All systems point to the downswing of the omicron wave which is good news for everyone. Risking being labelled a party-pooper, I hasten to add that we cannot let our guard down. Note that we have had 2 variants cause global upheaval in just the second half of 2021 plus still vast portions of the world's population lacking vaccination. It is a question of when, not if, the next variant of concern appears. (I'll rejoice if I'm proven wrong.) We need to accelerate all preventive measures including delivery of vaccines and boosters worldwide.
Continued Evidence That Vaccines Work
Last week saw publication of 4 new studies all showing the benefit of COVID-19 vaccination in protection against serious illness.
- A study released by CDC last Wednesday I think got a bit twisted by some press reports into appearing to advocate that being unvaccinated was a good plan. The study looked at surveillance data from California and New York. A summary statement, "By early October, persons who survived a previous infection had lower case rates than persons who were vaccinated alone," was the source of confusion. The confusion can be clarified by a nice graph that unfortunately only appeared as an online supplement to the report, meaning that readers would have had to read through some pretty dense tables of numbers or take an extra mouse click to get a good appreciation of the situation. Here is one of the supplementary figures from New York. Look at the distance between the solid blue line (unvaccinated) and the 3 lines at the bottom. Yes, it is true that vaccinated individuals with no prior infection history had slightly higher hazard rates for lab-confirmed infection, but not that much more compared to the rates of vaccinated or unvaccinated people who had prior infection. Being vaccinated was far better than being unvaccinated regardless of prior infection history. Also, these data are mostly before the omicron surge when everything changed, and it did not look at booster status.
- The second study did evaluate booster status and was highly encouraging. The authors, from the CDC, used the familiar test-negative study design and were able to judge the association between 3 doses of either the Pfizer or Moderna vaccine with protection from symptomatic COVID-19 disease against both delta and omicron variants. This study design only permits calculation of odds ratios rather than true relative risks but can be used as an approximation. Comparing 3 doses of an mRNA vaccine to being unvaccinated, odds ratios were 0.33 (95% CI 0.31-0.35) for omicron and 0.065 ((0.059-0.071) for delta, i.e. significantly in favor of vaccination. Similarly, a comparison of 3 versus 2 doses of vaccine showed odds ratios of 0.34 (0.32-0.36) and 0.16 (0.14-0.17) for omicron and delta, respectively. We'll need to see a follow-up to these data since the study extended only through January 1, 2022.
- Another study from CDC compared disease incidence and death rates among unvaccinated and fully vaccinated adults, with and without boosters in 25 sites in the US. Significant benefits were demonstrated for the vaccinated and boosted groups, especially for ages 50 years and above.
- Also worth noting is a study focused on benefit of third doses in preventing emergency and urgent care visits and hospitalizations in 10 states, covering the August 2021 through some of January 2022. As you might guess by now, the data lend further evidence of the benefit of vaccinations and boosters.
Ten Fingers Ten Days
Only those who have suffered through a clinical rotation with me on the infectious diseases service at Children's National Hospital will recognize the heading above. It is law #5 in Bud's Laws, a list of 10 aphorisms that I've been tweaking the last few decades. It refers to the fact that most antibiotic treatment durations we use are based on little to no evidence. I've maintained that if we were an animal species with 12 fingers we'd be treating everything for 12 days instead of 10.
Enter a new study, the SCOUT-CAP trial. (I won't even mention the origin of such an acronym, it's too long and too forced into forming the acronym itself. Other acronyms in the study are RADAR and DOOR; very cute.) This was a randomized double-blind placebo-controlled trial of 5 versus 10 days of oral antibiotic therapy for outpatient community acquired pneumonia taking place in 8 US cities. It included 380 children 6-71 months of age diagnosed with CAP and initially prescribed one of 3 commonly used regimens (amoxicillin, amoxicillin/clavulanate, or cefdinir) to treat CAP. Children were continued on their original therapy, then randomized to continue that treatment or switch to placebo on days 6-10. (Note that over 90% of the children received amoxicillin originally.) Fewer than 10% of children overall had an inadequate clinical response regardless of treatment assignment, probably indicative of the fact that most children had viral infections and never needed an antibiotic in the first place. However, the 5-day course individuals had better scores for resolution of symptoms and antibiotic-associated side effects, as well as a lower incidence of antibiotic-resistance genes detected in throat swabs obtained in a subset of the participants.
You might have correctly concluded that if the main thing about the study that I'm complaining about is use of acronyms, the study itself is pretty good. It's very difficult to design a study of mild CAP that both applies to real-world practice and has enough safeguards to prevent bias from study design, and this study achieved those goals.
I'm not suggesting we all amputate 5 of our fingers, but please think about shortening courses of antibiotics for outpatient pneumonia in relatively healthy children.
[Note this section was updated January 26 to clarify that this was basically a study of 5 versus 10 days of amoxicillin therapy, too few children received amox/clav or cefdinir to comment.]