Study: Universal flu immunization lowers death rates

Oct 29, 2008 (CIDRAP News) – When Ontario started offering free seasonal influenza vaccines to anyone over 6 months of age in 2000, Canadian researchers seized on a unique opportunity to compare the universal approach with a more targeted vaccination strategy, and they found that the universal model decreased flu-related deaths and healthcare visits.

The research team, mostly from the University of Toronto, published its findings today in the journal PLoS Medicine (Public Library of Science Medicine).

To compare the results of the two vaccination strategies, they analyzed mortality and hospitalization data from 1997 to 2004 for all 10 Canadian provinces. To gauge influenza patterns, the investigators looked at physicians' billing claims for hospitalizations, emergency department visits, and outpatient clinic visits. They also factored in annual population estimates and vaccination data.

To estimate influenza patterns, they generated a baseline of expected outcomes in the absence of flu activity and then computed influenza-associated events as the difference between the number of observed events and the expected baseline. They then ran statistical (Poisson regression) analyses for each outcome according to province and age-group to ferret out differences between universal and targeted vaccination.

Comparing vaccine strategies
When comparing pre– and post–universal vaccine time frames, they found that influenza vaccination rates for those age 12 and older increased 20 percentage points for Ontario (from 18% to 38%), compared with an 11-percentage-point hike for other provinces (from 13% to 24%).

Ontario held a vaccination-rate lead over other provinces for residents who were younger than age 64, but its rate increase actually lagged other provinces' for those aged 75 and older.

After the universal flu vaccine strategy was deployed, influenza-related mortality for Ontario residents decreased 74%, compared with 57% in other provinces, according to the study. However, in age-specific analyses, the researchers found that the mortality decreases in Ontario compared with the other provinces were statistically significant only in those aged 85 or older.

In analyzing influenza-related healthcare use patterns, the group found that hospitalization, emergency department use, and outpatient visits decreased more in Ontario than in other provinces.

When looking at age-specific patterns, they consistently saw greater decreases in Ontario residents who were younger than 65. In addition, Ontario's seniors had greater decreases in hospitalizations (ages 65 to 84) and emergency department use (ages 65 to 74) when compared with seniors in other provinces.

As mentioned, the researchers estimated that only about 38% of Ontario's population received annual flu immunization after universal coverage was implemented. Vaccination rates in Ontario for those aged 12 to 19 increased from 16% to 31% over the study period, compared with a rise of 6% to 11% in the other provinces, the group reported.

Though vaccine uptake was higher among older age-groups, "it is uncertain that levels required for appreciable herd immunity effects were obtained in the overall population, particularly in younger age groups," they wrote. They add, though, that the larger mortality decreases in Ontario seniors age 85 and older, fewer hospitalizations for those age 65 to 84, and emergency department use for those age 65 to 74 may suggest a herd immunity effect of the universal strategy.

"Herd immunity" refers to the protection received by even unvaccinated people when a high percentage of the population becomes immunized, making contact with infected people less likely.

Some shortcomings of the study
The investigators said the study does not allow them to tease out the direct benefits from the indirect benefits.

Though potential confounders such as strain-specific surveillance data and individual comorbidities could have influenced the findings, "we have no reason to believe that these factors change more over time in Ontario compared to other provinces," they wrote.

Also, they noted that they couldn’t rule out a "healthy population" bias because of the observational nature of the study. "We believe this study represents the best possible evaluation of a real-world natural experiment that Ontario's unique strategy to offer influenza vaccination to the entire population represents," they added.

In addition, the authors mention that mortality and healthcare use rates over the study period could have been lower in all provinces because of less severe influenza seasons.

Weighing universal immunization
The authors concluded that the study's findings don't settle the question of what flu vaccination strategy is better, but raises the possibility that universal vaccination may be an effective strategy for increasing a population's protection against seasonal influenza.

A commentary on the study—by Cécile Viboud and Mark Miller from the National Institutes of Health—appears in the same issue of PLoS Medicine. The two wrote that interest in alternative ways to protect seniors from influenza is high, given disappointing news in several recent studies that suggest the vaccine is less effective in this at-risk group.

The United States' recent expansion of the recommended vaccine target group to include all children ages 6 months to 18 years would not only reduce pediatric infections, but might also protect seniors, they wrote.

Viboud and Miller credited the authors of the study for the thorough sensitivity analyses they conducted to control for factors that weren't related to vaccine uptake. However, they noted that observational studies lack statistical power.

Though they wrote that the study was likely underpowered to evaluate the direct and indirect benefits of universal immunization, cluster-randomized trials might provide a more powerful tool to evaluate the effects.

"The increasing pressure to improve influenza control may eventually generate sufficient interest in sponsoring such large-scale and costly randomized studies," Viboud and Miller wrote. "Ironically, there is only a short time window of opportunity for these studies—once pediatric or universal vaccination policy becomes established, there is no ethical 'control' community left."

They also wrote that "the Ontario program was very successful in developing innovative channels for influenza vaccine delivery—an experience extremely useful in the context of pandemic preparedness."

Kwong JC, Stukel TA, Lim J, et al. The effect of universal influenza immunization on mortality and health care use. PLoS Med 2008 Oct;5(10):1440-52 [Full text]

Viboud C, Miller M. Health benefits of universal influenza vaccination strategy. (Perspectives) PLoS Med 2008 Oct;5(10):1423-25 [Full text]

This week's top reads

Our underwriters