Pandemic influenza spread rapidly throughout the Pacific despite enormous distances and relative isolation. PICTs that first identified pH1N1 were those that receive many tourists, either by cruise ship or plane, from countries where pandemic influenza had already been circulating. Only three jurisdictions did not report confirmed cases. Of these, Niue experienced an outbreak of influenza-like illness during high pandemic influenza activity in the region, so it is not possible to exclude the possibility that pH1N1 infection was present, despite several negative swab samples. Pitcairn’s tiny population of approximately 50 may have experienced a small number of influenza-like-illness cases, which went undetected, or perhaps their extreme isolation prevented introduction of the virus. Tokelau thus may have been the only country in the world that was able to keep the pandemic out, through strict maritime quarantine. To verify these findings, serosurveys could be considered.
Authorities in several PICTs placed great emphasis on instituting arrival health screening of airline passengers; in many cases, this came at significant expense and at the sacrifice of routine public health services because healthcare workers were diverted from their regular duties to staff airport arrival areas. Most of these screening efforts involved measurement of temperature, either with infrared detection devices or with plain thermometers. While it is difficult to draw any firm conclusions, it is likely that this approach was ineffective in substantially delaying arrival of influenza in any PICT, due to the low sensitivity of the temperature screening method [8
]; the fact that many passengers would have taken paracetamol or other NSAIDs to reduce a fever; and the possibility of arrival of infected passengers still in the incubation period. WHO and other agencies therefore advised focusing instead on community-level surveillance and control measures [9
Most PICTs were well-prepared for the arrival of the pandemic, as there had been a strong emphasis prior to the pandemic on pre-pandemic planning and capacity-building by WHO, USCDC, and SPC, the latter through the “Pacific Regional Influenza Pandemic Preparedness Project” (PRIPPP) funded by Australia and New Zealand. One challenge faced by PICTs during the pH1N1 pandemic was the difficulty in applying advice aimed at the global level to the unique local circumstances found in the Pacific. Notably, several PICT authorities experienced frustration at being advised that countries should not institute border closure, while at the same time being cognizant of the history of the 1918 influenza pandemic in the Pacific and the very real possibility of excluding infection by instituting strict border controls. All PICTs had pandemic preparedness plans in place, though, as in other countries, challenges were encountered when trying to apply these plans, developed for worst-case-scenarios along the lines of a pandemic of highly pathogenic avian influenza, to a pandemic with a much milder profile. For instance, several PICTs had plans that called for automatic closure of borders upon the declaration of Pandemic Phase 6. A few PICTs closed their air- and seaports to travelers for up to two weeks but found it impossible to sustain this measure for the long term. A major lesson learned, therefore, was the need to make future pandemic (or emerging disease) plans flexible enough to accommodate varying degrees of severity.
Four PICTs reported 63% of all cases of pH1N1. Each of these PICTs had laboratories able to confirm pH1N1, which likely resulted in a greater proportion of ILI cases being confirmed. Further, laboratory capacity in these PICTs resulted in their cases being reported closer to the true onset date of the case and so it appears that the pandemic arrived earlier in these countries than in countries that relied on overseas laboratories to confirm the diagnosis. Once pH1N1 was identified in a PICT, confirmatory testing in overseas laboratories was no longer considered a priority for those PICTs without laboratories, thus underestimating the number of confirmed cases in these countries.
The 2009 influenza H1N1 pandemic had a relatively low impact in the Pacific, as compared with previous pandemics such as 1918 [1
]. Nevertheless, the Pacific experienced significant morbidity and mortality, consis-tent with the experiences of other indigenous and low-resource settings throughout the world [10
]. The highest mortality rate was in the Cook Islands with 9.1 deaths reported per 100,000 inhabitants. The mortality rates in the PICTs were ten-fold higher than most other countries reported during the same period [13
]. The reasons for this are probably multifactorial, including: indigenous people and Pacific islanders have higher rates of pre-existing conditions, such as heart and lung disease, diabetes and obesity when compared with non-indigenous populations; there are relatively large numbers of children and pregnant women; access to health care and diagnostic services is often limited or delayed; and, larger family size and social networks, crowding, and poverty may increase the risk of infection. Our data also show that, during this pandemic, 19% of people who died did not have a pre-existing condition, a younger population was severely affected, and pregnant women were at risk for severe disease. These findings are consistent with other published reports [14
]. Unfortunately, because most PICTs collected demographics and risk factor data only on fatal cases of pH1N1, it was not possible to perform statistical analysis to determine which, if any, of these risk factors were significantly associated with a fatal outcome.
The computed case fatality rate (CFR) (1%), amongst laboratory-confirmed cases in the Pacific, appears high compared with published rates from other regions. However, these findings must be interpreted with caution. More severe cases of influenza are more likely to be tested and diagnosed with H1N1, which will over-estimate the CFR. Additionally, the majority of cases of influenza are not laboratory-confirmed due to limited laboratory capacity in PICTs, and limited influenza surveillance and detection of clinical case-patients in some places. Therefore, it is likely that the true number of case-patients of pandemic influenza H1N1 is far greater than diagnosed. It is also likely that not all deaths due to pandemic influenza H1N1 were recognised and reported. The mortality data have several limitations. Data collection was not standardised, as only three countries completed WHO case summary forms. The remaining case data were elicited by email or telephone and as a result data are incomplete for many case-patients. Pre-existing conditions are not well defined, so it is possible that different definitions are used within PICTs, e.g. heart disease may or may not include primary hypertension, and obesity may be defined as a body mass index greater than 30, 35 or even 40. Interpretation of the importance of pre-existing conditions such as heart disease and obesity is therefore not possible. Onset dates were estimated by reporting parties where information was not available. Finally, these data relate to a small number of cases.
Syndromic surveillance of influenza-like-illness showed a peak in cases in August and was a good predictor for subsequent laboratory confirmation of pH1N1. As with confirmed cases, the data on ILI should be interpreted with caution as ILI surveillance varied in extent between different PICTs and varied in intensity with the evolution of the pandemic. Further, the method of case ascertainment varied considerably amongst jurisdictions (for example, whether fever had to be documented or could merely be subjective), making it difficult to compare jurisdictions.