Please activate cookies in order to turn autoplay off

Attacking a new millennium's bug

Epidemiologists are using IT to track and tackle the H1N1 swine flu virus with a sophistication unavailable in earlier epidemics

Gloucester old spot piglets

Well spotted: organisations including the World Health Organisation and the US Center for Disease Control are using software to track the spread of swine flu. Photo: jiunlimited.com

1918 was a hard time for the human race. The influenza epidemic that broke out that year killed up to 40m people, ending more lives than the first world war. Now, people are drawing comparisons with H1N1 - the 'swine flu'. But as this virus continues to spread and mutate, we have one advantage on our side: information technology.

90 years ago, the closest thing anyone got to crowd sourcing was an Amish barn-raising. With the internet now connecting us more than ever before, networks are helping health workers to track the spread of the virus on a local, national and international level.

In the UK, the NHS has used its central NHS.uk website to promote awareness of the disease, and to warn the public what to do about it. The Health Protection Agency has also been working with other European agencies to help analyse the nature of the virus. It provided what it says is the first isolate of the virus for vaccine researchers to work with.

At an international level, the World Health Organisation had to move quickly to try to document the spread of the new bug. Its health technical manager for non-administrative IT systems Patrick Briand and his team rushed to roll out a daily reporting system for H1N1 outbreaks, along with a case summary reporting system. Both of them were cobbled together in little more than a couple of weeks using Microsoft Sharepoint.

"It was more prototyping for some of these applications than by-the-book development," says Briand, who was taken off his daily work routine and put into an emergency working team. Briand and his colleagues had to liaise with epidemiologists who were trying to understand what information they needed to track about this worrying new strain. Rarely has requirements gathering - the part of the software development cycle when the developers document what the users want - been conducted so quickly.

The organisation had already been working on some other applications when the outbreak occurred. A weekly reporting system called FluID, due to be rolled out any time now, will provide longer-term, broader data to help health workers understand infection trends. And an event management system will collate data from multiple sources including news sites, to provide a central point of intelligence about new infections, perhaps before they are even identified locally as H1N1.

From anthrax attacks to flu

In the US, the Center for Disease Control has been hard at work on three systems: BioSense, BioPhusion and the National Electronic Disease Surveillance System (NEDSS).

BioSense was created in 2003 in response to the anthrax attacks in the US. It retrieves data from acute healthcare institutions to help both acute healthcare workers and health administrators understand what was happening on the ground. It complements BioPhusion, which is a system to retrieve less structured information from a variety of sources, says Terry Boyd, director of the division of informatics and shared services within the National Center for Public Health Informatics at the CDC.

"It pulls information from a wide variety of sources to give a complete picture of what's happening," says Boyd. "They pull information from labs, veterinary areas for animal investigations, information about what's showing up on the news, and information about hits to the CDC web site to show how interested the public is in what's happening."

Finally, NEDSS provides a standard method for disease reporting in the US, and enables policy makers to more closely track nationally notifiable diseases, including emergency outbreaks such as H1N1.

The CDC's efforts contrast with more grassroots initiatives such as ProMED Mail, a mailing list organised by the International Society for Infectious Diseases. Originally started as a means of monitoring the potential use and effect of biological weapons, the system has expanded from a membership of 40 to more than 50,000. The site now uses RSS feeds and Twitter to send out quick reports. "Its purpose has become to be an early warning system for outbreaks of newly emerging diseases," says Larry Madoff, the service's editor, adding that ProMED Mail was the first to report publicly on SARS.

Madoff argues that this grass roots effort contrasts with those of more top-down institutions. "ProMED has long been an advocate of transparency. The official public health sector hasn't always been," he warns, pointing out that EPI-X, the CDC's communications network for disseminating epidemic information, is only open to healthcare professionals, rather than to the general public. "Ministries of health of various countries sometimes try to contain information for economic and political reasons."

One of the problems for health workers is getting all of this data into the system. The WHO, which deals with cases across the globe, has to cope with information in many different formats, says Briand. "We take the information how it comes. If it comes in an unstructured way, we have people here who can enter it," he says, adding that information may arrive in emails and spreadsheets. "We don't want the structure to be a barrier for collecting information, especially in poorer countries."

"The secret with informatics is to make it as easy as possible for folks to provide the information you're looking for, and in many cases that means doing it in the way they're already used to," agrees the CDC's Boyd. "But one thing that the Office of the National Co-ordinator for Health Information Technology is looking towards is the day when we can tie this information into electronic health records and not have to re-key any information."

That day will be fraught with privacy difficulties as the authorities work to strip identifiable patient information from the data for privacy reasons. But it would be one more step in the road to efficient epidemiological reporting. In the meantime, the many and varied reporting efforts already in place look set to help researchers mitigate the threat of this bug - even as their data shows that we're well past the point of containment.


Your IP address will be logged

  • Smart Healthcare email
  • Register here