The CDC director just got COVID. She got the new bivalent booster a month ago.

It's time to be honest about bad evidence and bad leadership

Rochelle Walensky just announced that she has COVID 19. She received the bivalent booster exactly one month ago at a CVS pharmacy. Right now she's probably in the window where the booster exerts the greatest protective effect it could possibly exert, yet still: look what happened.

If you ask the CDC director, “What is the vaccine efficacy of the bivalent booster you have received? What is it for any symptomatic disease? What is it for severe disease?” She won't be able to answer.

That's because the leadership at the White House has permitted this product to come to the US market without any credible evidence that it has any vaccine effectiveness. We simply don't have human randomized data for clinical endpoints. All we have is human data on antibody titers, which is a surrogate endpoint of no value in the current moment.

It will be hard for the American people to reconcile Rochelle Walensky's insistence that everyone rush to boost their 7-year-old, a decision she encourages, with her living proof that this vaccine cannot stop breakthrough infection. And knowing we know nothing about the benefit on severe disease in a child that young. Particularly one who had Covid.

In the beginning of the pandemic, in the first few weeks, it was acceptable to do things without good evidence. After all, we didn't know what would come.

But by April 2020, it was clear that a rational, scientific society would try things in the context of randomized control trials. This is the only way to separate what works from what doesn't. Many commenters, and I, argued this.

Of course, in a perfect world, observational studies could help. But immediately it was obvious that non-pharmacologic interventions were deeply polarized. Masking was tied to political affiliation, and a host of other precautionary measures. Places that decided to implement a mask mandate versus those that did not, did not do so randomly. They did so based, in part, on the valence of the region, and in part on the trends. Trying to take observational data and learn the efficacy of an intervention when it is plagued with confounders is a fool's errand.

Fast forward. We're now in 2022. The emergency phase of the COVID-19 pandemic is over. There's no reason to be taking risky gambles. We've got all the time in the world for careful studies. Those studies can result quickly if the path to products is only through the study, and we prioritize high risk populations.

Take the new bivalent booster. If I worked for the FDA, I would not approve this product for the United States without randomized trials that it can further lower the risk of severe disease. How would I do those studies?

I would demand Pfizer randomize nursing home residents. They are the highest risk population, and the most likely to have an efficacy signal. I would power the study for all cause mortality. I would leave it open-ended. In other words, the sample size can keep growing. This permits both randomization and a population vaccination campaign, saving time. We can look after a certain number of events, based on a pre-specified calculation, as is commonly done using one of several methods. If the trial is futile, we will abort mission.

Concurrently, I would run a pragmatic, open-ended randomized control trial among people 40 years and older. The trial would have a vast sample size that would permit interaction testing by age. Is it possible the booster helps people over 60 but not under 60? Over 70? The trial I envision would have power for that. Here the primary endpoint would be severe disease.

There should also be some discussion about running a trial in younger populations. But the moment you get the power calculation back from the statistician, it would be apparent that it is a fool's errand. The sample size would need to be so vast because the upper bound effect size is so poor. I would give it up. We wouldn't be rolling out bivalent boosters for 7-year-olds. In that respect, the United States would join our peer nations in Europe among places that are sensible and scientific.

Does Pfizer have the resources to run a randomized control trial in nursing homes? Powered for mortality? Absolutely, they have 100 billion reasons why that is possible. If you crack the whip, a pharmaceutical company can do anything. But if you lay down, and let Bourla do whatever he wants, they will just keep approving boosters on six mice, then four mice, then one mouse- shaped cell culture dish.

The role of government regulation is to compel private actors to generate evidence that helps people before we pay for costly therapies. We need to know the benefits are greater than the harms. And we need to know in whom the interventions work.

This White House is too cozy with Pfizer. They spend too much time talking to Pfizer and too little time talking to experts in evidence-based medicine. Sadly, their internal expertise does not seem to include medicine, evidence, or health policy.

Probably their biggest failure is that they picked the wrong people to run the show. They picked political loyalists, and not the best scientific minds.

They picked Ashish Jha because they saw him on cable news. And they saw him on cable news because he tweeted a lot. And they really liked him because—like a puppy dog—he loved the White House's decisions, even when they were blatantly negligent and clearly incompetent. He kissed up better than the best. And that's why he's running the show. Not because he's the most thoughtful scientist. His decision making suggests otherwise.

Here we find ourselves. A White House that is committed to the campaign that bivalent boosters, in young populations, is a critical step in the fall pandemic plan. They have no randomized evidence for that. They have no observational evidence for that. Their own CDC director who received this vaccine one month ago now has breakthrough infection. They have never asked Pfizer to do anything challenging, to earn the billions of dollars they’re forking over. This administration is incompetent, corporatist, and anti-evidence-based medicine.

They spend more time smearing their opponents as misinformation purveyors then they do demanding that drug companies generate information.

There have always been moments like this in history. You take advantage of the fear and the population to sell them interventions that are unproven, useless, or both.

The White House's bivalent booster campaign is a step back for evidence-based medicine, an insult to people like David Sackett, and insult to every practicing doctor who can see through their misleading rhetoric. It has revealed the failures and complacencies of the health journalists, who, because they like Biden more than Trump, have never held the Biden administration accountable for their innumerable COVID-19 policy failures.

Before we launch massive vaccination campaigns, we need good evidence that they actually benefit the people we are tasked with protecting and caring. Rochelle Walensky’s infection is just a reminder to the American people she doesn't know what she's talking about, because she has not asked for good evidence.

Nevertheless, I wish her well and hope she uses this opportunity to reflect on what the evidence should have been telling her, or at least her advisors.