[Source: World Health Organization, Weekly Epidemiological Record, full PDF document: (LINK). Extracts.]

Weekly epidemiological record / Relev? ?pid?miologique hebdomadaire / 29 NOVEMBER 2013, 88th year / 29 NOVEMBRE 2013, 88e ann?e, No. 48, 2013, 88, 509?520, http://www.who.int/wer

Review of the 2013 influenza season in the southern hemisphere

This report summarizes the epidemiology and virology of the winter influenza season in the temperate regions of the southern hemisphere and the countries in the tropical regions of South and Central America, from January to the end of September of 2013. The data presented have been derived primarily from reports published by national ministries of health or other official bodies reporting on their behalf, or reported to WHO through Flu-Net and FluID.(1, 2)

Southern cone of South America


The timing of the 2013 influenza season in the countries of the southern cone of South America was similar in Argentina, Paraguay and Uruguay, but earlier in Chile.

Argentina, Paraguay and Uruguay started to report active influenza transmission at mid-May, with a peak in early July. No virus was detected by early September. In Chile, this year?s influenza season began and peaked a month earlier than in other countries in the southern cone, however influenza virus activity was still detected at the end of September.

In Chile, this year?s influenza season was comparable to the historical 10-year average (excluding 2009, Figure 1). The peak of the influenza season in Argentina was about 2 months earlier and level of influenza activity was much higher compared to the historical 10-year average (excluding 2004 and 2009, Figure 1). For Paraguay and Uruguay, historical information was insufficient to make an appropriate comparison.

Overall, influenza A(H1N1)pdm09 virus predominated in the southern cone of South America, with the exception of Paraguay, where influenza A(H3N2) was detected.

A small peak of influenza A(H3N2) activity was also seen towards the end of the season in Argentina and Chile. Influenza type B virus was detected at lower levels in all 4 countries.

In Chile, during the period of January through August of 2013, influenza A accounted for 86% (1651/1919) of influenza viruses reported and influenza B accounted for 14% (268/1919). Of the influenza A viruses subtyped, 71% (1300/1820) were A(H1N1)pdm09 and 19% (351/1820) were A(H3N2).

In Argentina, during the period from January to the end of August 2013, influenza A accounted for 97% (5037/5352) of influenza viruses reported while influenza B represented only 3% (133/5352). Of the influenza A viruses subtyped, 81% (2716/3336) were A(H1N1) pdm09 and 19% (620/3336) were A(H3N2).

In Uruguay the pattern was similar to that seen in Argentina during the same period ? of the 72 influenza viruses detected, 99% were influenza A and 1% were influenza B. Of the influenza A viruses subtyped, 96% (64/67) were A(H1N1)pdm09 and 4% (3/67) were A(H3N2).

In Paraguay, during the period from January through August of 2013, influenza A accounted for 84% (876/1049) and influenza B for 16% (163/1049) of positive samples.

In contrast with other countries, influenza A(H3N2) was predominant in Paraguay and accounted for 88% (775/876) of influenza A viruses detected, while A(H1N1) pdm09 accounted for 12% (101/876).(3)

Morbidity and mortality

In Chile, this year?s influenza burden was much higher than compared to previous years. Influenza-like illness (ILI) peaked at the end of June, with a rate of 28.5 consultations per 100 000 population, which was higher than the past 7 years, with the exclusion of 2009 and 2010. Emergency department visits for a respiratory cause were comparable to the numbers seen in 2010?2012 for both adults and children, with a peak in early July.

Sentinel severe acute respiratory illness (SARI) activity was also much higher in 2013 compared to the 2011 and 2012 influenza seasons, with 686 influenzaassociated SARI cases in 2013 (January?August), compared to 331 and 187 influenza-associated SARI cases during the same period in 2012 and 2011 respectively.

Of the SARI cases positive for influenza reported from January through August of 2013, 18% required admission to intensive-care units (ICU). This percentage was higher than seen in 2012 (12%). The number of deaths reported to date in 2013 (n=70) was higher than the total reported in 2012 (n=19). Of the 70 reported deaths associated with influenza in 2013, 86% (60/70) were found to have A(H1N1)pdm09.(4)

In Paraguay, this year?s influenza burden was slightly more severe than last year, with a higher SARI rate and lower ILI rate. Nationally, ILI peaked in mid-July with around 11 700 cases, which was fewer than last year but higher than levels seen in 2010?2011. SARI activity was higher in 2013 compared to the 2012 influenza season, with 339 influenza-associated SARI hospitalizations in 2013 (in the period January?August 2013), compared to 250 influenza-associated SARI hospitalizations during the same period in 2012. Of the influenza-associated SARI hospitalizations, 35% were aged 15?59 years and 18% (60/339) of the influenza-associated SARI hospitalizations required admission to ICUs. Of the 29 influenza-
associated SARI deaths reported to date in 2013, 69% (20/29) were positive for A(H3N2).(5)

In Argentina, the cumulative numbers of cases of ILI, SARI and pneumonia reported at the end of August 2013 were lower than levels reported for the same period in 2011 and 2012. The majority of all ILI, SARI and pneumonia cases were seen in children ≤4 years of age. Similar to last year, Argentina also had a high detection rate of RSV in 2013, although lower than in 2012. Between January and August 2013, RSV accounted for approximately 61% of all laboratory-confirmed respiratory viruses and influenza accounted for 27% (compared to 90% and 4% in 2012, respectively).(6)

In Uruguay, SARI activity peaked at the same time as the positive influenza detections, in early July, with around 140 SARI cases in sentinel hospitals (approximately 3% of the total hospitalizations). The SARI activity and the timing of the peak was similar to the activity seen in 2012 (around 130 SARI cases and approximately 3% of the total hospitalizations), but lower and earlier than seen in 2010 (around 305 cases and 4% of the total hospitalizations with the peak in early August).

Tropical South America


In tropical South America, active influenza virus transmission was first noted in early March with 2 peaks during the season ? the first peak in mid-June, mainly driven by Brazil, Bolivia (Plurinational State of), Colombia and Venezuela, and the second peak in mid-July, mainly driven by Ecuador and Peru. The influenza season ended in most countries at the end of August, but the season for Ecuador and Peru ended in late September.

The predominant influenza virus in tropical South America was A(H1N1)pdm09, however A(H3N2) and influenza B viruses were also circulating throughout the region. In Colombia, Brazil, Peru and Venezuela, A(H1N1)pdm09 clearly predominated throughout the season. In Brazil co-circulation of A(H1N1)pmd09 and influenza B was observed, whereas in Colombia the co-circulating virus was mainly A(H3N2). In French Guiana, January?July, the most detected virus was influenza B, at 75% (152/204). Of the subtyped influenza A viruses in French Guiana, 84% (38/45) were A(H1N1) pdm09 and 16% (7/45) were A(H3N2). In Bolivia (Plurinational State of) all 3 influenza viruses were in circulation in January?September; A(H3N2) appeared at the beginning of the season, A(H1N1)pdm09 appeared towards the end and influenza B was seen throughout the season Influenza B was also reported in some areas of Brazil and Peru. In Venezuela, influenza activity peaked in late May/early June and A(H1N1)pdm09 predominated.

The influenza season in Ecuador was characterized by 2 peaks; A(H3N2) predominated in the first peak from the end of March until early June; A(H1N1) pdm09 predominated in the second peak in early August.(7)

Morbidity and mortality

In Colombia, both the acute respiratory infection (ARI) associated ICU admissions and ARI-associated hospitalizations peaked in mid-June and were slightly lower than the rates seen in 2012. Most influenza SARI hospitalizations occurred in patients aged 15?59 years.

In Ecuador, the percentage of SARI-related hospitalizations increased from 4% in mid-March to 8% during the first peak in mid-April, which was mainly limited to the southern region of the country. This was associated with circulation of RSV in children <5 years of age. Subsequently, SARI activity remained high while influenza activity increased in the west coast and central highland areas of the country, where a second peak was observed at the end of July (8% of SARI hospitalizations), with the 15?54 year age group being the most affected. The percentage of SARI-related ICU admissions increased from early June and peaked in mid-July at 29%.

The number of SARI deaths up to the 22 September (n=175) was lower than the number that were reported in 2012 (n=250).

In Peru, the number of ARI infections and pneumonia cases in children aged <5 years followed the same trend as was seen in the last 5 years. An increase was reported from the end of January/early February with a peak in early June; however, the pneumonia activity in those aged> 5 years increased in early May and peaked in early August, surpassing the epidemic threshold and showing higher levels of activity than during the 3 years
2010?2012. In Venezuela, ARI activity clearly showed a peak in early June above the expected levels of ARI activity reported over the last 7 years. ARI activity started to increase at the end of April and returned to expected levels at the end of July.

In Brazil, among all the SARI cases (n=30 970) seen in 2013 until the 22 September, 17.8% (n=5517) were confirmed for influenza. Among positive influenza cases, A(H1N1)pdm09 was the predominant virus (65%), followed by influenza B (21%) (reported largely since mid-May) and A(H3N2) (10.6%). The peak of SARI cases positive for influenza was seen in early June, with the south-east region of the country reporting the highest level of activity (almost half of all reported cases were from this area).

Among all the SARI-related deaths in 2013 reported until the 22 September (3469), 25% were confirmed for influenza. Among SARI deaths that were confirmed as positive for influenza, A(H1N1) pdm09 was the predominant virus (82%). The age group with the highest percentage of deaths with positive influenza samples was the 40?49 year age group, with a median age of 49 years (0?98 years).

Central America and the Caribbean


In Central America and the Caribbean, influenza activity was first detected in Cuba, the Dominican Republic and El Salvador at the beginning of April 2013. The influenza season in Cuba and the Dominican Republic was mainly characterized by A(H1N1)pdm09 transmission, whereas in El Salvador influenza A(H3N2) was identified as the predominant virus (representing 65% of positive influenza samples). In all 3 countries influenza transmission peaked in early June. At that time, the influenza season had started in Costa Rica, Nicaragua and Panama. Influenza activity was highest in these countries in early July and by the beginning of August positive influenza detections had decreased. In all these countries, co-circulation of A(H1N1)pdm09 and A(H3N2) viruses was reported, but in Costa Rica A(H1N1)pdm09 virus predominated (56% of the influenza positive samples), while in Nicaragua and Panama influenza transmission A(H3N2) virus predominated (60% and 70% of the influenza positive samples respectively). In Honduras, only a few samples were positive for influenza, mainly influenza B virus. The influenza season in Guatemala and Puerto Rico was characterized by 2 peaks, the first in early February and the second in August.(8)

Morbidity and mortality

In Honduras, SARI activity was still increasing in late August, whereas in 2012, the peak occurred in early August. SARI activity in Honduras was most likely due to respiratory syncytial virus (RSV), as low levels of influenza viruses were detected. Infants aged <1 year were most affected by SARI. In El Salvador, activity peaked in June 2013 and occurred slightly later than in 2012 (when it peaked in May). In Guatemala, no obvious peak in influenza activity was observed, however in 2012 the peak was clearly visible in March. Among SARI cases in Guatemala, A(H1N1)pdm09 virus was the predominant virus and children aged <5 years were the most affected group. In the Dominican Republic, influenza activity peaked in late May/early June. The cumulative ARI cases for 2013 (through mid-September) were about 14% less than levels reported for 2012. However, the number of SARI cases were similar to last year, but the number of SARI deaths in 2013 (n=24), exceeded the total number reported in 2012 (n=5).

South Africa


In South Africa, active influenza virus transmission was first detected in mid-April and peaked in early June, followed by a small second peak in early August. For both peaks, almost 65% of all ILI samples from sentinel sites of the Viral Watch Program tested positive for influenza. Compared to the historical 8-year average (Figure 1), the main peak of this year?s influenza season in South Africa occurred a month earlier and was much lower, indicating an early and mild season in 2013. The SARI cases that were positive for influenza virus had a similar time pattern as the ILI cases.

During the period January?September 2013, among ILI cases reported, influenza A accounted for 86% of all influenza viruses detected (728/845) and influenza type B represented 13% (117/845). Of the subtyped influenza A viruses, A(H1N1)pdm09 accounted for 82% (590/723) and A(H3N2) for 18% (133/723). Influenza A(H1N1)pdm09 transmission peaked around 9 weeks earlier than A(H3N2), resulting in a marked predominance of A(H1N1)pdm09 in the first 3 quarters of the season, with transition to A(H3N2) in the final quarter of the season.

The proportion of influenza virus types was quite similar among SARI and ILI cases; among influenza-associated SARI in 2013, influenza A was found in 94% (137/146) and influenza type B in 6% (9/146) of cases. However, it should be noted that the provinces included in the ILI and SARI systems do not completely overlap.(9)

Morbidity and mortality

In the period from January to 22 September 2013, the overall influenza detection rate among ILI cases reported through the ILI surveillance system in primary health care clinics was 9.9% (173/1738), with a peak of 23.0% at the end of April/early May. Among these ILI cases, the highest detection rate was seen in the 5?24 year old age group, namely 14.5% (62/428). In the same period, the overall influenza detection rate among SARI cases reported through the SARI surveillance system was 5.8% (138/2361), with the highest detection rate among the 25?64 year age group at 7.2% (64/890). The influenza detection rate among SARI cases peaked at the end of May/early June, with a rate of 24.2%. Among the influenza-positive SARI cases with available information, 17.4% (24/138) reported having underlying medical conditions (excluding HIV) and 41.4% (41/99) had been previously diagnosed with HIV infection. For 110 influenza-positive SARI cases, information was available on their outcome: the case-fatality rate was 7.3% (8/110) during the period from January to 22 September 2013.

Australia and New Zealand


Overall, the 2013 influenza season in Australia can be described as a late and moderate season and it occurred over a shorter period of time in comparison with previous seasons. In Australia, active influenza transmission was first detected at the end of June. This was approximately a month later than was seen during the seasons of 2003?2012 (excluding 2009) (Figure 1). The number of influenza notifications peaked at the end of August, but was lower than in 2012.

In New Zealand the 2013 influenza season started and peaked around 2 weeks later than in Australia. Compared to the historical 10-year average (excluding 2009) for New Zealand (Figure 1), this year?s influenza season started and peaked a month later. The influenza positivity rate was higher this year, with a peak around 70%, compared with an average peak rate of around 55% influenza positivity over the last 10 years. However, as described below, this peak in influenza positivity was not very representative for this year?s influenza burden in New Zealand.

In Australia, in the period from 1 January 2013 to 11 October 2013, there were 22 983 laboratory confirmed notifications of influenza reported to the National Notifiable Diseases Surveillance System (NNDSS). Across Australian states and territories the distribution of influenza types and subtypes varied. However, nationally 62% of these cases were reported as influenza A (of these 41% were not subtyped, 16% were A(H1N1)pdm09, 6% were A(H3N2), and 38% were influenza B. Although the majority of influenza A detections were not further subtyped, so far in 2013 around 15% of overall notifications have been reported as A(H1N1)pdm09, compared with <1% in 2012. Additionally, the overall proportion of influenza B this season has been higher than in recent years. A predominance of influenza B infections was seen in those aged <15 years and the influenza A infections peaked in the 0?4 and 30?34 year age groups. Consistent with A(H1N1)pdm09 dominant years, there were very few notifications of this subtype in those aged ≥65 years.

In New Zealand, in the period from January to 22 September 2013, influenza A and B virus types were reported at similar rates. Influenza B predominated in early winter from May to July followed by influenza predominance of A(H3N2) from August to September.

Overall, influenza type A accounted for 55.2% (845/1532) of all viruses detected, and influenza B for 44.8% (687/1532). Of the influenza A viruses with subtype information, 74% (374/506) were A(H3N2) and 26% (132/506) were influenza A(H1N1)pdm09.

Morbidity and mortality

In 2013, the overall impact of the influenza season appeared to be somewhat lower than average in Australia. At the seasonal peak, the number of influenza cases reported per week and ILI consultation rates were lower than last year. The Australian Sentinel Practices Research Network (ASPREN) reported that in the period from 1 January 2013 to 13 October 2013, 2113 ILI patients were tested for influenza, of which 18.5% tested positive. Of those who tested positive the majority had influenza A (12.0%). The Influenza Complications Alert Network (FluCAN) sentinel hospital surveillance system reported that in the period from 30 March 2013 to 11 October 2013, 10% of influenza hospital admissions have been admitted directly to ICU. This was slightly more than in 2012 (9%), but somewhat less than in 2011 and 2010 (14% and 21% respectively). Of the influenzaassociated hospital admissions in 2013, the majority were due to influenza A (68%), and 32% to influenza B.

Around 32% of the hospitalized patients were aged ≥65 years (median 57 years) and 78% of all cases had known medical co-morbidities reported. As of 11 October 2013, 28 influenza-associated deaths had been notified to the NNDSS, with a median age of 63 years (range 27?97 years). Of these cases 24 were reported as having an influenza type A infection. The number of influenzaassociated deaths was lower than in the same period in 2012 (n=84) and in 2011 (n=40). However, the number of influenza-associated deaths reported to the NNDSS was reliant on the follow-up of cases to determine the outcome of their infection and was therefore most likely to be an underestimate of the true mortality impact associated with this disease.(10)

In New Zealand, national sentinel general practitionerbased ILI surveillance indicated that the consultation rate in the 2013 New Zealand winter was at a low level compared to that of the past 22 years during 1992?2013. The 2013 consultation rate of 47.1 per 100 000 was the second lowest compared to the lowest rate (41.7 per 100 000) in 2000.(11)

As part of the Southern Hemisphere Influenza and Vaccine Effectiveness Research and Surveillance (SHIVERS) project, ILI and SARI surveillance were established for the Auckland and Counties Manukau District Health Boards (ADHB and CMDHB) with a population of 838 000. From 29 April to 29 September 2013, a total of 2016 ILI cases were reported to the ADHB and CMDHB in New Zealand. Of these cases 24.9% (448/1802) tested positive for influenza, mainly influenza A (60% (268/448), and of those with subtype information 85% (180/213) were A(H3N2)). The highest ILI-associated incidence in these districts was reported among those aged 5?19 years (639.7 per 100 000), among cases of Asian origin (573.2 per 100 000), and among persons with the highest socioeconomic status (SES) (NZDep1?2) (498.2 per 100 000). In the same period, 59 688 patients were urgently admitted to the hospitals in the ADHB and CMDHB hospitals.

Of these, 3537 with suspected respiratory infections were assessed. Among them, 1642 (46.4%) met the SARI case definition and of those 17.04% (215/1642) had positive influenza virus results. This detection rate was slightly lower than in 2012 (22%). Among the hospitalized SARI cases, SARIassociated influenza incidence was highest among infants aged <1 year (122 per 100 000), followed by the 65?79 year-old age group (72.3 per 100 000). The highest SARI-associated influenza incidence was reported for Pacific Peoples (50.6 per 100 000), followed by Maori (26.8 per 100 000). Furthermore, the highest SARI-associated influenza incidence was reported in persons with the lowest SES (NZDep9-10)(51.2 per 100 000). Of the SARI cases, 66 were admitted to ICUs (3.8%) and 7 deaths (0.4%) were reported during this period. Of those admitted to ICUs and those who died, 15.6% and 16.7% were positive for influenza respectively.(12)

Antigenic characterization

All influenza A(H1N1) viruses detected globally since January 2010 were A(H1N1)pdm09-like; no detections of seasonal A(H1N1) have been reported since 2009.

The majority of influenza A(H1N1)pdm09 viruses characterized by the WHO Global Influenza Surveillance and Response System were antigenically similar to the vaccine virus A/California/7/2009, contained in the trivalent seasonal influenza vaccine for the 2013 southern hemisphere season. The majority of recent A(H3N2) viruses have remained antigenically similar to the A/Texas/50/2012 virus and cell-propagated A/Victoria/361/2011-like viruses. For the influenza B viruses, the proportion of B/Yamagata/16/88 lineage viruses increased in many parts of the world. Most recently isolated B/Yamagata/16/88 lineage viruses were antigenically similar to B/Massachusetts/2/2012-like viruses. This finding led to a recommendation for change in vaccine composition for the 2013/14 northern hemisphere seasonal influenza vaccine.(13)

Antiviral sensitivity testing

The majority of A(H1N1)pdm09 viruses tested were sensitive to oseltamivir and all were sensitive to zanamivir. All of the oseltamivir-resistant influenza A(H1N1)pdm09 viruses detected had the H275Y substitution in the neuraminidase gene. The great majority of A(H3N2) and B viruses tested were sensitive to oseltamivir, peramivir and zanamivir, with the following exceptions: 3 A(H3N2) viruses which showed highly reduced sensitivity to zanamivir due to the E119V substitution in the neuraminidase gene, 2 B/Victoria-lineage viruses which showed highly reduced sensitivity to peramivir due to the H273Y substitution in the neuraminidase gene, and a small number of B/Victoria-lineage viruses which showed reduced sensitivity either to oseltamivir or to both oseltamivir and peramivir. In addition, all influenza A viruses tested remained resistant to adamantanes.(13)


The 2013 influenza season in the temperate countries of the southern hemisphere varied considerably from country to country with regard to timing and circulating influenza virus types. Regarding timing, the influenza season in the countries in the temperate and tropical regions of South America and South Africa started around March?May and ended around August? September. In Australia and New Zealand the season was several months later, starting in June/July and ending at the end of September/early October. With regard to the circulating influenza virus types and subtypes, overall A(H1N1)pdm09 was the most commonly detected virus. However, towards the end of the season, influenza A(H3N2) and B viruses were identified more often in Argentina, Chile, South Africa and Uruguay.

Influenza A(H3N2) virus was also commonly detected in Paraguay throughout the season. In Australia and New Zealand all 3 seasonal influenza viruses circulated simultaneously during the season. Influenza A(H1N1) pdm09 also predominated in the tropical and central regions of South America, with co-circulation of A(H3N2) and B viruses: in the tropical regions the predominant co-circulating virus was influenza type B, whereas in the central region the predominant co-circulating virus was A(H3N2).

Overall, the 2013 influenza season was milder than in 2012, particularly in Australia and New Zealand. However there were exceptions. In Chile, ILI activity, and especially SARI activity, was much higher (almost double) compared to previous years. This also resulted in more influenza-associated deaths in Chile. The SARI rate was also higher this year in Paraguay. As in years when influenza A(H1N1)pdm09 was predominant, little severe disease was seen in the >65 year age group. Severe disease was seen in the elderly only in countries where influenza A(H1N1)pdm09 was less predominant (Australia and New Zealand).

In most countries the virological data on transmission and epidemiological data on illness and mortality showed a similar pattern. However, in several countries there was a contradictory pattern. From the virological data from Chile (Figure 1), influenza positivity was relatively low compared to the previous years, but the number of influenza- associated SARI cases almost doubled, indicating a relatively severe season. A contrary situation was seen in New Zealand where the maximum level of influenza positivity reached during this year?s season was higher than the average influenza positivity seen in recent years. However, the illness and mortality data of New Zealand show that the ILI consultation rate barely reached the baseline level of activity and was measured at its lowest level since 2000.

These contradictory patterns illustrate that virological and disease activity data are complementary and necessary to provide a comprehensive picture of the influenza season in a specific country.

With increasing numbers of countries carrying out ILI and SARI surveillance, useful data are being generated to reveal patterns and trends of influenza activity over time within countries. However, as surveillance methods differ considerably among countries, the data require careful interpretation for comparison across countries.

More consistent reporting and the use of identical case definitions and common surveillance protocols would facilitate this comparison and improve understanding of influenza patterns around the world. 
  1. FluNet can be accessed online at http://www.who.int/influenza/gisrs_laboratory/flunet/en/; accessed October 2013
  2. FluID can be accessed online at http://www.who.int/influenza/surveillance_monitoring/fluid/en/; accessed October 2013
  3. PAHO Regional Influenza Surveillance is available at http://ais.paho.org/phip/viz/ed_flu.asp; accessed October 2013.
  4. For additional information, see the Chile progress report. Available at http://epi.minsal.cl/vigilancia-epidemiologica/enfermedades-de-notification-obligatoria/influenza/; accessed October 2013.
  5. For additional information, see the Paraguay epidemiological bulletin. Available at http://vigisalud.mspbs.gov.py/index.php?option=com_content&view=category&id=10&I temid=101; accessed October 2013.
  6. For additional information, see the Argentina monitoring bulletin. Available at http://www.msal.gov.ar/index.php/home/boletin-integrado-de-vigilancia ; accessed October 2013.
  7. PAHO Regional Influenza Surveillance is available at http://ais.paho.org/phip/viz/ed_flu.asp; accessed October 2013.
  8. PAHO Regional Influenza Surveillance is available at http://ais.paho.org/phip/viz/ed_flu.asp; accessed October 2013.
  9. For additional information, see the Weekly Surveillance Report of the National Institute of Communicable Diseases. Available at: http://www.nicd.ac.za/?page=seasonal_influenza&id=72; accessed October 2013
  10. For additional information, see the Australian influenza report. Available at: http://www.health.gov.au/internet/main/publishing.nsf/content/cda-surveil-ozfluflucurr.htm; accessed October 2013
  11. For additional information, see the Influenza Weekly Update. Available at: http://www.surv.esr.cri.nz/virology/influenza_weekly_update.php; accessed October 2013
  12. For additional information, see the Southern hemisphere influenza vaccine effectiveness research and surveillance report. Available at http://www.esr.cri.nz/competencies/shivers/Pages/SHIVERSReports.aspx; accessed October 2013
  13. See No. 44-45, 2013, pp.477-488