Modelling aims to predict virus outbreaks


Japanese data helps refine disease behaviour, potentially aiding public health efforts. Paul Biegler reports.


Advance warning is critical to public health efforts to contain virus-generated epidemics.
Advance warning is critical to public health efforts to contain virus-generated epidemics.
Motortion / Getty Images

China has a big problem with Hand Foot and Mouth disease (HFMD). According to the World Health Organisation, there were over 13,000 cases in February with one fatality, although that’s an improvement on the more than 130,000 cases reported in November 2017.

Like many infectious diseases, HFMD picks out young children, in whom it causes fever, painful mouth blisters, lung infection and, rarely, meningitis, paralysis and death. Its explosive prevalence in China is perhaps why citizens there can access a vaccine that remains, thus far unavailable in countries including Australia and the US.

A major challenge for the public health response to the enteroviruses that trigger HFMD is predicting when outbreaks will happen, and the signature or “serotype” of the culprit virus. Such information would be a boon for planning prevention measures such as vaccination, hand hygiene and mask use.

That cause has now been advanced by research, published in the journal Science, from Margarita Pons-Salort and Nicholas Grassly of the Department of Infectious Disease Epidemiology at Imperial College London in the UK, who have developed a model for predicting epidemic outbreaks.

The task is a tricky one, not least because viruses peak at intervals that vary between countries and across the various serotypes. For example, enterovirus A71, a major cause of HFMD, has a three-year cycle in Malaysia but a one year one in China. Complicating matters, HFMD itself can also be caused by other viruses, such as those from the Coxsackie group.

The task is made even more pressing by the wide repertoire of illness caused by enterviruses, which takes in nasties such as meningitis, heart inflammation and polio.

To build their model, the researchers sourced data from Japan covering the years 2000 to 2016. That nation sets itself apart with a proactive screening program of patients for four diseases caused by enteroviruses: viral meningitis; HFMD; herpangina (painful mouth ulcers) and haemorrhagic conjunctivitis (commonly known as pink eye but, in severe cases, the conjunctiva can literally bulge with blood).

They homed in on the finding that one of the offending viruses, Coxsackie A4, switched from a one-year cycle to a two-year cycle after 2004, theorising that Japan’s declining birth rate was to blame. The logic here is that youngsters yet to become immune through contact with the virus are more likely to harbour it. At times when there are fewer young folk around, the remaining population comprises a mostly immune population that resists infection, hence the longer cycle between disease spikes.

Plugging in data on known immunity, transmission rates, and birth and death rates, the researchers devised a model to test that hypothesis and to predict the behaviour of 20 different enteroviruses across the data collection period. The result was, mostly, successful.

“The model fitted the 15 years of incidence data remarkably well for 18 of the 20 serotypes examined,” the authors report.

They suggest several possibilities for the aberrant behaviour of the remaining two, including a viral evolution that altered their ease of transmission, ability to generate an immune response or manifest as disease – many enteroviruses infect without producing symptoms.

In an accompanying perspective article, Birgit Nikolay and Simon Cauchemez from the Institut Pasteur in Paris, France, note the facility to predict outbreaks may spur countries with less robust reporting to institute better surveillance of these diseases, with clear benefits.

“In the future, models could be used to anticipate when the next enterovirus outbreak might occur, how large it will be, and which serotype might cause it,” they write. “This information could help healthcare providers to better prepare for the detection of cases and the provision of appropriate care.”

Contribs paulbiegler 2.jpg?ixlib=rails 2.1
Paul Biegler is a philosopher, physician and Adjunct Research Fellow in Bioethics at Monash University. He received the 2012 Australasian Association of Philosophy Media Prize and his book The Ethical Treatment of Depression (MIT Press 2011) won the Australian Museum Eureka Prize for Research in Ethics.
  1. http://www.wpro.who.int/emerging_diseases/hfmd_biweekly_20180327.pdf
  2. http://www.wpro.who.int/emerging_diseases/hfmd_biweekly_report_20180102.pdf
  3. https://www.tandfonline.com/doi/abs/10.1080/21645515.2018.1442997
  4. http://healthywa.wa.gov.au/Articles/F_I/Hand-foot-and-mouth-disease
  5. https://www.cdc.gov/hand-foot-mouth/about/prevention-treatment.html
  6. https://emedicine.medscape.com/article/215241-overview
Latest Stories
MoreMore Articles