Geographic representativeness for sentinel influenza surveillance: implications for routine surveillance and pandemic preparedness

Objective: To review the guidelines for geographic representativeness applied to sentinel influenza surveillance as proposed in the Framework for an Australian Influenza Pandemic Plan (1999). Methods: The number of sentinel practices, participating general practitioners and their consultation rates per 100,000 population, by region, were described for the Victorian sentinel surveillance system for 2003 and 2004. Influenza-like illness rates per 1,000 consultations were calculated for all participating practices and for a subset of regular participators. Indicators of seasonal influenza activity, set according to pre-defined thresholds, were compared in the two groups. Results: During these two influenza seasons, a subset of approximately one-quarter (27%) of participating practices provided almost half (45%) of the patient swabs and detected the same level of influenza activity over two influenza seasons as all participating practices. However, this subset of GPs recorded only 0.3% of all GP consultations in Victoria in 2004. Conclusions: There should be an updated, evidence-based strategy for interpandemic influenza based on the number of general practice consultations. Requirements for surveillance during various pandemic phases also need to be reviewed.