糖心TV

Described as the 'Last Great Plague' of humankind, both seasonal and pandemic influenza continue to pose major challenges both in terms of optimal response, and understanding disease dynamics. How do new strains emerge and why? When should we shut schools? How effective are antiviral agents? Such questions require advanced methods for modelling and analysis of multiple data sources.

On this page we focus on our work about Pandemic Influenza; we also work on Seasonal Influenza.

---

Influenza pandemic predictability

Despite influenza pandemics being rare events, with it currently being nearly impossible to predict the next influenza emergence event, it may be the case that the virus itself provides us with outbreak signals that should prompt us to be more prepared. Specifically, knowing whether or not influenza pandemic occurrences are uniformly random in time can inform the intervention strategies that would be best suited to reduce the risk of further pandemic events.

To investigate the predicatability of influenza pandemics, we analysed the time periods between probable influenza pandemics since 1700. We depict below three alternative timelines of reputed historic influenza pandemics that underwent consideration:

Using a Bayesian model selection approach we compared plausible model hypotheses regarding the mechanisms driving influenza pandemic occurrences. We found the weight of evidence to favour influenza pandemic emergence being history-dependent, rather than a memoryless process. Conclusions drawn may allow quantitative predictions for number of expected pandemics in a specified time period to be made.

Though the modelling approach taken relies upon limited data, so is uncertain, it provides cheap, safe and direct evidence relating to pandemic emergence, a field where indirect measurements are often made at great risk and cost.

Publications

1. Hill EM, Tildesley MJ & House T (2017)  Significance 14(6): 28-33. doi:
2. Hill EM, Tildesley MJ & House T (2017) Sci. Rep. 7: 43623. doi:

---

H1N1 pandemic

The world-wide H1N1 pandemic of 2009 was the first influenza pandemic since 1968, and hence the first influenza pandemic since predictive modelling was able to inform policy. Upto 2009, the vast majority of modelling papers and public-health planning was focused towards avian-dervied H5N1 influenza spreading from S.E.Asia. H5N1 has the potential to cause substantial loss of life, with very high mortality levels observed. In reality, the 2009 pandemic began in Mexico, and despite initial concerns, it proved to be rather mild; in fact, although there was little immunity in the population again H1N1, there were fewer fatalities from this pandemic than during a normal 'flu season.

The timing and mildness of the H1N1 pandemic raised a number of issues. In a severe pandemic it is likely that most cases will have symptoms and therefore seek medical advice or in some other way be recorded; however for a mild pandemic it is unclear what proportion of people have sufficiently severe symptoms to be recorded. Or, to put in another way, are there current few cases (most of whom have symptoms) or are there many cases few of whom have symptoms. In addition, the epidemic began in the late spring and persisted through the summer, as this is not the usual 'flu season in the northern hemisphere it was unclear whether a larger epidemic with more severe cases could be expected in the winter.

Members of the (then called) 糖心TV Infectious Disease Epidemiology Research (WIDER) group had worked on pandemic planning issues before, either considering pandemic influenza or bio-terrorist smallpox release. We were therefore happy to analyse the UK and US data as it was being collected and to feed back findings to the relevant government agencies. In particular: Matt Keeling was a member of the UK's Dept of Health SPI-M (Scientific Pandemic Influenza - Modelling) group and provided input to the Dept of Health through several groups; Leon Danon was temporarily based in Harvard (USA) during the pandemic and assisted with their analysis of the US situation; Thomas House analysed the impact of school closures on the pandemic in particular to alleviate the burden on intensive care wards.

Publications

1. Black AJ, House T, Keeling MJ & Ross JV (2013) J. Roy. Soc. Interface 10(81): 20121019. doi:

2. House T, Inglis N, Ross JV, Wilson F, Suleman S, Edeghere O, Smith G, Olowokure B & Keeling MJ (2012) BMC Medicine 10: 117. doi:
3. House T, Baguelin M, van Hoek AJ, White PJ, Sadique Z, Eames K, Read JM, Hens N, Melegaro A, Edmunds WJ & Keeling MJ (2011) Proc. Roy. Soc. B 278(1719): 2573-2760. doi:
4. Keeling MJ & White PJ (2011) J. Roy. Soc. Interface 8(58): 661-670. doi:
5. Miller JC, Danon L, O'Hagan JJ, Goldstein E, Lajous M & Lipsitch M (2010)  PLoS ONE 5(5): e10425. doi: .
6. Goldstein E, Cowling BJ, O'Hagan JJ, Danon L, Fang VJ, Hagy A, Miller JC, Reshef D, Robins J, Biedrzycki P & Lipsitch M (2010) BMC Infect. Dis. 10: 211. doi:
7. Sigmundsdottir G, Gudnason T, Olafsson O, Baldvinsdottir GE, Atladottir A, Love A, Danon L & Briem H (2010) Eurosurveillance 15(49): pii=19742. doi:
8. Lajous M, Danon L, Lopez-Ridaura R, Astley CM, Miller JC, Dowell SF, O'Hagan JJ, Goldstein E & Lipsitch M (2010) Emerg. Inf. Dis. 16(9): 1488-1489. doi:
9. Danon L, House T & Keeling MJ (2009) Epidemics 1(4): 250-258. doi:
10. Keeling MJ & Danon L (2009) British Med. Bull. 92: 33-42. doi:
11. House T & Keeling MJ (2009) Epidemiol. Infect. 137(5): 654-661. doi:
12. Lipsitch M, Lajous M, O'Hagan JJ, Cohen T, Miller JC, Goldstein E, Danon L, Wallinga J, Riley S, Dowell SF, Reed C & McCarron M (2009) PLoS ONE 4(9): e6895. doi:
13. Wearing HJ, Rohani P & Keeling MJ (2005) PLoS Med. 2(7): e174 doi:

Funded by: MRC, EPSRC

SBIDER people involved:

Matt Keeling

Internal collaborators:

Andrew Easton

External collaborators:

(Manchester)

(LSHTM)

(Adelaide)

(Lancaster)

Ken Eames (formerly LSHTM)