About 1 in 38 hospitalised patients in the United States had at least one healthcare-associated infection in 2023, down from 1 in 31 in 2015, according to new survey data from the Centers for Disease Control and Prevention. The findings were published in the New England Journal of Medicine.
What the survey measured
The 2023 survey came from the CDC's Emerging Infections Program and covered 13,653 patients at 218 hospitals in 10 states. Trained staff reviewed medical records and applied national surveillance definitions to decide whether an infection counted.
CDC said patients in the 2023 survey were less likely to have an infection than those in 2015, even after accounting for differences in the patients included in the two surveys. A separate comparison put the decline at about 27% between 2015 and 2023.
"When federal agencies collaborate with specific goals and accountability, including use of quality measures, patients benefit from better care in clinical settings," said acting CDC director Jay Bhattacharya. "More needs to be done but this effort is notable in its progress on a very challenging issue."
Where the remaining risk sits
The absolute numbers are still large. CDC estimates that about 518,000 healthcare-associated infections occurred in US hospitals in 2023, costing billions of dollars and contributing to deaths.
Pneumonia, surgical site infections and gastrointestinal infections were the most common categories. One detail stands out: 61% of all such infections were not associated with a medical device or a procedure. That complicates the usual prevention playbook, which leans on catheter and ventilator protocols because those are easier to target.
Reporting and accountability
Hospitals report major infection types through the National Healthcare Safety Network, which CDC runs with the Centers for Medicare and Medicaid Services. The system combines reporting requirements, feedback to facilities, benchmarking and public transparency.
CMS administrator Mehmet Oz said the drop shows accountability and transparency are driving results, and that preventable harm must keep falling. Critics of the reporting system argue that measurement changes can flatter the trend, which is why the CDC compared surveys while adjusting for patient mix. The gap between 1 in 31 and 1 in 38 is real progress, but it leaves roughly 38 patients in every 1,000 leaving hospital with an infection they did not arrive with.