About FluTracking
Welcome to FluTracking.net, the project which harnesses the power of the internet for tracking influenza. By taking part, you'll not only be contributing to scientific research, you will be helping to track influenza in your local community and Australia-wide. Over the 6 years the survey has been running we have grown to over 10,000 participants per week who have collectively completed 875,500 surveys, and have had 117 years off work due to influenza-like illness!
A simple online survey that takes less than 15 seconds each week during flu season can tell us so much.
The FluTracking Program
FluTracking was launched in the Australian winter of 2006 with 400 "Flutrackers" completing a 10 - 15 second online survey about flu-like symptoms each week.
In the winter of 2007 this increased to approximately 800 participants, and then over 1600 people in the winter of 2008. In 2010 and 2011 this increased to over 10,000 participants a week.
For some information on participant demographics please click here
By comparing the rate of symptoms between vaccinated and unvaccinated particpants we were able detect when influenza struck because the unvaccinated people had much higher rates of illness than the vaccinated people.
The main aims of Flutracking are to develop a system that can rapidly determine:
- the onset of influenza in Australia and subregions
- the effectiveness of the influenza vaccine
- the severity of circulating influenza strains
- if influenza strains have changed
A burden of illness pyramid is method for estimating the relationship between influenza-like illness at the community level with national influenza laboratory reports. The figure and table below describe surveillance levels from cough and fever through to positive laboratory test for influenza (self-reported) among 1,811 Flutracking participants nationally, for the five weeks beginning week ending 18/9/11 to 16/10/11.
Our Burden of Illness Pyramid is available hereWeekly Survey
After the volunteers are recruited and provide informed consent they receive a weekly email from us during the traditional influenza season (usually from May 1 to October 30). The email contains a link to an online survey form.
On the first visit they provide the following information:
- Date of birth
- Receipt of influenza vaccine in the current and the preceding year
- Face to face contact with patients
- Postcode of residence
On subsequent weekly visits to the form, they are asked about the presence of typical flu-like symptoms like:
- Fever
- Cough
- Number of days absent from normal duties
- Muscle aches (deleted for 2008)
To see a short video of the survey screens see below:
2011 Survey Results
There were some small peaks of influenza-like-illness activity in June and August but overall this year was a fairly mild flu season.
2008 Survey Results
In 2008 we hoped to double our recruitment to approximately 1500 participants, however, by the first day of the winter survey (5th May 2008) we had 1,666 participants registered to answer on behalf of themselves and another 652 household members - for a total of 2,318.
The facility to allow a participant to answer on behalf of other household members was new in 2008 and is responsible for much of the increased recruitment. Given that some participants would be answering for multiple people, we reduced the number of questions on the survey to make it faster to complete.
Participants were only asked about cough and fever, if they answered yes to either then an additional question asks them about absence from work or normal duties. We deleted the question about muscle aches and acute onset of symptoms as it was felt that they contributed little to the discrimination of influenza from other respiratory illness.
Several large organisations agreed to circulate the invitation to participate across their corporate email systems, including faculties at the University of Newcastle and the Tasmanian Department of Health and Human Services. Excellent uptake within Tasmania should provide a useful comparison with findings from NSW.
The chart below shows Fever and Cough by Vaccination Status to 7th September 2008.
2007 Survey Results
2007 was a bad year for influenza, and despite rumours circulating in the media that the influenza vaccine "was not working", FluTracking data suggested it was protecting vaccinated people at least as effectively as in 2006.
Participation has increased over the 2 years of the FluTracking pilot. The most effective method of recruiting participants appears to be through invitations to join sent via corporate or government email systems.
Weekly survey emails are usually sent out on Monday mornings of those participants who respond, 50% answer the survey within 3 hours, 80% answer within 24 hours and 93% answer within 7 days of the email dispatch.
The Figure below compares the rate of fever and cough among vaccinated and unvaccinated participants and the peak week of influenza activity detected by other surveillance systems for influenza.
Indicators of influenza activity peaked in the week ending 20/7/07 for NSW laboratory influenza antigen tests; the week ending 17/8/07 for NSW Emergency Department ILI surveillance; and the week ending 24/8/07 for both NSW Death Register data for influenza and pneumonia related deaths and NSW laboratory influenza serology testing.
Rates of cough and fever among vaccinated and unvaccinated FluTracking participants began diverging substantially in late June and more markedly in mid-August suggesting significant influenza activity.
In future influenza seasons FluTracking will focus on early season surveillance by commencing in later April or early May.
