Influenza epidemiology and vaccine effectiveness following the 2009 pandemic

Abstract

Influenza A(H1N1)pdm09 was identified in April 2009 and spread rapidly around the globe. The public health response in Victoria was undertaken in accordance with the Australian Health Management Plan for Pandemic Influenza (AHMPPI) and included intensive case follow up, school closure, antiviral distribution and a vaccination program. However, evidence soon emerged that most cases were relatively mild compared to previous pandemics. This thesis sought to assess how the epidemiology of influenza A(H1N1)pdm09 differed from expectations in pandemic planning and how the control measures of school closure and antiviral distribution within the AHMPPI were applied and performed, and to investigate the role of infection severity in driving the initial spread of influenza A(H1N1)pdm09. It also sought to examine how the epidemiology of seasonal influenza in Victoria changed following the emergence of influenza A(H1N1)pdm09, and measure the effectiveness of influenza vaccine in prevention of laboratory confirmed influenza infection prior to, during and following the emergence of influenza A(H1N1)pdm09. Investigation of these questions utilised a variety of methodological approaches, including: analysis of influenza-like illness (ILI) and laboratory confirmed influenza surveillance datasets in general practice, locum service, hospital, notifiable disease and reference laboratory settings; systematic review of the literature on influenza A(H1N1)pdm09 viral shedding; deterministic mathematical modelling; and application of sentinel surveillance influenza laboratory testing data to a novel variant of the traditional case control study design to measure vaccine effectiveness. Although it spread rapidly and primarily affected younger age groups, influenza A(H1N1)pdm09 morbidity and mortality were mild compared with previous pandemics. However, the intensity of the public health response was not commensurate with the severity and magnitude of the disease. Transmission of influenza A(H1N1)pdm09 was largely driven by those effectively invisible to the health system and the virus was therefore well-established by the time it was detected. The delay in detection and high proportion of relatively mild infections meant that school closures and antiviral distribution to notified cases and their contacts were ineffective. Pandemic plans need to be revised to accommodate such a scenario and ensure trust from public and professionals in future pandemic responses. Influenza A(H1N1)pdm09 replaced the previously circulating seasonal A(H1N1) and remained dominant in Victoria in 2010. Higher proportions of A(H3N2) and type B influenza were observed in 2011 before dominance of A(H3N2) in 2012, accompanied by an increase in severe infections in older people especially. Whilst ILI surveillance suggested influenza seasons of moderate magnitude from 2010-2012, notifiable disease surveillance indicated a considerable increase in influenza testing by medical practitioners. Influenza vaccine effectiveness (VE) in Victoria varied considerably in the years preceding, during and following the 2009 pandemic. With the exceptions of high influenza A(H1N1)pdm09-specific seasonal VE in 2010 and 2011, and no protective effect of seasonal vaccine against influenza A(H1N1)pdm09 in 2009, type and subtype-specific VE were inconsistent across seasons, and had little correlation with the percentage match between circulating and vaccine strains. Further investigation of the role of previous immunity and antigenic similarity by phylogenetic analysis is needed to better understand the determinants of influenza VE.