Statistical Bamboozlement with Shingles Vaccines

In this skill-testing question, tell me which of these statements is true: 

1. The shingles vaccine is 90% effective in preventing shingles. 

1. The shingles vaccine prevents shingles in ONE in every 32 people vaccinated. The other 31 people see no benefit.

Answer: both are true.

Guess which statement dominates the marketing of Shingrix, the most popular shingles vaccine in America?

90% effective. 90% effective. 90% effective.

That’s the main marketing message coming from the Centers for Disease Control and Prevention discussing the efficacy of the vaccine. So common is this sort of statistical sleight of hand that the CDC doesn’t even seem to see any real problem in how those effects are presented to the public. Because they are so impactful, and are essentially marketing messages, and they need to be deconstructed. 

We can look to professional health reporters working for the mainstream media to deconstruct these numbers and explain what they all mean, right? 

Hmm. Not quite. 

I was in for a surprise when I went to see what the New York Times had to say about this allegedly 90% effective vaccine. Remember, this is a news outlet giving us all the news that’s fit to print. 

Earlier this year the New York Times published a remarkable article describing associations between certain vaccines and aging, headlined: “Vaccines Are Helping Older People More Than We Knew.” It starts out boldly:

Let’s be clear. The primary reason to be vaccinated against shingles is that two shots provide 90 percent protection against a painful, blistering disease that a third of Americans will suffer in their lifetimes, one that can cause lingering nerve pain and other nasty long-term consequences.

Let’s “be clear” alright. 

The New York Times reporters and editors are smart people, smart enough to make sure their narrative about the use of vaccines in older people flows in the “right” direction. It’s relatively straightforward to produce a plausible bit of vaccine cheerleading when you don’t have to actually critically examine the numbers someone hands you. Thus, the story follows a predictable direction honed to please the many drug and vaccine advertisers whose ads are keeping the Times afloat. 

In this case, the copy follows a standard vaccine-reporting template we’ve come to expect: It overstates a vaccine’s benefits, downplays or ignores its harms, and hammers home the dreadfulness of the disease under the illusion of certainty and medical consensus. As I’ve discussed in an earlier Brownstone article, the net result of this kind of biased reporting is essentially to broadcast the “Ask your doctor” tagline to the unwashed masses from one of America’s biggest pulpits. 

Walt Whitman the great American poet, could have been speaking of the NY Times when he wrote of sounding his “barbaric yawp over the roofs of the world.” 

We’re talking herpes zoster here, commonly known as shingles, and caused by a reactivation of the varicella-zoster virus (VZV), the same virus that causes chickenpox. 

It can be nasty, but not always. It typically involves tingling, burning, itching, or pain along a single nerve pathway, sometimes with a blistering rash which typically heals over in 2 to 4 weeks. There are quite useful and relatively cheap antiviral medications to help speed recovery if taken soon enough. More serious is when shingles develops into long-term nerve pain called postherpetic neuralgia (PHN). Sometimes it gets into your eyes, which can be truly nasty. But thankfully it’s rare and most shingles cases clear up on their own within a month. 

Going through a succinct independent analysis of the randomized trials of the major shingles vaccines provides a unique case study for determining how actual science can become so contorted by professional reporters at the New York Times.

So where does the “90% effective” come from?

Whenever you see a number this big you know that it is a relative number. Throw it out and start again I always say. 

Relative risk reductions such as “90% effective” are promotional shorthand, a misleading description of what an individual drug or vaccine recipient can expect. 

Here’s what people think of the words “90% efficacy:” if 100 people get vaccinated there will be 90 fewer cases compared to a group of 100 people who didn’t get vaccinated. That’s how most peoples’ brains will interpret the message being conveyed. And if you’re facing the chance of getting a painful blistering rash, you might say, why not? 

If you go to buy something for $100 and it has a “90% off” sign on it, what will you pay? Ten dollars right? That’s not how it works when describing the effects of drugs or vaccines. 

The 90% describes the proportional reduction in disease among vaccinated compared with unvaccinated people during a trial. Typically the big number hides the crucial fact that the baseline risk in this case is incredibly low. 

In the two major trials of Shingrix, involving one group of 50+ patients and another of 70+ patients, shingles occurred in the placebo group at a rate of 3.4% over 3.5 years. In the Shingrix group the rate was 0.3%. This is among over 16,000 patients, so a big enough cohort to draw some conclusions. 

So when you drop from 3.4% to 0.3% that’s a “90% drop” right? 

Put it this way, if you had $3.40 in your pocket and a robber stole 90% of that, what would be left? That’s right, 30 cents. 

When you tell your friends “Robbers stole 90% of my money!” they would conclude that you’re now in the poorhouse, because you likely lost thousands of dollars. 

Not so, you’ve only lost $3.10. Barely pocket change. 

In the case of the vaccine the difference between those vaccinated with Shingrix and those vaccinated with a placebo is 3.1% (subtract the two rates). 

Stated more honestly, vaccinating 100 people over almost four years prevents roughly 3 cases of shingles; This means that you would have to vaccinate 32 people with Shingrix to prevent one case of shingles, over almost four years. This term, NNV or Number Needed to Vaccinate is key to better determining your odds of whether a vaccine might be worth it to you. Your chances of benefit are one in 32. 

There is one huge caveat to these numbers: we assume that these data have been collected without any fraud, misrepresentation, or fabrication. As far as I can tell there has been no fraud, misrepresentation, or illegal activity reported regarding the clinical research done on the Shingrix vaccine, even though the company making and studying it has faced historic healthcare fraud settlements for other drugs in the past. Just saying. 

So back to the “90% effective.” 

I have a pretty strong reaction with these sorts of numbers games because without telling you what the baseline risk is, the “90% effective” essentially overstates the effects of the vaccine. We already saw this rodeo trick around the Covid vaccines, with the manufacturers and the CDC declaring them “95% effective” when the absolute likelihood of being injured by Covid was astronomically tiny. And therefore the absolute benefit of the mRNA vaccines was also very tiny. 

Which brings us to the second problem of these sorts of data, exaggerating the baseline risk. On that, let me ask you: 

What is the lifetime risk of death for the average American?

It’s 100%. Sorry to be the bearer of bad news. 

That number becomes a lot smaller when you consider your annual risk of death. There might be drugs or vaccines or surgeries that might delay that inevitable conclusion, but nothing will ever reduce your lifetime risk of death to zero. 

Your lifetime risk will always be, and forevermore 100%.

Let’s keep this fact in mind when we say that 1/3^rd of Americans will get shingles in their lifetime. This figure is misleading as it forgets to suggest the corollary: that 2/3rds (ie: 66%) of Americans will never, ever get shingles in their lifetime. 

A case of the glass is half-full. “One-third lifetime risk” also exaggerates what your annual risk is. For the average American over 50 your risk in any given year of getting shingles is between 0.3 and 1%. It’s a bit more when you get older, but, in absolute terms, not much. 

Another way to say this: 

“Boomers and Gen Xers, listen up: In any given year, your chances of NOT getting shingles is more than 99%.”

Let that sink in for a second.

If you are the kind of person who thinks that going to Vegas to win big or buying lottery tickets are both good strategies to fund your retirement plan, I’d say you need to get better at understanding probabilities. 

Don’t feel bad; most people don’t think in terms of probabilities, and strangely they don’t consider probabilities concerning their health. Clearly most people think: “My mom had shingles and it was really nasty. Therefore I don’t care if the vaccine gives me a one in 32 chance of “winning.” Fair enough. Roll the dice, hope you win.

As for the cost of these vaccines. They are not cheap, and never “free.” 

Last time I checked, vaccines aren’t ever “free.” Someone pays for them. The money comes from the patients’ pockets, your employer health plan, or government’s coffers. The real question is, is it ultimately worth it? 

Let’s recap:

Shingles and shingles vaccines, by the numbers: 

• 0.3 to 1%: Per-year chance the average American over 50 getting shingles. Basically, the likelihood you WON’T catch shingles every year is over 99%. This increases slightly as you get older, but not by much. 

• 32: Number of people over 50 getting vaccinated with Shingrix to prevent ONE shingles case (over 3.5 years). This NNV or Number Needed to Vaccinate means the vaccine has no effect for 31 out of 32 people vaccinated. 

• 333: Number of people over 50 who need to get vaccinated with Shingrix to prevent ONE case of Postherpetic Neuralgia (332 won’t avoid a case of PHN—because they were extremely unlikely to get it in the first place)

• 11 to 25: Number of people over 50 who get the vaccine who end up with ONE case of systemic reaction (effects include fatigue, weakness, headache, shivering, and stomach problems). This is called NNH or Numbers Needed to Harm. In other words, between one in 11 and one in 25 people vaccinated will experience these side effects and possibly be out of action for several days. There is a small increased chance of getting Guillain-Barré syndrome (GBS), a rare neurological disorder where your immune system mistakenly attacks your peripheral nervous system. 

• $400: The average cost someone in the US would pay, without insurance, for the two-dose series of Shingrix administered two to six months apart. Obviously, Medicare, Medicaid, and private employer health insurance plans would pay less. 

• $12,800: Total cost of the shingles vaccine to prevent ONE case of shingles in people over 50 (over 3.5 years). 32 times $400.
• $132,800: Total cost of the shingles vaccine to prevent ONE case of PHN in people over 50 (over 3 years). 332 times $400.
• 125.5 million: Number of American citizens over 50 (about 36% of the US population)
• $1.6 trillion: The cost of vaccinating every American over 50 with the shingles vaccine (paying the uninsured rate). 

I hope you follow my math. 

Giving a “free” shingles vaccine to every American over 50 might cost consumers, Medicare Part D, US private employer plans, and Medicaid $1.5 trillion if everyone got it. 

Remember, this is a disease that 99% of the population will never get in any given year, where bothersome vaccine side effects are much more likely than any of shingles’ rare complications. 

We will never have affordable healthcare when this volume of money gets thrown at small problems.