Customize your JAMA Network experience by selecting one or more topics from the list below.
Drugstores and supermarkets begin exhorting customers to come in for a flu shot when they’re more likely to be shopping for sunscreen and potato salad than cough drops and Halloween candy.
For years, the US Centers for Disease Control and Prevention (CDC) has advised getting a flu shot by the end of October, if possible, before influenza viruses begin to circulate, but the question of whether it’s ever too early to get immunized remains unsettled. That’s become more of an issue because influenza vaccine has become available sooner each year than it used to be.
“These days, manufacturers are so efficient that they start delivering the vaccine earlier,” explained William Schaffner, MD, a professor of preventive medicine and infectious diseases at the Vanderbilt University School of Medicine.
“We have had reports of vaccine being distributed as early as the end of July,” said Jill Ferdinands, PhD, a research epidemiologist in the CDC’s influenza division.
And drugstores and supermarkets are motivated to immunize people sooner rather than later because if they don’t, customers might go somewhere else or do without, said Marc Lipsitch, PhD, an epidemiology professor who focuses on infectious diseases at the Harvard T.H. Chan School of Public Health. “The incentives in the private sector are not necessarily aligned with public health,” Lipsitch said.
The question of too-early immunization is relevant to anyone older than 8 years. (Children aged 6 months through 8 years who are getting vaccinated for the first time or who received only 1 dose of vaccine in a previous flu season are supposed to get 2 doses at least 28 days apart, so they should get their first dose as soon as it becomes available, according to the CDC.)
While some studies suggest that influenza vaccine’s effectiveness might wane before the end of flu season, public health researchers worry that shortening the recommended period for influenza vaccine could cause logistical problems, leading some people to skip getting immunized altogether.
If flu season began the same week and ended the same week every single year, figuring out the optimal timing of vaccinations would be a whole lot simpler.
However, “there are a lot of uncertainties,” Ferdinands said. “Early flu seasons are as common as late flu seasons.” And there’s no correlation between when flu seasons begin and how long they last, according to the CDC. To complicate matters, flu season doesn’t start at the same time throughout the country, noted Edward Belongia, MD, director of the Marshfield Clinic’s Center for Clinical Epidemiology and Population Health.
The only thing that’s certain is that the beginning of flu season is defined as the first week in which the number of laboratory-confirmed cases represented at least 2% of the total number of laboratory-confirmed cases for the season.
During the past 18 years, excluding the pandemic season of 2009-2010, the annual US flu season typically began in December or early January, according to the CDC. However, in 3 of those years, flu season started by the beginning of December, and in another 3 years, flu season started in or later than the third week of January. The average flu season lasted about 13 weeks, but some lasted as long as 17 weeks.
“I think the length is variable because sometimes there are multiple waves of [influenza] A and [influenza] B, and those tend to be out of sync with each other,” Lipsitch said.
The question of how early is too early to immunize would be moot if vaccine effectiveness remained stable throughout the flu season. But Ferdinands’ and Belongia’s research and other studies suggest that is not the case.
Ferdinands, Belongia, and their coauthors examined the association between vaccine effectiveness and time since vaccination among patients 9 years of age and older who received medical care for respiratory illness and were tested to see if they had the flu.
The researchers used data pooled from the US Flu Vaccine Effectiveness Network for the 2011-2012 through 2014-2015 flu seasons. Those who tested positive for influenza were the cases, while those with negative test results were the controls. Based on the odds of testing positive for influenza, they concluded that vaccine effectiveness declined about 7% per month after vaccination for H3N2, a type of influenza A, as well as for influenza B, and 6% to 11% per month after vaccination for H1N1 viruses, another type of influenza A.
In another recent study, researchers identified a cohort of 49 272 Kaiser Permanente patients who had received the flu vaccine from September 1, 2010, to March 31, 2017, and were subsequently tested for influenza A and B and respiratory syncytial virus (RSV). Those vaccinated 42 to 69 days before testing were 32% more likely to test positive for influenza compared with those who’d been vaccinated 14 to 41 days prior to testing.
The likelihood of testing positive for influenza increased by approximately 16% for every additional 28 days since vaccination. By the time that at least 154 days had passed since their immunization, people were twice as likely to test positive for influenza compared with those who’d been immunized only 14 to 41 days earlier. Meanwhile, the rate at which cohort members tested positive for RSV remained stable throughout the study duration.
If people who received their vaccination on October 1 had delayed it until November 26, 9 influenza episodes would have been prevented for every 1000 immunizations, the authors concluded.
If the observed decline in effectiveness is real, it could be due to characteristics of the vaccine itself, antigenic changes in the circulating influenza viruses, or a combination of both.
“I think it has to do with the actual components of the vaccine,” said Nicola Klein, MD, PhD, a coauthor of the Kaiser Permanente study and director of the Kaiser Permanente Northern California’s Vaccine Research Center. “I think it’s been pretty well established that the vaccine is not optimal,” Klein said, adding that this is not unique to the flu vaccine. “Pertussis vaccines [also] wane quite substantially.”
Although waning of effectiveness appeared to have occurred in most of the seasons they studied, “these kinds of studies always have the potential for some confounding that we’re not taking into account,” said Klein’s coauthor G. Thomas Ray, MBA, a senior data consultant at the Kaiser Vaccine Research Center.
In their article, Ferdinands and Belongia acknowledged the possibility of uncontrolled confounding. But they also noted that a definitive observational study probably is not feasible because of the large cohort that would be required and the need to measure vaccine effectiveness at multiple time points.
Another approach, they said, would be a trial in which participants were randomized to receive the vaccine at different points in time before flu season starts. But even randomized trials can be prone to biases that make it look like effectiveness is waning when it really isn’t.
“Those who do and do not get vaccinated may differ—for reasons other than the vaccine—in their risk of exposure to influenza infection, their risk of getting clinically ill…if infected, or…their tendency to seek care if ill with a respiratory infection,” Lipsitch noted in an editorial accompanying the Kaiser Permanente study.
Plus, he wrote, influenza vaccine is among the “leaky” vaccines that reduce but don’t eliminate a vaccinated individual’s infection risk. “I’m not completely convinced that there is waning,” Lipsitch said.
If vaccine effectiveness does indeed wane, the rate at which it occurs isn’t known, Ferdinands said. Findings have been inconsistent, she said, in part because most of the studies used to answer the question were not designed for that purpose. “The closest thing we have as a metric would be to look at the drop in [influenza] antibody titers, an imperfect correlate of protection.”
Research suggesting that the effectiveness of influenza vaccine wanes is strongest in older adults, who are at a greater risk of complications than younger, healthy adults. Using historical data from the 2010-2011 through 2015-2016 flu seasons, a recently published study evaluated the impact of delaying influenza vaccinations in US adults 65 and older.
The findings suggest that in a model in which it took a year for vaccine effectiveness to wane to zero, the time to start immunizing older individuals to obtain maximum population benefits would be the end of August or the beginning of September. And in a model in which vaccine effectiveness waned to zero in just 6 months, the optimal time to begin immunizing that population would be late September.
Delaying vaccination could have reduced the disease burden by up to 5.11% in the slow-waning model and up to 11.97% in the faster-waning model, depending on the flu season, the authors wrote. A reduction of just 1% equates to more than 1000 hospitalizations prevented, they said.
Whether Fluzone high-dose flu vaccine—licensed for people aged 65 and older and first used in the 2010-2011 flu season—might remain effective longer than other vaccines isn’t known, said study coauthor Melissa Stockwell, MD, MPH, an associate professor of pediatrics and population and family health at Columbia University’s College of Physicians & Surgeons and the Mailman School of Public Health.
It’s tricky to balance the potential for the vaccine’s effectiveness to wane over the course of the flu season with the goal of immunizing as many people aged 6 months and older as possible.
Still, Ferdinands said, “we do feel that it would be prudent not to encourage a shift of vaccination into the July and August time frame.”
Schaffner echoed Ferdinands. “Most of us would look askance at people getting vaccinated in August” or early September, he said. “If you’re 22 years old and in robust health, maybe that doesn’t matter.” But a frail 72-year-old with diabetes might want to wait until October to get a flu shot, Schaffner said.
Carving that message into stone in the form of practice guidelines is another matter.
Despite her study’s findings, Klein said, “we at Kaiser haven’t changed any of our clinical practice [related to flu immunization]. It’s better to get vaccinated early than not at all.”
One concern is that if physicians tell patients they see in August that they should come back later for a flu shot, the patients won’t return.
“We’re not even sure that the current infrastructure could get all those people vaccinated in a shorter time frame,” Ferdinands said. “That is a trade-off we’re worried about.”
The advice to get vaccinated by the end of October makes sense on a population level, and clinicians could explain to patients they see in August that they’d be wise to return in a month or two for their flu shot, Stockwell said.
However, she added, while in a perfect world it might make sense to wait at least until September to start administering flu vaccine, “we don’t live in a perfect world, and life interferes.”
Note: Source references are available through embedded hyperlinks in the article text online.
Rubin R. Is It Possible to Get a Flu Shot Too Early?. JAMA. 2018;320(22):2299–2301. doi:10.1001/jama.2018.18373
Create a personal account or sign in to: