Finding the flu

Blog Author: Tara Lyon [3]

Average citizens generally don’t have much practical use for up-to-date information on where influenza outbreaks are occurring (unless you’re that guy on the bus wearing the surgical face mask). For public health officials however, tracking flu trends is key in predicting strain on health care resources, understanding outbreak patterns, and potentially identifying and preventing a developing pandemic.

Traditionally, agencies like the US Centers for Disease Control and Prevention (CDC) [4] and the Public Health Agency of Canada (PHSA) [5] have been the go-to authorities for data on flu symptoms and outbreaks.

Each Friday, Canada’s national surveillance system, FluWatch (on the PHSA site) reports data collected from labs, hospitals, physician offices, and health ministries to detect flu outbreaks, monitor strain types, and provide information that can be used by the World Health Organization to determine the best vaccine to use for seasonal flu shots.

Recent advances in web technology and social media have allowed for the development of new flu-tracking resources, providing a modern spin on tracking the spread of influenza symptoms.

While symptom tracking in the past has relied on data generated from patients who’ve sought medical care, Google Flu Trends [6] creates geographical reports based on web searches for flu symptoms, which may in some ways be a more reliable--and certainly timely--indicator of illness than traditional methods of symptom reporting.

Google Flu Trends has so far proven accurate in predicting CDC-reported trends, about 2 weeks sooner than the CDC can gather and process data to generate their reports. Recognizing the significance of Google’s results, the CDC links to the Google Flu Trends site each week in their FluView report.

A new website still in its beta phase uses data from social media networks to generate maps of where various illnesses are occurring. Sickweather.com [7] allows users to search by symptom and area to find out which illnesses are trending in their neighbourhood. Whether such “crowdsourcing” data will prove useful in flu tracking remains to be seen, but the site relies on a few factors that may seriously affect the accuracy of its data:

• It assumes that the majority of people in an area are on social media (many rural areas still do not have access to high-speed Internet, and per capita Internet usage is lower than in urban areas).

• It relies on people to be over-disclosers on social media (people who post all the gory details of their illness symptoms in their status updates/tweets).

• It assumes that someone who is down for the count with flu symptoms (fever, aches, vomiting, etc.) will be feeling up to getting on a computer and interacting on social media.

Whether you’re scouring the net for data on when flu season will hit your area, or just relying on your friends’ status updates to get a general picture of how people around you are feeling, flu prevention tips remain the same: get your yearly flu shot, develop good handwashing habits, and cough/sneeze into your elbow rather than your hand.

And please, spare us all your sputum-related status updates.