is from the CDC's variant tracker; note the striking and rapid appearance of BA.2.

Again, what is a hopeful sign is that we have seen BA.2 virtually take over most of the country, but without a rapid rise in healthcare resource strain. The fully assembled puzzle may be showing us that while BA.2 rapidly became the predominant strain, it did not result in a major illness surge. The next few weeks in the US will reveal a clearer picture.

But Wait, There's More

I was super-excited to see the FDA's new industry guidance for COVID-19 vaccination, the first update in about a year. Other than the vaccine industry, I may be the only other person to be thrilled to see this guidance. The press and even most of my healthcare alerts seem to have ignored it. It's pretty dense, boring reading, but the meat is in Appendix 2 on page 21 where the approach to vaccines for new variants is discussed. Although the FDA always has a disclaimer that these are all nonbinding recommendations, you can bet Pfizer, Moderna, and the other vaccine players will be paying close attention to this roadmap for future trials, likely later this year.

I deliberately chose the term "omen" at the top of this post, feeling like I may as well be reading tea leaves or using similar methods to divine the future. Nonetheless, my puzzle work today reminds me to look forward to my summer vacation with our 1000-piece jigsaw puzzle, this one with a bunch of trees that all look alike. If our plans stay intact, we'll have (and need) the whole family working on this one.

Data continue to look encouraging in terms of case rates, hospital bed availability, and other pandemic tracking across the country, notably with exceptions and still at substantial transmission levels in many areas. Let's look today at some slightly conflicting reports of vaccine effectiveness (VE) and a bit of reflection 2 years after the start of all this.

Which Numbers are Correct?

The answer, of course, is that both are likely correct, within the imperfect data analyzed. Last Monday, February 28, the New York State Department of Health posted a pre-print (non-peer reviewed) study that showed low rates of VE in children 5-11 years of age compared to 12 years and older who received a higher dose of the Pfizer mRNA vaccine. I had questions about the accuracy of the data in the study, a problem with all studies like this one that uses administrative databases. If the authors can answer those questions satisfactorily when it is peer-reviewed, it did indicate that VE wanes faster in the children who received the lower dose.

Then on March 1 the CDC released their own data looking at the same question. It also is based on administrative datasets but has the advantage of being more established and more likely to be free of serious errors. Also, getting through the CDC review process is a bit more like peer-review, though most of the peers are CDC personnel which could introduce unintentional bias. I'll focus on the CDC numbers because I think they are more reliable. Here is the bottom line for VE against laboratory-confirmed COVID-19-associated emergency department (ED) and urgent care (UC) clinical encounters and hospitalizations (H). Note that time after vaccination varies because the younger age group was authorized for vaccine only recently. Also, some estimates have very wide confidence intervals (CI) because the number of events is too small to be more precise.

Age Group/Vaccine StatusVE (95% CI)
ED/UC*5-11 yo, 2 vaccine doses 14-67 days earlier51 (30 - 65)
12-15 yo, 2 doses 14-149 days earlier45 (30 - 57)
12-15 yo, 2 doses > 150 days earlier-2 (-35 - 95)
H**5-11 yo, 2 doses 14-67 days earlier74 (-35 - 95)
12-15 yo, 2 doses 14-149 days earlier92 (79 - 97)
12-15 yo, 2 doses > 150 days earlier73 (43 - 88)
*omicron period only; **combined delta and omicron periods

VE is similar relatively soon after receiving 2 doses of vaccine, suggesting less of an effect from the vaccine dose itself. Remember that 5-11 year-olds received 10 mcg doses compared to 30 mcg in the 12-15 age group. The numbers of events for these children receiving a third dose was too low to calculate anything, but in the 16-17 year-olds a third dose seemed to produce a terrific rise in VE. Undoubtedly we'll see more reports about this from other jurisdictions as we have more time elapsed to observe VE.

Learning Now to Prepare for the Future

I'll close with a quick plug for 2 opinion items I read in the last few days. First is journalist Joel Achenbach's article in the Sunday Washington Post Magazine about 10 lessons learned so far from the pandemic. I especially noted #7: pandemics end psychologically before they do biologically. How true. Let's not get too complacent yet.

Second is a piece released this week in the New England Journal of Medicine talking about the need to develop capabilities to produce a vaccine within 100 days of the start of a new pandemic. The authors note that it took 326 days from the SARS-CoV-2 genetic sequence release in January 2020 to the emergency use authorization of the first COVID-19 vaccine. I've said before that this speed approached miracle status, so proposing lowering that to 100 days will take a bit of work. Let's hope we don't repeat past behavior and lose our research momentum when this pandemic calms down.

I must admit a bit of surprise that so many "blue" states are relaxing NPIs (Non-Pharmaceutical Interventions) for the pandemic at a time when new case rates, while falling, are still quite high. Is this all too soon? Too late?

What Do the Models Predict?

Most of us don't have the mathematical background to critique the various pandemic models, but let's look a little closer at one of them. The Institute for Health Metrics and Evaluation (IHME) is an independent population health research organization based at the University of Washington. IHME was founded in 2007 and is one of my favorite sources for data and forecasting during the pandemic.

Let's look at their most recent data compiled February 17. This is a link I'd really encourage you to explore because as you will see the graphs are interactive. Although I've set the link to open for the view of the United States, you can search anywhere.

It's important to look at several different outcomes, especially in the omicron era, to get an accurate picture. We have lots of variability in healthcare seeking and testing behavior, plus as my wife reminded me most of those rapid home test results don't get reported to any public health authority. However, hospitalization rates are a reasonable assessment of what's going on plus focus on an outcome that we care about most. Here's a screen shot of IHME's current projection for hospital bed use in the US.

This is certainly encouraging, though projections will depend on what future variants have in store for us.

Variant Alphabet Soup

Writing in the BMJ, journalist Elisabeth Mahase reminds us that the World Health Organization first designated omicron a variant of concern on November 26, 2021. A lot has happened in the last 3 months. In Pango-speak (Phylogenetic Assignment of Named Global Outbreak lineages, a software tool) it is designated B.1.1.529.1, now BA.1 for short. It accounts for >90% of cases globally currently, including in the US. BA.1 seems to have increased transmissibility but lower severity, although as we've seen it can still stretch our healthcare resources beyond capacity. It exhibits immune escape in terms of infection, though vaccine and natural immunity still provide good to excellent protection against severe disease, depending on individual circumstances.

The subvariant BA.2 (B.1.1.529.2) is a bit in the spotlight now, comprising a few percent of US cases as we hold our breath to see if it precipitates another infection wave. BA.2 certainly has a faster replication rate than BA.1 and likely is more transmissible. Although cases have been reported throughout the world, it's still a bit early to judge relative severity.

BA.3 is much less common so far. Its spike protein mutations aren't that different from BA.1 and BA.2; much remains to be learned about its ability to spread and cause severe disease.

NPI relaxation across the US leads us into a new era of the pandemic. We'll see what this experiment reveals in the coming months.

I guess we all should be accustomed to the ups and downs of the pandemic. We continue to see good news with waning of the omicron surge around the world, but parents of children under 5 had a bit of a jolt on Friday with the news that the Pfizer vaccine will not be discussed by the FDA next week as originally planned. We now await ongoing data from the trial which has begun a booster dosing phase. Regular readers of Pediatric Infection Connection will know that I oversee this trial at Children's National Hospital, but I have no knowledge of the data submitted by Pfizer to the FDA so can't provide any independent opinion.

Booster News?

Speaking of boosters, CDC released some important new information about boosters, some of which pertains to children. First, an early release in MMWR highlighted waning of effectiveness of 2- and 3-dose mRNA vaccines. This is sort of a glass half empty or full view, I was actually more encouraged by the continuing effectiveness, particularly for boosted individuals, against severe disease and hospitalization. Just looking specifically at the data from the omicron-dominant time period, vaccine efficacy (VE) in preventing emergency or urgent care visits was 87% 2 months following a third or booster dose though dropped to 66% beyond 5 months (note few data available for this latter estimate). VE in preventing hospitalization was 91% and 78% for those 2 intervals post third dose. This is by no means the final word, lots of limitations in this study and also it looked only at individuals 18 years of age and older and did include a significant number of people with immunocompromise and other risk factors for severe COVID-19 disease. No information about how these risk factors specifically affected VE.

Also, as promised the CDC issued an update to guidance for vaccination of those with moderate or severe immunocompromise to include a fourth dose in some circumstances. This update applies to people down to 12 years of age and provides a good road map for vaccination of those individuals.

Variant Viewing

I'd definitely be happier stargazing or watching butterflies and bees in season, but lately I've been keeping an eye on the BA.2 subvariant of omicron. It has started to appear in our CDC data (after clicking on this link, choose Variants and Genomic Surveillance, then Variant Proportions from the left-side menu). BA.2 now comprises 3.6% of isolates as of Feb 5 compared to 96.4% omicron BA.1) and similarly has increased slightly in the UK. As I indicated last week, it is more transmissible than the original omicron variant, but it isn't clear whether prior infection with omicron BA.1 confers some immunity against BA.2. That is probably the key factor in determining if we will see another surge due to BA.2.

I also note some encouraging news in the UK that the reproductive number is now estimated at 0.8 to 1.0, a milestone that signals flattening or decrease in the pandemic. Of course it could be just another ride on the roller coaster but I choose to take this as a further good sign.

Apparently Punxsutawney Phil saw his shadow this week, but maybe you didn't know about this important groundhog connection to the world of infectious diseases. Woodchucks (aka groundhogs) are affected by WHV (woodchuck hepatitis virus) and have been used in hepatitis B research. Unfortunately, Phil's forecasting accuracy rate for winter weather duration of 40% might be better than most models of COVID-19 future waves.

Important Conversations at ACIP

I didn't see much about this in the lay press, but the Advisory Committee on Immunization Practices met on February 4, primarily to discuss the recent FDA approval for the Moderna mRNA vaccine that will now be marketed as Spikevax. That all went fine, but I was more interested in the afternoon discussions.

  • Tracking of vaccine-associated myocarditis continues with further data from both the US and international sites. Although there are no large randomized controlled trials directly comparing the 2 mRNA vaccines, it seems likely that this risk is slightly higher with the Moderna vaccine than with Pfizer's. However, both risk rates are considerably lower in comparison to the risk of myocarditis from natural infection. I think the best way to use this information might be to advise a male adolescent or young adult, the highest myocarditis risk group, to choose the Pfizer vaccine when starting a primary series or considering a booster. The vaccine-associated myocarditis still seems to be very mild with no long-lasting sequelae, but well-designed long term follow-up studies are ongoing.
  • The other discussion I found very interesting was regarding the interval between first and second doses for a primary mRNA vaccine series. Canada and other countries now have a fair amount of data on this; in the US this has not been well-studied, and in the initial vaccine research trials virtually all of the participants had second doses at around 3 (for Pfizer) or 4 (for Moderna) weeks after the first. Based on the meeting presentations, consensus of ACIP members seemed to lean heavily towards recommending an 8-week interval between the first 2 mRNA vaccine doses, especially in times of low community COVID-19 activity where risk of infection during that more prolonged interval between doses will be less.
  • Also, stay tuned in the next few days (I think) for more specific and very helpful guidance on immunization of individuals with moderate or severe immunocompromise. This will be very helpful to those patients as well as to providers who advise transplant and other patients on this topic.

Alas, Poor Yorick

I must admit I don't know much about Denmark: it is the origin of the name for epidemic pleurodynia, Bornholm disease, and it is the setting for what is perhaps Shakespeare's best play. Now, Denmark is in the COVID-19 spotlight for 2 reasons.

First, a recent preprint shows us that the new omicron "subvariant," BA.2, has become very prominent in Denmark. The study looked at primary cases of COVID-19 from December 20, 2021, to January 11, 2022, to identify secondary attack rates in those households. Tracking of secondary attack rates ended January 18, and testing of household members was relatively high. From the 8541 primary cases in households, 5702 out of a potential 17,945 exposed household members were infected. When broken down further, the secondary attack rates for BA.1 (the original omicron) and BA.2 were 29% and 39%, respectively. This provides further evidence that BA.2 may be slightly more infectious than BA.1. Vaccination of exposed individuals didn't seem to affect secondary infection rates, but note that this included asymptomatic infections; there was no evidence of increased disease severity of BA.2.

I'm especially interested to watch how this all unfolds in Denmark, both because they are experiencing relatively high percentages of BA.2 and because they have just instituted one of most relaxed mitigation strategies anywhere.

It's hard to guess whether BA.2 has the potential to cause additional new disease surges after BA.1 across the world. Try asking Punxsutawney Phil.