As China moves toward developing a national flu vaccination program, health officials will face a timing challenge, because the northern and southern parts of the country have different seasons, with some areas in between showing two annual flu activity spikes, researchers reported today.
According to an analysis of 7 years of flu activity, flu in northern provinces peaks in January and February, as in other Northern Hemisphere locations, while flu in the southern provinces peaks in the spring. The timing is important for flu vaccination planning, because a key strategy is to immunize people a few weeks before local flu activity ramps up to maximize the temporary protection afforded by flu shots.
The study group included researchers from the Chinese Centre for Disease Control and Prevention (China CDC) and Cecile Viboud, PhD, with the National Institutes of Health (NIH) Fogarty International Center in Bethesda, Md. Their findings appear today in PLoS Medicine.
Vaccine use expected to grow
Flu vaccination has increased in China since 1998, but coverage is still low—about 2%, according to the authors. However, the country's health officials expect use of flu vaccine to rise alongside rapid economic development. The authors said the goal of the study was to help China CDC get a better handle on flu's seasonality and strain circulation in different parts of the country, along with a clearer picture of the factors that drive virus activity.
The researchers collected weekly reports of lab-confirmed influenza A (H3N2) and B from sentinel hospitals in 30 Chinese provinces from 2005 through 2011, excluding the 2009 H1N1 pandemic months. They also compiled data on population size and density, mobility patterns, socioeconomic patterns, and daily weather for the cities in the flu surveillance network.
Then they applied spatiotemporal models to sort out the seasonality of the two flu strains, what factors affect seasonality, and where the broad epidemiologic regions are located in China.
They found that China has three regions with distinct influenza A seasons: the north with its winter peak, the south with its spring peak, and provinces at intermediate latitudes that have two peaks, one in January and February and one in the summer.
The most intriguing finding was a different seasonality for influenza B in China, which predominated in colder months throughout the country, the group wrote.
The researchers noted that China's roughly north-to-south flu spread seems to resemble that of Brazil, another large country that has diverse climates.
No single climate factor stood out as the main flu driver in China, but the team found that minimum temperature, hours of sunshine, and maximum rainfall seemed to influence winter, spring, and more complex patterns.
Multiple vaccination seasons
Concerning the implications for vaccination programs, northern China would follow the timing recommended for the Northern Hemisphere, and vaccination in the country's southernmost provinces would occur mainly in February and March, somewhat similar to the timing used in Southern Hemisphere countries, the authors wrote. They added that better surveillance is needed in the provinces between those two areas to help flesh out where they fall along the timing and locations spectrums.
The group pointed out two main limitations of their study: a short surveillance period and spotty surveillance information from some southern and mid-latitude provinces.
Cecile Viboud, PhD, one of the study coauthors and research scientist at the NIH's Fogarty Center, told CIDRAP News that one of the other surprising findings was semi-annual flu activity around Shanghai, despite its relatively high latitude, which would typically suggest a winter pattern. "There may be local factors specific to China that we don't yet understand," she said.
India is another large country with diverse climates that would be interesting to explore with a similar research design, Viboud said. "Influenza surveillance is still pretty limited there but there is good evidence that seasonal patterns vary in different regions of this large country."
Though the new H7N9 virus in China shows some signs of increased activity during colder months, it is still a zoonosis, and scientists still know very little about its ecology in avian reservoirs, Viboud said, adding, "A head-to-head comparison of the seasonality of human flu versus avian flu would be very interesting."
To build on the new findings, she said the researchers plan to do more detailed geographic analyses at the city level, especially in China's mid-latitude regions that have more complex flu season patterns, she said. The same group will also compare the genetic and antigenic match between the strains circulating in China and the World Health Organization's vaccine recommendations.
China CDC has many competing infectious and chronic disease priorities, she said, adding that her colleagues and China are committed to setting up a national flu immunization program sometime in the years ahead.
The geography of influenza B
In a commentary that accompanied the study, Steven Riley, PhD, with the MRC Centre for Outbreak Analysis and Modelling at Imperial College London, wrote that one of the study's other striking findings was that the proportion of influenza B increased from 20% in the northern provinces to nearly 50% in China's southern provinces. He said the difference adds another challenge to vaccination program design and should prompt phylogeographical and serotype studies of influenza B in China.
The decision to synchronize vaccination programs in southernmost China with the Southern Hemisphere timetable is tempting, but might be premature, Riley wrote. "Genetic data from even a small subset of the viral isolates used for this study could give a definitive picture of the ancestral relationship between viruses circulating in southern China relative to viruses in northern China and Southern Hemisphere populations," he said.
He commented that the diverse flu circulation patterns in China and other countries may lead to a refinement of the current practice of selecting two sets of vaccine strains each year, and that the growth of vaccine programs in regions that straddle the two hemispheres might provide settings to explore the benefits of more rapid vaccine production and locally driven strain selection.
Yu H, Alonso W, Feng L, et al. Characterization of regional influenza seasonality patterns in China and implications for vaccination strategies: spatiotemporal modeling of surveillance data. PLoS Med 2013 Nov 19;10(11):[Full text]
Riley S. Complex disease dynamics and the design of influenza vaccination programs, commentary. PLoS Med 2013 Nov 19;10(11):[Full text]
