ResistanceMap and Get Smart About Antibiotics Week

Among the many skills admired by journalists (always finding the free food; never going out of the office; yes, those are said in jest), skidding in* just *under deadline may be the most valued. I am making a flourish of that skill here, sliding in a mention of an important observance this week just as the week ends.

(Sorry. There was a lot of news this week.)

So: In case you hadn't yet noticed, this is (was) Get Smart About Antibiotics Week, 2010.

The Get Smart week is co-sponsored by the CDC and the FDA, and its goal is to alert people to the continuing overuse of antibiotics in human medicine. (A separate but just as important issue as overuse of antibiotics on the farm, which I've talked about a number of times here.)

If you're concerned about this — and who shouldn't be — the CDC's sites have a plethora of information about this crucial issue. There are briefings, posters, brochures, radio and TV PSAs, a blog by federal researchers and others and — possibly the most directly useful to parents in the whole campaign — a form letter that doctors can download and fill out to give to working parents, explaining why a child with a sniffle or a sore ear can be admitted to daycare without an antibiotic prescription.

For the first time this year, the Get Smart week included a focus on antibiotic use in health care that tackled why antibiotic stewardship, as it is sometimes called, is such a crucial issue.

But my favorite feature of the week is ResistanceMap: an elegant and brutally clear set of time-maps engineered by the researchers at Extending the Cure, whose work I've written about before. The maps document the emergence of four resistant organisms — MRSA, imipenem-resistant Acinetobacter, cipro-resistant E. coli, and TMP-SMX resistant E. coli — in the US over the decade 2000-2009. (The 2009 image from the MRSA map is above right.)

EtC says about the data behind the display:

These series of maps explore the spatial distribution of regional resistance levels for four common combinations of antimicrobials and organisms... The maps cover the 2000-2009 period and are organized by the nine US census divisions. For a list of states in each census sub-region, please see www.census.gov/geo/www/us_regdiv.pdf.

At the core of the data are individual bacterial isolates tested for resistance to certain antimicrobial agents. Isolates are collected electronically from participating clinical microbiology laboratories. Each laboratory reports on isolates gathered from one or more hospitals. The participating institutions are nationally representative and come from more than 150 zip codes across the country. The number of hospital beds represented by the data is set to roughly match the hospital beds in each state.

The numbers used to make the maps are restricted to outpatient samples. State-level data is aggregated at the regional level. For each drug-bug combination, regional resistance rates are calculated as the ratio of strictly resistant isolates to the total number of collected isolates in a given region. The total sample size (N) used to make the maps is 5,587,728 isolates.

A limitation of the data is that laboratories may join or leave the database over time. Generally, no more than 10% of the sample will change over the course of a year. However, the number of participating sites may vary more across years, especially for the first and last year of the period. As a result, in some of the maps, certain states are not represented in all years. The aggregation by region is necessary to overcome this problem and cover the entire country.

Even if you think you know something about antibiotic resistance, watching these maps unfold is a revelation: As the percentage of organisms that are resistant increases, state after state flushes as red as an inflammation. It's chilling. It's a reminder that antibiotic overuse ought to be a priority not just this week, but every week.