As a resident of the corporate world (and as I have to explain to my children every year when their elementary school lets out for summer), not a whole lot changes about my job over the summer. My daily routine changes a little since I don’t have to see the kids off to school, coworkers might be gone for a week or two at a time for vacation, and there’s usually a company picnic or kayak trip. But by and large July is no different from October or February, with one notable exception: government fiscal years.
My company’s customers are primarily state and local health departments, which means most of our income is tied to government budgets and grants. Those budgets and grants tend to run on fiscal years, and those fiscal years all tend to be similar. As a result, most of our revenue comes in over a relatively short period of time towards the late summer. And every year, there’s always some uncertainty about which customers will have funding, which states will have actually passed budgets, and whether last year’s cash will hold out until this year’s starts arriving. So far it has always worked out, and this year has been better than most, but the dance goes on. I suspect many faculty experience similar cycles with their grants.
Just because the corporate calendar doesn’t have semesters and summer breaks doesn’t mean it’s always business as usual; new challenges pop up all the time. Our most recent opportunity arose around the political conventions. Whenever you get a lot of people together from different parts of the country and put them all in the same space, there are concerns about the spread of diseases, not to mention violence or unforeseen health emergencies. Our public health customers tend to go in high alert for these big events, and this year both conventions were occurring in states whose health departments we serve.
Those health departments asked us to help them keep on top of the health condition in the convention host cities. Our usual monitoring of emergency departments is already intended to detect upticks in infectious disease activity, but at times like this the health departments want to watch everything more closely. One challenge they have is anticipating unanticipated circumstances. They don’t just want to detect the events they can imagine–food poisoning, heatstroke, etc.–but also the ones they haven’t. Our usual methodology is to look for predefined sets of words in what the patients report about their condition, words like ‘cough’ or ‘fever’ or ‘nausea.’ But for events like these conventions, we’ve also started looking at all the words and developed techniques that answer the questions “What words are people using to describe their illness or injury today that they’ve never used before?” or “What words are people using today in large numbers that they rarely used in the past?” We answer those questions in pretty simplistic ways, but the results are still helpful in making sure we are prepared for as many scenarios as possible. And even as we try to detect the unexpected, we find answers to questions we didn’t think to ask, like whether we can detect when a hospital has hired someone new to register patients, or when the registrar shift changes during the day.
Of course, the twist in working in public health is that you prepare to detect and respond to all sorts of known and unknown scenarios, but hope that none of them actually happen. The best possible outcome for all the work that we did in preparing to monitor these conventions is that nothing actually happened and we are certain of that because if something had happened, we would have seen it. Hopefully as you are reading this, the only newsworthy developments from the conventions are the political ones. (And if you want to know what to expect from those, you’ll need to talk to Joshua Wu.)