The above results take other local interventions into account only in that the reproduction number in Australia is assumed to be the same as in the source region. That is, Australian interventions other than border control measures are assumed to be the same as in the source region.

i Limitations

The results are strongly influenced by the assumption of a fixed 2 day incubation period. This, in combination with the peaked infectiousness function, results in nearly half an infected person’s offspring being generated before the onset of symptoms. This puts upper bounds on the effectiveness of both border screening and early presentation in preventing an epidemic being initiated. In slight contrast, Ferguson et al. (2005) assume a variable incubation period, with a mean of 1.5 days, and infectiousness spiking immediately following this, though rapidly declining thereafter. Under their model, very early presentation (i.e. several hours or less) following symptom onset reduces transmission more effectively, although border screening will remain similarly ineffective. If the incubation period is variable, partial home quarantine of arriving travelers may not be as effective because, for some infected individuals, the quarantine period may be over before onset of their symptoms.

ii Further work needed

One important aspect that has not yet been included in our calculations, is the fact that resources available in States and Territories can target cases very effectively while the number of cases is small, but much less effectively when the number is large. In other words, achievable intervention can reduce R effectively (most likely bringing it below 1) during the very early stages while the number cases is small, but such a level of intervention can not be achieved once transmission gathers momentum. Calculations that incorporate this real-world constraint should be conducted.

iii Contrasting the results with those for SARS

Border control measures, such as border screening and active surveillance of arrivals from at-risk regions, have greater potential to delay a local epidemic for any respiratory infection with a very short incubation period and for which infectiousness is not appreciable prior to the onset of symptoms. The latter is true for SARS, but border screening is not effective for SARS because its asymptomatic latent period lasted for several days. Empirical evidence that border control measures were not particularly effective for SARS are given by Wilder-Smith et al. (2004) and Samaan et al. (2004).

Document download

This publication is available as a downloadable document.

Using Mathematical Models to Assess Responses to an Outbreak of an Emerged Viral Respiratory Disease(PDF 873 KB)