Bulletin of the World Health Organization

Influenza seasonality and vaccination timing in tropical and subtropical areas of southern and south-eastern Asia

Siddhartha Saha a, Mandeep Chadha b, Abdullah Al Mamun c, Mahmudur Rahman d, Katharine Sturm-Ramirez c, Malinee Chittaganpitch e, Sirima Pattamadilok e, Sonja J Olsen f, Ondri Dwi Sampurno g, Vivi Setiawaty g, Krisna Nur Andriana Pangesti g, Gina Samaan h, Sibounhom Archkhawongs i, Phengta Vongphrachanh i, Darouny Phonekeo i, Andrew Corwin j, Sok Touch k, Philippe Buchy l, Nora Chea m, Paul Kitsutani n, Le Quynh Mai o, Vu Dinh Thiem o, Raymond Lin p, Constance Low p, Chong Chee Kheong q, Norizah Ismail r, Mohd Apandi Yusof s, Amado Tandoc t, Vito Roque u, Akhilesh Mishra b, Ann C Moen v, Marc-Alain Widdowson v, Jeffrey Partridge w & Renu B Lal c

a. Center for Disease Control and Prevention, Influenza Programme, c/o US Embassy, Shanti Path, Chanakyapuri, New Delhi, India.
b. National Institute of Virology, Pune, India.
c. International Centre for Diarrhoeal Disease Research Bangladesh, Dhaka, Bangladesh.
d. Institute of Epidemiology, Disease Control and Research, Dhaka, Bangladesh.
e. National Institute of Health, Ministry of Public Health, Nonthaburi, Thailand.
f. Center for Disease Control and Prevention, Influenza Programme, Nonthaburi, Thailand.
g. Ministry of Health, Jakarta, Indonesia.
h. Center for Disease Control and Prevention, Jakarta, Indonesia.
i. Ministry of Health, Vientiane, Lao People’s Democratic Republic.
j. Center for Disease Control and Prevention, Influenza Programme, Vientiane, Lao People’s Democratic Republic.
k. Ministry of Health, Phnom Penh, Cambodia.
l. Pasteur Institute, Phnom Penh, Cambodia.
m. World Health Organization, Phnom Penh, Cambodia.
n. Center for Disease Control and Prevention, Influenza Programme, Phnom Penh, Cambodia.
o. National Institute of Hygiene and Epidemiology, Hanoi, Viet Nam.
p. Ministry of Health, Singapore.
q. Ministry of Health, Kuala Lumpur, Malaysia.
r. National Public Health Laboratory, Kuala Lumpur, Malaysia.
s. Institute of Medical Research, Kuala Lumpur, Malaysia.
t. Research Institute for Tropical Medicine, Alabang, Philippines.
u. Department of Health, Manila, Philippines.
v. Centers for Disease Control and Prevention, Atlanta, United States of America.
w. Center for Disease Control and Prevention, Influenza Programme, Hanoi, Viet Nam.

Correspondence to Renu B Lal (e-mail: rbl3@cdc.gov).

(Submitted: 11 May 2013 – Revised version received: 17 November 2013 – Accepted: 21 November 2013 – Published online: 24 February 2014.)

Bulletin of the World Health Organization 2014;92:318-330. doi: http://dx.doi.org/10.2471/BLT.13.124412

Introduction

Influenza is a vaccine-preventable disease; safe and effective vaccines have been used for more than 60 years to mitigate the impact of seasonal epidemics.1 Annual human influenza epidemics occur during the respective winter seasons in the temperate zones of the northern and southern hemispheres – i.e. between November and March in the northern hemisphere and between April and September in the southern hemipshere.13 Data on influenza strains circulating globally are used to predict which strains have the highest probability of circulating in the subsequent season and this information is used to generate recommendations for seasonal influenza vaccine composition.4

The effectiveness of influenza vaccines depends not only on the right match between vaccine strains and circulating viral strains, but also on the vaccine-induced immune response in the target population.5,6 Several types of influenza vaccine are regularly available on the market. The most commonly used one – non-adjuvanted, influenza-virus-containing vaccine (the “inactivated influenza vaccine”) ­– induces neutralizing antibody responses that wane during the year.5 Therefore, appropriate timing of vaccination is a very important consideration in efforts to improve vaccine effectiveness. Better understanding of influenza seasonality and viral circulation is essential to selecting the best time for vaccination campaigns, which should precede the onset of the influenza season by several weeks.

Despite substantial influenza-associated morbidity and mortality and increasing local vaccine manufacturing capacity, many tropical Asian countries have yet to improve population-wide routine influenza vaccination.710 Although influenza surveillance data have been very useful in developing vaccination strategies in temperate regions, fewer data are available from countries in tropical and subtropical areas.4,1113 Current progress in influenza surveillance and the widespread use of highly sensitive molecular assays for influenza diagnosis and viral typing have shown more complicated patterns of influenza activity in the tropics and subtropics than in other areas, with year-round circulation in some regions and bi-annual peaks of circulation in others.13,14 Such undefined patterns of influenza activity further complicate vaccination recommendations and, in particular, the selection of an appropriate time for vaccination.2,15 We undertook the analysis of influenza surveillance data from 10 countries in tropical and subtropical parts of southern and south-eastern Asia to characterize common trends in influenza circulation and therefore identify the most appropriate time for vaccination.

Methods

Data collection

Nasopharyngeal swabs from patients presenting with influenza-like-illness were collected at various clinics within each country as part of an influenza surveillance network across the Asian region. The number of specimens tested and the number of samples positive for influenza virus or viral nucleic acid were collected directly from weekly surveillance data or extracted from FluNet for 10 selected countries in Asia: Bangladesh, Cambodia, India, Indonesia, the Lao People's Democratic Republic, Malaysia, the Philippines, Singapore, Thailand and Viet Nam.14 FluNet data are regularly collected by surveillance systems in geographically disparate sentinel sites that register cases of influenza-like illness or severe acute respiratory infection.13,1627 Data on specimen positivity and on viral subtypes were based upon specimens from these surveillance systems; all samples were tested using reverse transcription polymerase chain reaction assays,28 with the exception of samples from India collected before 2009, which were tested with viral isolation methods.19

Since influenza surveillance in the different countries was started in different years, data for India, Malaysia, the Philippines and Viet Nam were available for 2006 to 2011; data for Cambodia, Indonesia, Singapore and Thailand, for 2007 to 2011; and data for Bangladesh and the Lao People's Democratic Republic, for 2008 to 2011.

Data analysis

Data on laboratory-confirmed influenza for each country were analysed individually in MS Excel (Microsoft, Redmond, United States of America) and PASW Statistics 18 (SPSS Inc., Chicago, USA). Monthly influenza activity was calculated by adding the weekly number of specimens that tested positive during a given month. The monthly data were then plotted as the percentage of all positive specimens during the calendar year that corresponded to that month. Means and standard errors were calculated from the cumulative data for each country over the period evaluated. Data from 2009 were processed separately owing to the emergence of influenza virus A(H1N1)pdm09, which did not follow the usual seasonal pattern of influenza viruses.

Results

Circulating influenza viruses

Of a total of 253 611 specimens tested in the 10 participating countries, 45 282 (17.9%) were positive for viral nucleic acid (or for influenza viruses from specimens from India obtained before 2009) (Table 1).

Bangladesh

Of the 12 583 specimens tested between 2008 and 2011, 1747 (13.9 %) were positive (Table 1). Analysis of monthly data showed that influenza activity was highest from June to September during most years (Fig. 1 and Fig. 2), with peak activity in July and August (Fig. 2). Subtype A H1 and type B influenza viruses were the ones most frequently reported during 2008 (Fig. 3): they were found in 39.3% and 46.7%, respectively, of all specimens that tested positive. Influenza A(H1N1)pdm09 emerged in mid-2009 (42.7%) and its circulation persisted in 2010 (39.2%); it co-circulated mainly with subtype A H3 viruses (51.1%) in 2009 and with type B viruses (43.6%) in 2010 (Fig. 3). A H3 viruses were the ones most frequently reported in 2011 – 50.4% of all positive specimens – followed by type B viral strains (47.1%) (Fig. 3).

Fig. 1. Monthly distribution of samples testing positive for influenza virus or viral nucleic acid by year and country of southern or south-eastern Asia, 2006–2011
Fig. 1. Monthly distribution of samples testing positive for influenza virus or viral nucleic acid by year and country of southern or south-eastern Asia, 2006–2011
a This represents the samples testing positive on any given month, as a percentage of all the samples that tested positive during the year.
Note: Figures were obtained from weekly surveillance data.
Fig. 2. Monthly patterns of influenza trends in samples testing positive to influenza virus or viral nucleic acid in 10 countries of southern and south-eastern Asia, 2006–2011
Fig. 2. Monthly patterns of influenza trends in samples testing positive to influenza virus or viral nucleic acid in 10 countries of southern and south-eastern Asia, 2006–2011
a This represents the mean value for the samples testing positive on any given month from 2006 to 2011 as a percentage of all the samples that tested positive during each year.
Note: Data from 2009 were excluded because of the presence of pandemic A(H1N1)pdm09, which did not follow the usual seasonality pattern. The latitude of each country’s capital city is shown at the top of each panel.
Fig. 3. Influenza virus types and subtypes identified, by year and country of southern or south-eastern Asia, 2006–2011
Fig. 3. Influenza virus types and subtypes identified, by year and country of southern or south-eastern Asia, 2006–2011
a This represents the percentage of samples that tested positive for a specific viral type or subtype during a given year, as a total of all samples testing positive for influenza virus or viral nucleic acid the same year.
Note: Inconsistencies arise in some values due to rounding.
Cambodia

Of the 10 105 specimens tested from 2007 to 2011, 1574 (15.6%) were positive (Table 1). Analysis of monthly data showed influenza activity primarily from July to December for most years (Fig. 1 and Fig. 2), with discrete peaks in September to October (Fig. 2). In 2007, influenza activity peaked in March and in August to September, while a peak of activity was identified between September and November for 2008 and between August and November for 2010 and 2011 (Fig. 1). Subtype A H3 (29.5% and 53.7%) and type B (56.8% and 34.1%) viral strains were the ones most frequently reported in 2007 and 2008. Influenza A(H1N1)pdm09 virus (22.9%) emerged in mid-2009 and its circulation persisted in 2010 (32.8%) and 2011 (25.2%), with co-circulation of subtype A H3 viruses (53.1%, 40.2%, and 9.2% in 2009, 2010 and 2011, respectively) and type B viral strains (22.9%, 27.0% and 65.6% in 2009, 2010 and 2011, respectively) (Fig. 3).

India

Of the 28 024 specimens tested from 2006 to 2011, 2803 (10.0%) were positive (Table 1). Between 2006 and 2008 all samples were studied by virus isolation methods. Analysis of monthly data showed increased influenza activity after June for most years (Fig. 1 and Fig. 2), with discrete peaks between June and August (Fig. 1). In 2008, influenza activity remained above background activity throughout the year (Fig. 1). India had highly divergent monthly patterns between the northern and the southern states (data not shown). Subtype A H1 and type B viruses were the most frequently reported in 2006 and 2007 (35.0% and 38.0% for subtype A H1, and 38.5% and 47.0% for type B strains, respectively, Fig. 3). Subtype A H3 (39.7%) and type B (50.0%) viruses co-circulated in 2008. Subtype A H3 viruses circulated in the first half of 2009, followed by A(H1N1pdm09) virus (42.0%), which emerged mid-2009 and persisted in 2010 (40.2%) and 2011 (12.1%, Fig. 3). Type B (8.3%, 52.1%, 32.3%) and subtype A H3 influenza viruses (44.1%, 7.6%, 55.6%) co-circulated with A(H1N1pdm09) during all three years.

Indonesia

Of the 15 150 specimens tested from 2007 to 2011, 2511 (16.6%) were positive (Table 1). Analysis of monthly data showed year-round influenza activity for most years and different months of peak influenza activity in some years (Fig. 1 and Fig. 2). Circulating influenza types and subtypes were available only for 2010 and 2011. Influenza subtype A H3 and type B viruses were the most frequently reported (37.1% and 45.5%, for subtype A H3 virus and 49.9% and 24.8% for type B viruses, respectively). Identification of A(H1N1)pdm09 in the country varied from 11.8% of positive specimens in 2010 to 29.0% in 2011 (Fig. 3; no virus identification data were available for 2009).

Lao People's Democratic Republic

Of the 5949 specimens tested from 2008 to 2011, 1302 (21.9%) were positive (Table 1). Analysis of monthly data showed influenza circulation primarily from August to December for most years (Fig. 1), with discrete peaks in August, September and October (Fig. 1 and Fig. 2). The Lao People's Democratic Republic reported circulation of subtype A H1 and type B viruses in 2008 (43.5% and 56.5%, respectively). Influenza A(H1N1)pdm09 virus emerged in mid-2009 (47.5%) and persisted in 2010 (56.1%). Type B viruses were also identified in the country in 2010 (31.0%) and 2011 (31.3%). In 2011, A H3 was the most frequently reported viral subtype (63.5% of positive specimens) (Fig. 3).

Malaysia

Of the 10 323 specimens tested from 2006 to 2011, 894 (8.7 %) were positive (Table 1). Analysis of monthly data showed year-round influenza activity for most years and different months of peak influenza activity in different years (Fig. 1 and Fig. 2). Influenza A H1, A H3 and type B viruses circulated at varying levels in 2006 to 2008 (14.0–65.2%), while A(H1N1)pdm09 virus (24.5%) emerged in mid-2009 and persisted in 2010 (56.3%); it co-circulated with subtype A H3 (52.3%) and type B viruses (21.8%) in 2009 and mainly with type B viruses (39.6%) in 2010 (Fig. 3).

The Philippines

Of the 69 108 specimens tested from 2006 to 2011, 12 607 (18.2%) were positive (Table 1). Analysis of monthly data showed influenza activity primarily from June to October (Fig. 1 and Fig. 2), with discrete peaks in July, August and September (Fig. 2). Subtype A H1 (46.1%) and type B strains (44.0%) were the predominant viruses in 2006; A H1 and A H3 viral subtypes were the ones most frequently reported in 2007 (34.1% and 59.7%, respectively), as were type B strains in 2008 (75.7%). Influenza A(H1N1)pdm09 virus (92.8%) emerged in mid-2009 and persisted in 2010 (33.4%), when it co-circulated with type B (50.1%) and subtype A H3 viruses (16.4%). Influenza A(H1N1)pdm09 was the virus most frequently (53.2%) reported in 2011, followed by type B influenza viruses (31.8%) (Fig. 3).

Singapore

Of the 55 449 specimens tested from 2007 to 2011, 12 801 (23.1%) were positive (Table 1). Analysis of monthly data showed year-round influenza activity and different months of peak influenza activity in different years (Fig. 1 and Fig. 2). Type B viruses were the most frequently reported influenza viruses in 2007 (79.1%) and 2008 (45.7%). Influenza A(H1N1)pdm09 (79.2%) emerged in mid-2009 and persisted in 2010 (55.8%) and 2011 (44.2%); it co-circulated with subtype A H3 (22.5% and 33.8%, respectively) and type B viruses (21.7% and 22.0% respectively; Fig. 3).

Thailand

Of the 17 421 specimens tested from 2007 to 2011, 3802 (21.8%) were positive (Table 1). Analysis of monthly data showed that increased influenza activity occurred from January to March and from June to November or December on some years (Fig. 1 and Fig. 2), with discrete peaks in August to September (Fig. 2). Subtype A H3 (43.3% and 27.2%) and type B viruses (40.1% and 42.9%) were the most frequently reported in 2007 and 2008, respectively. Pandemic A(H1N1)pdm09 emerged (60.6%) in mid-2009 and persisted in 2010 (53.3%); it co-circulated with type B (12.9%) viruses and with subtype A H1 virus (14.0%) and A H3 viruses (12.6%). A H3 (58.5%) was the viral subtype reported most frequently in 2011, followed by type B viral strains (33.7%, Fig. 3).

Viet Nam

Of the 29 499 specimens tested from 2006 to 2011, 5241 (17.8%) were positive (Table 1). Analysis of monthly data showed year-round circulation (Fig. 1 and Fig. 2), and some peak activity in July to August (Fig. 2). Subtype A H1 and type B viruses were the most frequently reported in 2006 and 2008, at 43.5 to 53.7% and 40.8 to 41.3%, respectively, whereas A H3 subtype (73.9%) and type B viruses (25.1%) were the ones most frequently reported in 2007. Influenza A(H1N1)pdm09 emerged (46.6%) in mid-2009 and persisted in 2010 (28.0%), when it co-circulated with subtype A H3 (71.5%) viruses. In 2011, A(H1N1)pdm09 was the influenza virus most frequently reported (74.1%), followed by type B viruses (22.1%, Fig. 3).

Circulating influenza types and subtypes

Different types and subtypes of influenza viruses circulated across southern and south-eastern Asia during the years studied: type A (subtypes H1 and H3) and type B strains (Fig. 3). Influenza A H1, A H3 and type B viruses were reported across the countries in 2006, 2007 and 2008. As in the rest of the world, pandemic A(H1N1pdm09) appeared in southern and south-eastern Asia in 2009; it was detected at different times in the various countries and it co-circulated with other influenza viruses (Fig. 1 and Fig. 3). In 2009, A(H1N1)pdm09 and subtype A H3 (plus some subtype A H1 and type B viruses) co-circulated in Bangladesh, Cambodia, India, the Lao People's Democratic Republic, Thailand and Viet Nam, whereas A(H1N1)pdm09 was the predominant subtype reported from the Philippines and Singapore in 2009. In 2010, A(H1N1)pdm09 co-circulated mainly with type B viruses in some countries (Bangladesh, India, the Lao People's Democratic Republic, Malaysia, the Philippines and Thailand), and with subtype A H3 viruses in others (Cambodia and Viet Nam). In 2011, some countries (Bangladesh, India, the Lao People's Democratic Republic and Thailand) showed almost no circulation of A(H1N1)pdm09 and varying degrees of co-circulation of subtype A H3 and type B viruses, while others (Cambodia, Indonesia, the Philippines, Singapore and Viet Nam) continued to have co-circulation of A(H1N1)pdm09 and influenza B (Fig. 3).

Influenza circulation and vaccination timing

Two major patterns of influenza circulation emerged from this study (Fig. 2). Some countries (Bangladesh, Cambodia, India, the Lao People's Democratic Republic, the Philippines, Thailand and Viet Nam) presented influenza peak activity between June and October, and three of them (India, Thailand and Viet Nam) showed additional minor peak activity during the northern hemisphere winter. Other countries (Indonesia, Malaysia and Singapore) had year-round influenza activity and variable peak activity in some years (Fig. 4).

Fig. 4. Months of peak influenza activity in 10 countries of southern and south-eastern Asia, 2006–2011
Fig. 4. Months of peak influenza activity in 10 countries of southern and south-eastern Asia, 2006–2011
Note: The map illustrates the months in which influenza activity peaks in each country, as well as the latitudes for the capital cities.

Cumulative analysis of monthly influenza activity between 2006 and 2011 in the 10 countries revealed that > 60% of influenza cases were reported in June to November for Bangladesh, Cambodia, India, the Lao People's Democratic Republic, the Philippines, Thailand and Viet Nam (Table 2), whereas countries closer to the equator (Indonesia, Malaysia and Singapore) had similar circulation of influenza viruses in both semesters (Table 2). The best vaccination times for each country, as proposed, are indicated by an arrow in Fig. 2.

Discussion

Weekly influenza surveillance data collected between 2006–2008 and 2011 were analysed for 10 countries in tropical and subtropical southern and south-eastern Asia. Despite differences in surveillance methods, demographic characteristics and environmental factors,2,3,12,30,31 it was possible to show peaks of influenza circulation in some countries, while in others influenza activity was detected throughout the year. These findings are consistent with the reports from the countries.13,1627 Host-related and environmental factors (e.g. lower temperatures and decreased humidity) may influence both viral transmission and host susceptibility.3,32,33 Several countries, such as Bangladesh, Brazil, India and Viet Nam, have reported high influenza activity that coincides with rainy seasons,15,16,19,25 while other countries, such as the Philippines, Singapore and Thailand, have reported semi-annual peaks that are not necessarily associated with rainfall.13,24 Thus, the seasonality of influenza in the tropics and subtropics appears to be country-specific.

Throughout the surveillance period, influenza virus types A and B co-circulated across all countries included in this study.13 In 2009, influenza A(H1N1)pdm09 virus appeared in the area and persisted, with limited circulation, in 2010 and 2011. Some other countries also continued to report A(H1N1)pdm09 in 2011, and again during the winter of 2012.34,35 In 2011, A H3 virus was the subtype most frequently reported in several countries (Bangladesh, India, the Lao People's Democratic Republic and Thailand). This resurgence of subtype A H3, explained as an antigenic drift of the virus,3638 has led to the selection of A H3 Victoria for the vaccine formulation for the northern hemisphere in 2012–2013.35

During the study period, we observed that most countries in tropical and subtropical southern and south-eastern Asia (Bangladesh, Cambodia, India, the Lao People's Democratic Republic, the Philippines and Thailand) experienced peak influenza activity between June/July and October, ahead of most counties in the northern hemisphere, where influenza vaccine formulations are usually available around October (Fig. 4). Since country-specific data on influenza seasonality for most tropical and subtropical countries are sparse, vaccine licensing and use, and the timing of vaccination campaigns, have been historically regulated in accordance with their hemispheric location. Nevertheless, as influenza vaccination should precede peak activity periods in order to confer maximum protection, tropical and subtropical countries should consider starting vaccination campaigns earlier than other countries in the same hemisphere; i.e. during April to June each year for countries of southern and south-eastern Asia (Fig. 2).This is consistent with reports from Brazil stressing that the designation of influenza seasons by hemisphere may not apply to tropical and subtropical countries.15

There is no ideal vaccine timing for equatorial countries with year-round influenza activity and without defined peaks of disease activity, such as Indonesia, Malaysia and Singapore. For these countries, the best recommendation would be to use the most recent vaccine formulation recommended by the World Health Organization (WHO) in that year. In Singapore, the Ministry of Health issues advisories once or twice a year when new vaccines with a change in formulation are released; since vaccines are purchased at an institutional level, residents usually receive the most recent influenza vaccine in their regular medical visits (Raymond Lin, personal communication). For countries that cover large areas with different average temperatures, an advisable option would be to use subnational data to guide sequential vaccine purchases and to select the time for vaccination campaigns. As an example, surveillance data in Indonesia show seasonal peaks of influenza in December and January for most of the islands, while other islands have peaks of infection from May to July.21 Similarly, countries with a large latitudinal span, like India, may also benefit from using subnational surveillance data to identify locally appropriate vaccination times. A summary of current and proposed influenza vaccination timing is shown in Table 2.

Although decades of public health research and global experience have shown that vaccination against influenza confers the best protection against influenza-associated deaths and complications,5 influenza vaccines are not available as part of routine public health programmes in most low- and middle-income countries in Asia.29,39 Nevertheless, the recent pandemic alarm in 2009, plus ongoing outbreaks of emerging influenza strains in other countries, such as A(H5N1) and, most recently, A(H7N9), as well as the number of deaths and complications in some regular influenza seasons, have increased the interest in influenza control in the area. To address the growing need for more affordable and accessible influenza vaccines during a pandemic, WHO has assisted countries in Asia (India, Indonesia, Thailand and Viet Nam) in developing their domestic vaccine production capacity.40 As a result, many countries have already established influenza vaccine manufacturing capabilities and some used locally produced vaccines during the 2009 pandemic.41,42 Furthermore, a recent pilot programme of influenza vaccination in the Lao People's Democratic Republic demonstrated that public–private partnerships can improve vaccination programmes in developing countries.43

The data reported here have limitations. The surveillance data collected in the countries during the six-year period only allow analyses of trends based on the percentages of nasopharyngeal specimens that tested positive, but not the calculation of incidence rates. In addition, comparison between the countries is difficult because of differences between countries in surveillance procedures and/or data collection methods (i.e. different sampling strategies, case definitions, age group selection, outpatient versus inpatient selection and biological tests), as well as in the criteria for submitting data to FluNet. Furthermore, the reporting systems might have changed over time, especially after 2009; different countries started surveillance at different times and some expanded or changed their surveillance sites over the study period. Nevertheless, these data provide a good overview of influenza seasonality and peaks of influenza circulation over the years; therefore, we were able to discern yearly trends for influenza circulation in each country, which revealed consistent patterns applicable to several tropical and subtropical countries in Asia.

In summary, we identified some patterns of circulation of influenza viruses in 10 tropical and subtropical Asian countries that are different from the patterns commonly seen in temperate countries. An early peak of influenza activity was detected between June and October, followed in some countries by a second, winter peak in December to February. Countries located close to the equator exhibited year-round circulation of the virus. Consequently, vaccination campaigns should adapt to the natural history of the virus in each country – i.e. vaccination campaigns should start before vaccination in countries of the temperate northern hemisphere showing early peaks of infection. On the other hand, for countries without epidemic peaks, year round vaccination with updated licensed vaccines might be required. These data can help decision-makers to better target vaccination programmes and inform regulatory authorities regarding the licensing of vaccines best timed to prevent seasonal epidemics in their country.


Acknowledgements

The authors thank Eduardo Azziz-Baumgartner, Nancy Cox, Josh Mott and Tim Uyeki for their valuable comments.

Competing interests:

None declared.

References

Share