Group items tagged

This is the largest nationally representative study to date of the toll taken by sepsis and pneumonia, two conditions often caused by deadly microbes, including the antibiotic-resistant bacteria MRSA. Such infections can lead to longer hospital stays, serious complications and even death. "In many cases, these conditions could have been avoided with better infection control in hospitals," said Ramanan Laxminarayan, Ph.D., principal investigator for Extending the Cure, a project examining antibiotic resistance based at the Washington, D.C. think-tank Resources for the Future. "Infections that are acquired during the course of a hospital stay cost the United States a staggering amount in terms of lives lost and health care costs," he said. "Hospitals and other health care providers must act now to protect patients from this growing menace." Laxminarayan and his colleagues analyzed 69 million discharge records from hospitals in 40 states and identified two conditions caused by health care-associated infections: sepsis, a potentially lethal systemic response to infection and pneumonia, an infection of the lungs and respiratory tract. The researchers looked at infections that developed after hospitalization. They zeroed in on infections that are often preventable, like a serious bloodstream infection that occurs because of a lapse in sterile technique during surgery, and discovered that the cost of such infections can be quite high: For example, people who developed sepsis after surgery stayed in the hospital 11 days longer and the infections cost an extra $33,000 to treat per person. Even worse, the team found that nearly 20 percent of people who developed sepsis after surgery died as a result of the infection. "That's the tragedy of such cases," said Anup Malani, a study co-author, investigator at Extending the Cure, and professor at the University of Chicago. "In some cases, relatively healthy people check into the hospital for routine surgery. They develop sepsis because of a lapse in infection control—and they can die." The team also looked at pneumonia, an infection that can set in if a disease-causing microbe gets into the lungs—in some cases when a dirty ventilator tube is used. They found that people who developed pneumonia after surgery, which is also thought to be preventable, stayed in the hospital an extra 14 days. Such cases cost an extra $46,000 per person to treat. In 11 percent of the cases, the patient died as a result of the pneumonia infection.

Statistical validation of results, as Shaffer described it, simply involves testing the null hypothesis: that the pattern you detect in your data occurs at random. If you can reject the null hypothesis—and science and medicine have settled on rejecting it when there's only a five percent or less chance that it occurred at random—then you accept that your actual finding is significant. The problem now is that we're rapidly expanding our ability to do tests. Various speakers pointed to data sources as diverse as gene expression chips and the Sloan Digital Sky Survey, which provide tens of thousands of individual data points to analyze. At the same time, the growth of computing power has meant that we can ask many questions of these large data sets at once, and each one of these tests increases the prospects than an error will occur in a study; as Shaffer put it, "every decision increases your error prospects." She pointed out that dividing data into subgroups, which can often identify susceptible subpopulations, is also a decision, and increases the chances of a spurious error. Smaller populations are also more prone to random associations. In the end, Young noted, by the time you reach 61 tests, there's a 95 percent chance that you'll get a significant result at random. And, let's face it—researchers want to see a significant result, so there's a strong, unintentional bias towards trying different tests until something pops out. Young went on to describe a study, published in JAMA, that was a multiple testing train wreck: exposures to 275 chemicals were considered, 32 health outcomes were tracked, and 10 demographic variables were used as controls. That was about 8,800 different tests, and as many as 9 million ways of looking at the data once the demographics were considered.