This year, for the first time in recent memory, parents and doctors have a choice to make. Two vaccine advisory groups — the Centers for Disease Control and Prevention (CDC) and the American Academy of Pediatrics (AAP) — issued conflicting recommendations on which influenza vaccine to give to children.

Why?

We'll start at the beginning.

In 1947, Thomas Francis, working at the University of Michigan, took influenza virus, grew it in eggs, purified it, and inactivated it with a chemical. The final product was administered as a shot. The first year, the vaccine worked pretty well. The next year it didn't. The reason: natural influenza virus mutates from year-to-year. As a result, a new influenza vaccine—one that accounts for the new circulating strains—has to be given every year.

Whereas most vaccines are about 90 percent effective, the influenza vaccine, at its best, is only about 60 percent effective. For that reason, ever since the days of Thomas Francis, scientists have been trying to make a better influenza vaccine. Fifty years passed before they did. Again, it was invented at the University of Michigan. This time by a man named John Maassab.

Instead of purifying influenza virus and killing it with a chemical, Maassab weakened it by adapting it to grow at colder temperatures. This live, weakened virus could reproduce itself in the nose (which is about 2 degrees cooler than core body temperature), but not in the body. The vaccine was administered as a nasal spray. Its trade name was FluMist.

The rise of FluMist has been as dramatic as its fall.

In 2003, FluMist was licensed for people ages 5 to 49. Both the CDC and AAP recommended using either FluMist or the inactivated vaccine. Neither group expressed a preference for one vaccine over the other.

In 2007, the license was extended to children as young as 2 years old.

In 2014, because three separate studies had shown that it was better, both the CDC and AAP recommended FluMist for children instead of the inactivated vaccine. At the time, countries such as Canada, Israel, Germany, and the United Kingdom had already made that choice.

Then, the bottom dropped out.

During the 2013-2016 seasons, FluMist was significantly less effective than the inactivated vaccine. Against one particular strain, the so-called H1N1 strain, it wasn't effective at all. For this reason, both the CDC and the AAP recommended against using FluMist for the 2016- 2018 seasons. Not only had FluMist lost its preferential recommendation, it had lost its recommendation entirely.

What happened? Why had FluMist been so effective before 2013 and so ineffective after? When FluMist is sprayed into the nose, the four viruses contained in the vaccine need to reproduce equally to induce a proper immune response. If one of the strains doesn't reproduce well, then it's not going to induce protective immunity. As it turned out, the H1N1 vaccine strain used in FluMist after 2009 didn't reproduce very well. After the manufacturers found the problem, they substituted a heartier H1N1 vaccine strain. With much better results.

Studies done with the new FluMist formulation have shown that:

  • The new H1N1 strain reproduces itself well in human nasal cells in the laboratory.
  • The new H1N1 strain spreads readily from one cell to the next in the laboratory.
  • The new H1N1 strain reproduces itself well in children.
  • The new H1N1 strain induces an immune response in children similar to that found when FluMist clearly protected against disease caused by that strain.

The problem, apparently, has been solved.

Nonetheless, the CDC and AAP disagree on how to interpret this new information. Whereas the CDC states that children can now be immunized with either FluMist or the inactivated vaccine, the AAP prefers the inactivated vaccine, stating that FluMist should be given only to children who absolutely refuse the shot.

Because many children are uncomfortable getting a shot, and because some doctors' offices probably won't stock FluMist following the AAP's preference for the inactivated vaccine, fewer children might be vaccinated. Indeed, before it was no longer recommended, about two-thirds of children given an influenza vaccine received FluMist.

The influenza vaccine remains our best weapon in the fight against this devastating infection. Even though last year's influenza vaccine was only 42 percent effective, it still prevented 84,700 hospitalizations, 2.64 million medical visits, and 5.29 million influenza illnesses. For those who didn't get the vaccine, the efficacy was 0 percent.

The question in medicine isn't when do you know everything; it's when do you know enough. The CDC is comfortable that we now know enough to recommend FluMist. The AAP, by waiting for more data, might cause some children not to get a needed vaccine, in which case their hesitancy — in the name of being cautious — might inadvertently do more harm than good.

Paul A. Offit, MD, is the director of the Vaccine Education Center at the Children's Hospital of Philadelphia. Although he is a member of an FDA vaccine advisory committee, the opinions he expresses are his own.