With this year’s flu season nearing epidemic levels — it’s widespread in 43 states, according to the Centers for Disease Control and Prevention (CDC) — prevention (and preparation) is certainly your best medicine.
Fortunately, there is no shortage of apps and websites that can help. Flu tracking is a popular subject, and ever since Google Flu Trends launched in 2008, it seems everyone wants to know how prevalent flu is — not just in their state and in their city, but in their neighborhood and even their office building. (Facebook can even help you figure out which of your friends might have given you the flu but tracking members’ posts about the illness and its symptoms.)
And while none are perfect, there are a few good ones.
The CDC’s FluView:
It gathers data from clinics, urgent care centers, doctor’s offices and hospitals and reports the number of people coming in with flu-like symptoms — fever, cough, sore throats, muscle aches — and people who test positive for the bug. It offers a good snapshot but it has some shortcomings. Because it logs symptoms in addition to diagnosed cases, there’s a chance that some of the fever and respiratory problems could be something else entirely. It also doesn’t record the countless people who likely just decide to weather out their illness at home with over-the-counter remedies or some chicken soup. There’s also a lag in the reporting, which means FluView can’t provide a real-time look at what the virus is doing at a given time.
The CDC has an epidemiologist studying alternative ways of collecting flu information, including crowdsourcing, to see if the CDC can provide more real time data. “There is potential there,” says Lynnette Brammer, an epidemiologist in the influenza division at CDC. “But with any data set you’ve got to know how to appropriately interpret that data and when you might be over reaching.”
This app is more granular, portraying in real time the actual level of flu activity in a given area. It can pinpoint your location down to the street and give you a low-moderate-high reading on flu activity. It’s based on self reports from people who register on the app and voluntarily provide information on their sniffle status on a weekly basis. Blue dots indicate people who are still symptom-free, while yellow dots indicate people who might have some of the symptoms of flu — including fever, coughs, or sore throats — and red dots represent people who meet the CDC criteria for influenza-like illness: fever over 100F and a cough or sore throat that’s not caused by any other known infection.
Self-reporting may not provide an entirely accurate picture either, however. To address such confounding factors, Flu Near Your deletes the first two reports by newcomers, to reduce the possibility that new users are just playing around with the app, but there’s still no way to verify the symptoms that people log in. But so far, Mark Smolinski, director of global health for Skoll Foundation Global Health Threats, which created the app, is confident that the reports are valid, since they track pretty well with the CDC data.
Alexis de Belloy, who pores over the Flu Near You data, says that registration for the app is up 40% compared to last year, and the proportion of them who actively respond to the weekly surveys is also up; the more participants, the stronger and more reliable the signal generated by the users.
Jeffrey Shaman, professor of at the Mailman School of Public Health at Columbia University, decided to combine a bit of both the CDC and real-time strategies into his flu forecaster, which he launched last year. You can select your city and receive a chart of when flu is likely to peak where you live. It’s based on the same predictive modeling that weather forecasters use — combining information from past flu seasons and current trends to make educated guesses about the ebb and flow of cases in the future. “There are problems with all data, so I don’t think any data should be used in isolation,” he says. “We should consider them all until we have that gold standard.”
And it’s not just an academic exercise. Smolinski points out that critical public health information can be gleaned from flu tracking and forecasting, such as how effective a particular flu season’s vaccine is. By comparing rates of illness among people who have been vaccinated and those who haven’t, doctors can get a good sense of whether the shot is a good match for circulating flu strains or not. It can also help doctors, hospitals and pharmacies to make informed decisions about stocking flu remedies like Tamiflu so everyone who needs them will have access to them.
In Australia, crowd-based online surveillance is becoming the country’s go-to resource on flu tracking. Their system, FluTracking.net, asks participants about whether they have had a fever or cough every week during the flu season; since the program began in 2006, more than 16,000 people now complete the survey regularly. In Europe, 10 countries participate in Influenzanet, an online flu tracking system that also relies on volunteers to report on their symptoms weekly.
“It’s what’s coming down the pike, and what’s going to be in our future,” says Shaman of the real time information from the public. “We already get pollution levels and the pollen count. Why not have a real time flu forecast?”