Advertisement

SKIP ADVERTISEMENT

Why the Vaccine Safety Numbers Are Still Fuzzy

Putting a risk in context: The rate of blood clots is extremely low, but the pause in the Johnson & Johnson vaccine could reveal more cases.

Image
A mass vaccination site in Elgin, Ill., closed for a day because of the pause put on the Johnson & Johnson vaccine.Credit...Rick West/Daily Herald, via Associated Press

When federal officials paused administration of the Johnson & Johnson vaccine after six cases of a rare clotting disorder, one fatal, among the 6.9 million people who had received the vaccine, many critics noted that the chance of a serious ailment was so rare as to be negligible — less frequent than being struck by lightning.

But that roughly one-in-a-million rate is far from certain. Doctors may ultimately find the vaccine is not responsible for the ailment. However, if the two are linked, it’s also possible that the chance of an adverse effect will be higher, even if it remains low.

“Numbers seem quite solid, like, ‘Oh, it’s 10,’” said Caitlin Rivers, an assistant professor at Johns Hopkins University, who studies infectious disease. She said epidemiologists deal with similar matters of uncertainty at the beginning of disease outbreaks. “But they’re estimates, and they will need to be refined, and they may need to be refined a lot, especially since they are small numbers.”

If there is a connection between the vaccine and this rare syndrome, new cases are likely to emerge now that the word is out. Regulators announced the pause in part to alert doctors to the existence of this syndrome; as people begin looking, they may be more likely to find and report it. With numbers so low, the addition of even a few more cases could increase the rate. (In the last few days, Johnson & Johnson has reported two more possible cases, one in a woman, and one in a man.)

Cases of rare blood clots

Reported cases of cerebral venous sinus thrombosis in those receiving the Johnson & Johnson vaccine, compared with the number of cases that would normally be expected in that many people over 20.5 days.

Cases Total vaccinated Incidence rate
Initially reported 6 6.9 million 1 in 1,115,000
If cases double 12 6.9 million 1 in 570,000
Normal life (low end) 1.9 6.9 million 1 in 3,600,000
Normal life (high end) 7.7 6.9 million 1 in 900,000
Note: The clotting syndrome seen in vaccinated women includes both a clot and low platelet levels, a combination that is less likely than the "normal life" rates of 0.5 to 2 per 100,000 shown here. Total vaccinated includes data as of April 12. Since then, an additional 800,000 shots have been administered.·Source: Total vaccinated and normal incidence rates from C.D.C. presentation on April 14.

If there’s a link between the vaccine and the syndrome, more people who already got shots might still develop the clotting problem, since it appears to show up within a few weeks of vaccination. About half of Americans who received the Johnson & Johnson shot got it this month, according to government estimates. One reason the Centers for Disease Control and Prevention vaccine safety committee wants to wait longer before updating any guidance on the shot is to see what happens with this group. Since the pause was first recommended, the government count of Americans who have received the shot has increased to 7.7 million.

It may turn out that only some segments of the population are at high risk of this problem, in the same way that some populations are at higher risk of serious issues from certain diseases. Most of the cases so far have been in women between 18 and 50. If we look at six cases in that population, the syndrome looks somewhat more common, though still very rare.

Cases of rare blood clots in women 18-50

Reported cases of cerebral venous sinus thrombosis in women 18-50 receiving the Johnson & Johnson vaccine, compared with the number of cases that would normally be expected in that many people over 20.5 days. Experts say that normal rates for this specific age and sex group are likely higher than those shown.

Cases Total vaccinated Incidence rate
Initially reported 6 1.4 million 1 in 230,000
If cases double 12 1.4 million 1 in 120,000
Normal life (low end) 0.39 1.4 million 1 in 3,600,000
Normal life (high end) 1.58 1.4 million 1 in 900,000
Note: The clotting syndrome seen in vaccinated women includes both a clot and low platelet levels, a combination that is less likely than the "normal life" rates of 0.5 to 2 per 100,000 shown here. Total vaccinated includes data as of April 12.·Source: Total vaccinated and normal incidence rates from C.D.C. presentation on April 14.

If more cases are reported, it’s also possible that this gendered pattern will disappear.

Dr. Tom Shimabukuro, a vaccine safety expert at the C.D.C. who presented numbers to the vaccine safety board this week, said all of the current calculations are still “crude.”

It’s hard to tell right now. Studies of such events typically compare people who are given a medication or vaccine with a control group of people who didn’t. With a rare disorder like this, that comparison couldn’t be easily made using clinical trials. Researchers are conducting a large study of the health records of 12 million patients called the Vaccine Safety Datalink, comparing medical records of people who are vaccinated earlier with those who get their shots later — a system that doesn’t rely on voluntary reporting. Those results will take a while.

Researchers also look at what’s called a background rate of serious events: the odds someone could have a health problem even if he or she never got a vaccine. Comparing the rate of events among people who get a vaccine with the rate in the overall population can give a sense of whether a given patient’s outcome may be because of the vaccine, or is more likely to just be a coincidence.

The tables above show the background rates for the particular type of blood clot that authorities have seen in the vaccinated patients — a blockage in the brain called a cerebral venous sinus thrombosis. These numbers aren’t a perfect comparison, because there aren’t good background statistics for this type of clot combined with low blood platelets, the specific syndrome seen in the vaccinated women.

Women under 50 — the group that may be at risk of the vaccine side effect — are also more likely than the general population to have these blood clots just by being alive.

Many medications given to sick people can have serious side effects for some fraction of those who take them. Doctors and patients routinely weigh such risks against the benefits of medical treatment.

Birth control pills with estrogen have been frequently discussed this week because they are a common medication carrying a risk of blood clots. Clots caused by birth control pills are different from the syndrome associated with the Covid vaccines, and some experts caution about comparing them directly. The kind of clots caused by oral contraceptives typically form in patients’ legs, not in their brains, but they can still be serious. The pills more than double a typical woman’s risk of such an event, meaning between three and nine women out of 10,000 taking the pills for a year will develop a clot. (Pregnancy, the condition birth control pills are often prescribed to prevent, causes an even higher risk of blood clots.)

Risks of other blood clots for women

Annual risk of venous thromboembolism, a blood clot usually forming in a deep vein of the leg, thigh or pelvis.

Low estimate High estimate
Not pregnant or on birth control pills 1 in 10,000 1 in 2,000
On birth control pills 1 in 3,300 1 in 1,100
Pregnant 1 in 2,000 1 in 500
12-week postpartum period 1 in 250 1 in 150
Note: Incidence rates are rounded.·Source: 2012 article by the American College of Obstetricians and Gynecologists

“I’ll often say the risk of getting a blood clot with birth control pills is kind of similar to having a really serious reaction to penicillin,” said Dr. Raegan McDonald-Mosley, an obstetrician-gynecologist and the C.E.O. of Power to Decide, a group devoted to reducing unintended pregnancy. She frequently discusses blood clot risk with her patients, telling them the increase in risk and the overall magnitude of that risk. Most patients, she said, select their form of birth control based on other considerations.

Penicillin, a widely used antibiotic, causes serious allergic reactions in between one and five patients for every 10,000 who take it.

For vaccines, however, the threshold for safety is generally higher than for other kinds of medications. As many researchers have noted, Covid-19 puts people at risk of serious blood clots, too — much more so than any plausible estimate of the vaccine effect. But not everyone who fails to get vaccinated is going to get sick.

“The disease you get by chance, and the vaccine you get by choice, and that’s what makes it harder,” said Dr. Steven Black, an emeritus professor of pediatrics at Cincinnati Children’s Hospital, who studies vaccine safety.

For other vaccines, the risk of serious adverse events is much lower than for birth control pills or penicillin — they generally occur in fewer than 1 in 100,000 who receive a given vaccine. That rate is “clearly much, much less than would be tolerated for a drug,” said Dr. Nicola Klein, director of the Kaiser Permanente vaccine study center, who is involved in the Vaccine Safety Datalink study.

Most other vaccines protect against diseases that tend to be rare. By contrast, Covid-19 remains widespread throughout the United States and many parts of the world. Given the seriousness of the illness and its ease of spread, the value of vaccination may be higher now than it is when such trade-offs are usually considered.


Josh Katz contributed production.

Margot Sanger-Katz is a domestic correspondent and writes about health care for The Upshot. She was previously a reporter at National Journal and The Concord Monitor and an editor at Legal Affairs and the Yale Alumni Magazine. More about Margot Sanger-Katz

Alicia Parlapiano is a graphics editor and reporter covering politics and policy from Washington. She joined The Times in 2011 and previously worked at The Washington Post and the Pew Research Center. More about Alicia Parlapiano

A version of this article appears in print on  , Section A, Page 6 of the New York edition with the headline: Why the Numbers Regarding a Vaccine’s Safety Still Don’t Provide a Clear Picture. Order Reprints | Today’s Paper | Subscribe

Advertisement

SKIP ADVERTISEMENT