Between June and December of last year, 17 patients at the Clinical Center of the U.S. National Institutes of Health in Bethesda, Maryland, came down with a bloodstream infection; six of them died. Doctors knew that a patient had arrived with Klebsiella pneumoniae in June, but it wasn’t clear how the bacterium, a common culprit in hospital-acquired infections, was passed around, or whether several different patients had simply brought it in with them.

By taking bacterial samples from the patients and certain hospital equipment and analyzing the genomes of the different strains, researchers traced the organism’s meandering path as it cut a deadly swath through the facility. The study, published today in Science Translational Medicine, represents the first such use of whole-genome scanning in a hospital during an outbreak.

The sequencing was possible because it is now cheap and fast enough to do in near-real time, says Julie Segre, an author of the paper and a senior investigator at the National Human Genome Research Institute. Each of the genome sequences cost about $2,000 last year; the price has since dropped to about $500.

The scans revealed that the germs originated from one patient, that their path through the hospital was different than expected, and that current methods for combating such hospital-acquired infections are inadequate. Instead of jumping from the first patient who showed signs of the blood infection to the second, as the epidemiologists had expected, the bacterium had taken a much more complex route. Patient 1, a 43-year-old New Yorker, passed it independently to patients 3, 4, and 8, whose cases appeared more than a month later. Patient 8 didn’t pass it further, but the strain from patient 3 was found in patients 5 and 2, and the rest of the patients got it from patient 4, the analysis showed.

Researchers believe the germ was transmitted on hospital workers’ hands. Moreover, scans of bacteria from one patient’s ventilator showed that the typical cleaning process was inadequate to rid the machine of the bacteria. Although the germs were not transmitted via the ventilator in this case, they may very well be transmitted that way elsewhere, says Tara Palmore, an infectious disease physician and deputy hospital epidemiologist at the Clinical Center, who was an author of the study.

George Weinstock, associate director of the Genome Institute at Washington University in St. Louis and a professor of genetics and microbiology there, says that the extra level of detail provided by whole-genome sequencing allowed epidemiologists to truly understand the path of disease for the first time. The lessons learned from such detailed tracking, he says, “are going to lead to more sophisticated epidemiological studies in public health, so that we’re more often going to get the true answer to what’s going on, rather than the simplest.” The next step, he says, will be to do this kind of analysis in real time, so that far fewer people will die before an outbreak is stemmed.

The outbreak highlights the strong need for improved infection control protocols. “We have very few antibiotics left,” Segre says. “The only way to stop this is with better infection control. The best way to keep the American people healthy is to keep them from getting sick.”