Every flu season doctors note that many people mistake having a bad cold for the flu, but that their symptoms, even if flu- like, do not mean they have the flu. So what do you think happened when the Centers for Disease Control called 10,000 people by phone and asked them if they had flu-like symptoms? Well, golly gosh, it turns out that one in five kids had “flu-like” symptoms in the past month (you know, the month when kids returned to school to give each other their germs). The CDC claims that most of the kids probably had swine flu, which, naturally, led some people in medialand — The Los Angeles Times, MSNBC and several others who appear to have toned down their headlines during the day– to claim that they actually had swine flu.
But – cough – how could anyone possibly make an accurate diagnosis of swine flu over the phone, when so many people commonly refer to “cold-like” symptoms as having the flu? We have privileged access to our thoughts and feelings, and can tell a telephone interviewer with 100 percent accuracy whether we are happy or sad; but we cannot, alas, discern between viruses when we’re sick; we just know we’re sick with something.
Aside from the limits of self-diagnosis, a telephone survey of this kind is also prey to other kinds of bias, such as whether people were more inclined to affirm that they had flu-like symptons (rather than claim a mere cold) due to the proliferation of swine flu stories in the media.
The false certainty created by ‘guessurveying’ the incidence of swine flu may well be designed to encourage parents to get their children vaccinated, but laudable ends are rarely well served by such obviously lame means. It just makes the CDC look unscientific. And that’s the kind of development that can metastasize into full-blown lack of credibility, when, eventually, some news organization starts to ask awkward questions.
And despite the prevalence of stenography in the press, the CDC’s “guessurvey” comes a day after an investigation by CBS News found that the Centers told states to stop doing real testing for H1N1 in July — and to stop counting actual cases, decisions which the network reported left some public health experts perplexed. On top of that, CBS’s analysis of state data shows that H1N1 was less prevalent over the summer than expected and that claims based on apparent symptoms and not actual testing — such as the alleged outbreak of swine flu at Georgetown University — were more likely to over-estimate the incidence of disease.