I can say without a doubt I have had food poisoning from eating at restaurants. Two of the most severe episodes I can even state with a high level of probability that I know what caused it. In Sydney, NS while on my honey moon it was definitely the Digby Clams...the one that tasted slightly off, but not so bad you spit it out. In France, it was the warmed milk I put in my coffee at the lunch I had in a quaint little Inn near the Millau Viaduct in Southern France.
Why you ask can I say with certainty that the Digby clams or the warmed milk were the culprits? By deduction my dear Watson. In both cases, my husband and I had eaten the exact same things all day. With the clam I distinctly recall thinking, hmmmm perhaps I should not have eaten that and with the milk...well my husband takes his coffee black.
According to a report publishes last week in the MMWR (Health Department Use of Social Media to Identify Foodborne Illness — Chicago,Illinois, 2013–2014) an estimated 55 million to 105 million people in the US experience acute gastroenteritis caused by foodborne illness each year. In Canada, 1 in every 8 people get sick due to foodborne illnesses. The unfortunate truth is that most people (me included) do not seek treatment which results in underreporting so the statistics are likely far higher. The Chicago Department of Public Health (CDPH) and its civic partners launched FoodBorne Chicago and used Twitter Tweets to try to track potential foodborne illnesses.
Using a Twitterbot (an algorithm) plus a new online complaint form. It searched for tweets geo-located to Chicago and its surrounding suburbs that mentioned "food poisoning." Human staff then read the tweets and marked them as relevant or not relevant, to help improve upon the algorithm to learn what tweets to pull in the future. Staff members then responded to the relevant tweets themselves. For example, the Twitterbot would locate a tweet such as "Guess who's got food poisoning? This girl!" and the staff members would reply back with "That doesn't sound good. Help us prevent this and report where you ate here (link to Foodborne Chicago and a web form to report the illness)." Over 10 months, the Twitterbot helped the department identify 133 restaurants for inspections and accounted for 6.9% of the 1,941 health inspections of food establishments prompted by complaints during the study period. For the 133 restaurants identified through the use of the Twitterbot, 21 failed inspection and 33 passed with "critical or serious" violations. Overall, Foodborne Chicago complaints contributed to 4% of the restaurants the city shut down for violations during the study period. Most importantly, the city likely would have never caught the majority of those complaints without the Twitterbot.
The use of Twitter has been widely discussed for tracking Pandemics and other outbreaks and certainly with this example shows that will a little work in developing the Twitterbot can help to identify foodborne illness. I wonder when we will see Twitterbots start investigating topics such as Hand Hygiene and cleaning and disinfection? Twitter just may be the tool to identify who is or who's not washing their hands and where the dirtiest surfaces or unclean conditions within hospitals and long-term care facilities are. Talk about peer pressure on steroids!