Predicting AH1N1 2009 influenza epidemic in Southeast Europe

Aim To use the data on the AH1N1 2009 influenza epidemic in the Southern hemisphere countries to predict the course and size of the upcoming influenza epidemic in South-Eastern Europe (SEE) countries and other regions of the World with temperate climate. Method We used a comparative epidemiological method to evaluate accessible electronic data on laboratory-confirmed deaths from AH1N1 2009 influenza in the seasons 2009/2010 and 2010/2011. The studied SEE countries were Albania, Bosnia and Herzegovina, Bulgaria, Croatia, Greece, Hungary, Kosovo, Macedonia, Montenegro, Romania, Serbia, and Slovenia, while Southern hemisphere countries were Argentina, Australia, Chile, New Zealand, Paraguay, Uruguay, and South Africa. Results In influenza season 2009/2010, Southern hemisphere countries with temperate climate reported 1187 laboratory-confirmed influenza AH1N1 2009 deaths (mortality rate 0.84/100 000; 95% confidence interval [CI], 0.50-1.24). SEE countries with similar climatic conditions reported 659 deaths and similar mortality rates (0.86/100 000, 95% CI, 0.83-1.10). In the whole Europe without the Commonwealth of Independent States countries (CIS, former Soviet Union), there were 3213 deaths (0.60/100 000; 95% CI, 0.65-0.93). In 2010/2011, Southern hemisphere countries reported 94 laboratory-confirmed deaths (mortality rate 0.07/100 000; 95% CI, 0.02-0.28) or only 7.9% of the previous season. SEE countries by the end of the 11th epidemiological week of 2010/2011 season reported 489 laboratory-confirmed deaths, with a mortality rate of 0.64/100 000 (95% CI, 0.26-0.96) or 74.2% of the previous season, which was significantly higher than in the Southern hemisphere countries (χ21 = 609.1, P < 0.001). In Europe without CIS countries, there were 1836 deaths, with a mortality rate of 0.34/100 000 (χ2 = 153.3, P < 0.001 vs SEE countries). Conclusion In the 2009/2010 season, SEE countries and Southern hemisphere countries had similar influenza AH1N1 2009 mortality rates. In the 2010/2011 season, the forecast of 10% increase in total mortality in SEE countries and Europe compared with the 2009/2010 season was significantly exceeded, while the expected impact of type-specific vaccines against influenza AH1N1 2009 was not achieved. Consumption of epidemic potential has greater importance for the prognosis of the course and size of influenza epidemic than the degree of vaccine immunity.

Chi-square test was used to test the differences between data for the two seasons, using Statistica 7 software package (StatSoft, Tulsa, OK, USA).

ReSultS And diScuSSion
In Tables 1-4, the number of laboratory-confirmed influenza deaths is summarized and compared between the countries of Europe, North America, Northern hemisphere, Western hemisphere, Eastern hemisphere, equatorial area, and the world. Out of 214 world countries, with 6724.3 million inhabitants, 127 countries (3743.4 million inhabitants) adequately monitored and regularly reported on the number of laboratory-confirmed deaths.
The outbreak of influenza AH1N1 2009 in Mexico in April 2009 was announced as the first catastrophic pandemic -the 'young people killer' influenza virus -of the third millennium. With the surrounding media frenzy, the influenza scare traveled quickly around the world (13)(14)(15)(16)(17)(18)(19)(20), and frightened governments were forced to take action by activating the pandemic plans prepared after the pandemic avian influenza in 2005 (70)(71)(72)(73)(74)(75)(76)(77). However, the first findings from Mexico, together with the course of the epidemic in the Southern hemisphere countries where local winter was starting, indicated that there was no exceptional, out-of-season outbreak of dangerous 'killer virus. ' Instead, it was the usual epidemic of highly contagious influenza virus AH1N1 2009, accompanied by the usual low mortality and lethality as is the case for the ordinary seasonal influenza (15)(16)(17)(18)(19)(20)(48)(49)(50). Indeed, the mortality in the 2009/2010 season was much lower than had been predicted by the WHO and the USA Center for Disease Control (77)(78)(79). Based on the number of laboratory-confirmed deaths from influenza in Southern hemisphere countries, the expected mortality in Northern hemisphere countries, where winter was starting, was 0.84/100 000 population. It was estimated that the epidemic would be a single-wave, lasting 10-16 weeks, with the onset 2 months earlier than usual (80). It was expected that SEE countries would have lower mortality rates than the Southern hemisphere countries, due to the availability of the type-specific vaccine against influenza virus AH1N1 09 (80)(81)(82).   . This was despite the fact that in 2009/2010 season fewer than 20 000 persons (0.4% of the population) were vaccinated, due to delayed administration of influenza vaccine H1N1A 09, compared with more than 550 000 (more than 12% of population) in the second season ( Figure 1).
In 2010/2011, an even higher mortality rate than in Croatia was recorded in Greece, which also had a higher mortality in the 2009/2010 season (Tables 1 and 2). In 2010/2011, only 3.5% of Greek population was vaccinated (9). In contrast, the UK, in which more than 70% of the risk-group population was vaccinated, also had a high laboratoryconfirmed mortality in 2010/2011 (22).
Poland is the only European country that refused the purchase of type-specific vaccine against influenza AH1N1 2009 in the 2009/2010 season. If the last available vaccination coverage for Poland from 2008 is used to estimate the current vaccination coverage (10% of the population older than 65 years), it is clear that the type-specific vaccine cannot be the reason for significantly lower overall laboratory-confirmed influenza mortality in Poland (0.78/100 000)  As the current data on vaccine coverage are not available for 22 EU countries, the data for the seasonal influenza in 2008 could be used (Figure 2). Since our practice in the field shows that the desired vaccination coverage is not achieved rapidly and that the habit of vaccination is not rapidly lost, similar vaccination coverage could be expected for the 2010/11 season. In 2009/2010 season, when type-specific vaccines against influenza AH1N1 2009 were not timely available in EU countries, laboratory-confirmed deaths mortality rate in the >65 age group linearly increased as the vaccination coverage decreased (Figure 2). In 2010/2011 season, when the vaccines were available, greater vaccine coverage was not followed by a lower mortality rate (13,14,(21)(22)(23)(24)(25)(26)(27)(28)(29)(30)(31)(32)(33)(34)(35)(36)(37)(38)(39)81).
The comparison of laboratory-confirmed AH1N1 2009 mortality rates among countries, even without sero-epidemiological field investigation, and detailed analysis by time, age, disability, geographic distribution, and with no valid data on the volume of applied non-pharmaceutical and pharmaceutical measures (vaccines and antivirals for preventive purposes) demonstrated that anti-epidemic protection measures by type-specific vaccines against AH1N1 2009 influenza virus in 2010/11 in SEE, Europe, and all Northern hemisphere countries did not result in a lower expected mortality rate than in Southern hemisphere countries.
Countries of the Western hemisphere, which had a higher rate of laboratory-confirmed influenza AH1N1 2009 deaths, ie, greater active infectious immunity by natural contact with influenza virus during the 2009/10 season, and thus greater estimated consumption of epidemic potential, had a significantly smaller number of laboratory-confirmed deaths in the 2010/11 season (Figure 3) than in 2009/2010, compared with the Eastern hemisphere (Table 2, Figure 3).
With the growing evidence on the questionable efficacy of influenza vaccines (86)(87)(88)(89) and despite the lack of reliable sero-epidemiological studies, the pattern of laboratoryconfirmed AH1N1 2009 mortality indicated that Western hemisphere countries achieved a higher degree of acquired active, infectious immunity than other world regions   (Table 4, Figure 3). Also, the expected epidemiological effect of vaccine immunity was not achieved. This suggests that infectious immunity has greater epidemiological significance than vaccine immunity. This is supported by the fact that people older than 55 years had lower morbidity in all parts of the world (7-67), indicating that they had been actively immunized with the infectious virus AH1N1 2009 during their life. No less important is the fact that infectious immunity provides a high degree of crossed immunity to other influenza viruses (90), while vaccine immunity lasts no longer than a year, is highly type-specific, and the possibility of crossed immunity is minimal (88,89,91). We believe that these findings clearly demonstrate that, from an epidemiological point of view, the vaccination for influenza during its outbreak should be reserved for health-risk groups. Healthy people need not get vaccinated, especially not in mass vaccination.
Data obtained by type-specific diagnostic tests in the two AH1N1 2009 influenza seasons brought into the question the previous estimates of 250 000 to 500 000 annual deaths from influenza (77)(78)(79). It was estimated that in the 2009/2010 season, the overall laboratory-confirmed influenza mortality rate in 214 countries with 6.742 million population would be 0.57/100 000, ie 38 500 deaths (Table  4). Taking into consideration the Centers for Disease Control estimate that only a third of influenza deaths are laboratory-confirmed (48), this number could reach 115 500 deaths. So far in the 2010/2011 influenza season, which is at its very end, the final estimated death toll is 25 000.
In addition to AH1N1 2009 influenza virus, other influenza viruses A and B were in circulation in different parts of the world (92). In the countries that used laboratory confirmation of influenza deaths, the total mortality of other influenza deaths did not exceed that of the AH1N1 2009 influenza (13), suggesting that the total annual number of deaths from all types influenza viruses did not exceed 50 000 in 2010/2011. From this perspective, the estimates of 250 000 to 500 000 annual deaths should be replaced by 50 000 to 150 000 annual deaths.
By studying the natural experiment of the influenza AH1N1 2009 epidemics during two seasons and by using the number of laboratory-confirmed deaths, we can provide a more accurate and reliable prognosis for the upcoming 2011/2012 season. Because of the attenuation of the epidemic potential, countries of West hemisphere will not have epidemic occurrence of influenza a H1N1 2009 above the usual average of any seasonal influenza.
The countries of Eastern hemisphere will have a low epidemic occurrence only in the countries and areas where the total mortality rate was below 1.25/100 000 during the previous two seasons. The equatorial area countries will have the usual epidemic activity for another season but of lower intensity than in the previous season (Table 4, Figure  3). Influenza virus AH1N1 2009 will continue its circulation around the world as the common seasonal influenza virus did at the beginning of April 2009.
ethical approval: not required.
declaration of authorship: MS was the chief organizer of the project of data collection on the morbidity and mortality of influenza A H1N1 2009 in the whole world ever since the announcement of the epidemic in the April of 2009. He will continue with data collection as long as the data will be published, available, and usable. AS organized and conducted data collection form more than 250 web links from more than 180 countries, international organizations, and institutions. MM performed the statistical analysis of data and prepared a presentation on economic situation in the countries of the world, with a special emphasis on the size of the tourist traffic, which will be analyzed in a future study.
competing interests: All authors have completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: no support from any organization for the submitted work; no financial relationships with any organizations that might have an interest in the submitted work in the previous 3 years; no other relationships or activities that could appear to have influenced the submitted work.