Advertisement

SKIP ADVERTISEMENT

Comments

See the Data on Breakthrough Covid Hospitalizations and Deaths by StateSkip to Comments
The comments section is closed. To submit a letter to the editor for publication, write to letters@nytimes.com.

See the Data on Breakthrough Covid Hospitalizations and Deaths by State

Serious coronavirus infections among vaccinated people have been relatively rare since the start of the vaccination campaign, a New York Times analysis of data from 40 states and Washington, D.C., shows. Fully vaccinated people have made up as few as 0.1 percent of and as many as 5 percent of those hospitalized with the virus in those states, and as few as 0.2 percent and as many as 6 percent of those who have died.

There is still a lot we do not know about so-called breakthrough infections — when fully inoculated people contract the virus. And there is some evidence that these cases are becoming more common as the more transmissible Delta variant surges. While vaccines have done a remarkable job at protecting a vast majority of people from serious illness, the data in the Times analysis generally spanned the period from the start of the vaccination campaign until mid-June or July, before the Delta variant became predominant in the United States.

Breakthrough Covid-19 Hospitalizations and Deaths by State

Among fully vaccinated people in each state since vaccination began.

StateBreakthrough hospitalizationsAs a pct. of all Covid hospitalizationsBreakthrough deathsAs a pct. of all Covid deaths
Alabama 77 0.3% 21 0.4%
Alaska 17 2.0% 2 3.3%
Arizona 379 0.5% 34 0.3%
Arkansas 4.7% 2.6%
California 843 0.4% 88 0.5%
Colorado 312 1.2% 52 3.1%
Delaware 22 0.4% 6 3.5%
Georgia 85 0.1% 21 0.2%
Idaho 37 0.9% 6 1.3%
Illinois 563 0.6% 151 2.2%
Note: Arkansas did not provide counts of breakthrough hospitalizations or deaths, and West Virginia did not provide a count of breakthrough hospitalizations.

Data on less serious breakthrough infections is not widely available, though it is possible those cases are rising. Data was not available for several states in which the virus has been surging, including Florida and Missouri.

Until recently, the Centers for Disease Control and Prevention estimated that fully vaccinated people accounted for less than 3 percent of coronavirus hospitalizations nationwide and less than 1 percent of virus deaths. But last week, the agency noted that those figures did not reflect new data involving the Delta variant and said it was actively working to update them. Only about 50 percent of people in the United States are fully vaccinated.

Looking at how many hospitalizations and deaths have involved fully vaccinated people is a common but crude measure of how well the vaccines are working.

As more people get shots, the percentage of hospitalizations and deaths among fully vaccinated people should rise. This may seem counterintuitive, so it’s important to understand why.

In a state with a high vaccination rate, a higher percentage of breakthroughs may simply reflect that fully vaccinated people are a bigger chunk of the population, or that there are few hospitalizations and deaths overall. Imagine a state where just two people are hospitalized but both are vaccinated — breakthroughs would account for 100 percent of the hospitalizations in that state, even though these cases were very rare.

Assessing breakthrough rates

Some epidemiologists and state health officials instead recommend comparing how likely a vaccinated person was to be hospitalized or to die, compared with an unvaccinated person.

To do this, the Times used the number of breakthroughs and the vaccination rate over time to estimate rates of hospitalizations and deaths for each group in every state for which data was available.

In those states, people who were not fully vaccinated were hospitalized with Covid-19 at least five times more often than fully vaccinated people, according to the analysis, and they died at least eight times more often.

How Rates of New Covid-19 Hospital Admissions and Deaths Compare

Among fully vaccinated people and among people who were not fully vaccinated in each state since vaccination began.

Hospitalization rate per 100,000Death rate per 100,000
State For vaccinated people For unvaccinated people For vaccinated people For unvaccinated people
Alabama 9 696 75x higher for unvaccinated people 3 124 48x higher for unvaccinated people
Alaska 9 154 17x 1 11 10x
Arizona 27 1,306 47x 2 182 73x
California 9 647 68x 1 58 58x
Colorado 24 567 22x 4 37 8x
Delaware 7 978 148x 2 26 14x
Georgia 5 735 161x 1 99 87x
Idaho 11 288 25x 2 30 16x
Illinois 20 1,001 48x 5 68 11x
Indiana 9 547 57x 4 29 7x
Note: Table shows rates for the entire reporting period for each state calculated with the average vaccination rate for the total population over that period. Rates per 100,000 are rounded. Rates used to calculate comparisons were unrounded. Data for unvaccinated individuals includes partially vaccinated people, unvaccinated people, and people with unknown vaccination status.

“In an ideal world, you would be able to calculate the rates based on the number of people who were actually exposed,” said Kristen Panthagani, a geneticist at Baylor College of Medicine who runs a blog explaining complicated scientific concepts, including breakthrough infections. “But that number is really hard to figure out.”

More breakthroughs as the outbreak grows

In interviews, epidemiologists said that the United States is likely to see more breakthroughs, especially in areas where cases are surging. Essentially — the more that the virus circulates, the more exposures you can expect, and the more breakthroughs you can expect.

“The more infection rates go up in the background, the more you’re going to see disease among people who were immunized,” said Dr. Paul McKinney, associate dean at the University of Louisville’s school of public health in Kentucky. “People need to be aware that, to the extent that we can keep the incidence down, the better off everyone’s going to be.”

It does not help that, in recent weeks, new research has shown that vaccinated people, if they become infected, can carry high levels of the coronavirus.

Although at least 80 percent of people 65 and older are vaccinated in the United States, surging cases could still present an elevated risk for them. It is possible, epidemiologists said, that seniors and people with compromised immune systems could be particularly vulnerable in a surge, even if they were fully vaccinated, because their bodies might not produce a strong immune response from the vaccine. The Food and Drug Administration is speeding efforts to authorize booster shots for people with weakened immune systems.

In Mississippi — where only about 35 percent of the population is fully vaccinated and where infections and hospitalizations have been surging — vaccinated older people and people with weaker immune systems have been overrepresented among those hospitalized and dying, according to the state health department.

“We’re seeing a pretty dramatic spillover effect from the transmission in the community to our more vulnerable parts of our population,” Thomas Dobbs, Mississippi’s state health officer, said last week, though he emphasized that a higher proportion of breakthroughs did not mean the vaccine had become less effective.

A need for better data

A lack of comprehensive data, paired with growing concern over the Delta variant, may be helping to drive fears about breakthrough infections.The C.D.C. does not provide data for breakthrough instances at the state or local level, nor does it provide the data over time, which makes it difficult to see if fully vaccinated people may be becoming more vulnerable to the virus in certain areas.

The agency, along with some states, stopped tracking mild breakthrough cases in May, focusing instead on breakthrough infections that result in hospitalization or death. The decision has recently garnered criticism from several lawmakers, who have argued that having detailed information on the infections is critical to understanding how the virus is behaving.

Several epidemiologists told The Times that collecting comprehensive data on mild cases is nearly impossible because those infected may have such mild symptoms that they do not bother getting tested (or they may have no symptoms at all).

“Most of the breakthrough cases will not be diagnosed because they tend to be mild or more like a cold,” said Dr. Chighaf Bakour, an assistant professor of epidemiology at the University of South Florida, explaining that the cases that are diagnosed would be more likely to be moderate to severe because those people would seek out testing or treatment.

But epidemiologists said that having better data at the local level could help public health experts identify problems — including more dangerous or transmissible variants, and also flaws in vaccine storage or manufacturing.

Beyond that, they said it is critical to understand how breakthrough cases might be affecting different demographic populations — including different age groups and racial groups, groups that received different vaccines, and groups that received their vaccines earlier in the year versus later in the year. That, they said, is important to keeping the virus at bay.

“The data is our access; we have to collect the data,” said Debra Furr-Holden, a Michigan State epidemiologist and associate dean for public health integration. “At this point, we are not trying to eradicate Covid. We are now trying to figure out how to mitigate its impact.”

Methodology

The Times asked all 50 states and Washington, D.C., to provide data on breakthrough infections, hospitalizations and deaths of fully vaccinated people, roughly spanning back to the first months of the vaccination campaign at the end of 2020 and beginning of 2021. Many of the states provided data through June or July, which may not account for the recent surge in Delta variant infections. In every state providing information, hospitalizations and deaths among fully vaccinated people accounted for a small minority of the totals.

Forty-four states, plus Washington, D.C., shared some breakthrough hospitalization or death information with The Times, though some provided only two months or less of records. Those states — Connecticut, Hawaii, Maryland and Wyoming — were excluded from the analysis. Three additional states — Iowa, Missouri and Pennsylvania — said they were unable to provide data on breakthroughs because their own systems did not effectively track them. Florida, Kansas and New York did not provide any data on breakthrough hospitalizations or deaths.

The states define breakthrough cases in different ways and provided data for different periods, so exercise caution when making comparisons. Some count fully vaccinated people who tested positive for the virus, even if they had no symptoms, while others do not. Some considered a case a breakthrough only if the person had been fully vaccinated for 14 days or more. Some states said that they did not know the hospitalization status of all of their breakthrough infections or that they only received data from a subset of hospitals. Some states said their breakthrough data was preliminary and subject to change.

Some hospitalizations and death records lacked vaccine status altogether. Data for individuals who were not fully vaccinated includes partially vaccinated people, unvaccinated people, and people with unknown vaccination status.

The Times used data reported by hospitals to the Department of Health and Human Services for new Covid-19 hospital admissions to calculate total hospitalizations in each state across the same time period, as well as to calculate the number of hospitalized people in each state who were not fully vaccinated. The data can overestimate the amount of people hospitalized if the same person is admitted to a hospital multiple times, and the data can lag. The Times counted confirmed and probable virus hospitalizations in adults and children. For Rhode Island, the Times used total hospitalization data provided by the state instead of federal data because it was more complete.

The analysis used state data for total Covid-19 deaths when it was possible to do so. If it was not available, The analysis used death data from The New York Times coronavirus database.

The rates for fully vaccinated and not fully vaccinated groups were calculated from the average number of people in each group across all days within the period for which data was provided. Because only a total number of breakthrough hospitalizations or deaths for the entire period was available, rates cover different periods for each state. Rate calculations are based on the entire fully vaccinated and not fully vaccinated population, not any measure of who has been exposed to Covid-19.