By Stan Deresinski, MD, FACP, FIDSA

Clinical Professor of Medicine, Stanford University

Dr. Deresinski reports no financial relationships relevant to this field of study.

SYNOPSIS: One-day prevalence studies demonstrated that there has been a 16% reduction in the risk of healthcare-associated infections from 2011 to 2015.

SOURCE: Magill SS, O’Leary E, Janelle SJ, et al. Changes in prevalence of health care-associated infections in U.S. hospitals. N Engl J Med 2018;379:1732-1744.

Using the Emerging Infections Program, researchers at 10 participating sites recruited as many as 25 hospitals (general, women’s, and children’s) in their areas to participate in a one-day prevalence of healthcare-associated infections (HCAI). Each hospital selected a study date from May 1, 2015, through Sept. 30, 2015. The survey included 12,299 patients in 199 hospitals, and the results were compared to a 2011 survey involving a similar number of hospitals and patients. The patients in the two surveys were similar regarding the proportion in critical care units (15%), the median interval from hospital admission to the survey (three days), and the proportion with an HCAI who died (approximately 11%). However, in 2015, the percentages with a urinary catheter were lower (18.7% vs. 23.6% in 2011), as were those with a central venous catheter (16.9% vs. 18.8%).

In 2011, 452 of 11,282 (4.0%; 95% confidence interval [CI], 3.7-4.4) patients had one or more HCAI. In 2015, using the same definition, this was true of only 394 of 12,299 (3.2%; 95% CI, 2.9-3.5), a difference that was significant (P < 0.001). Pneumonia was the most frequently identified infection, followed by gastrointestinal (mostly due to Clostridioides difficile) and surgical site infections. Just more than three-fourths of the latter were deep incisional or organ-space infections.

The most frequently isolated pathogens were C. difficile, Staphylococcus aureus, and Escherichia coli. Of the 47 isolates of S. aureus for which susceptibility test results were available, 21 (45%) were methicillin-resistant. Carbapenem resistance was detected in only three of 66 (5%) isolates of E. coli, Klebsiella spp., and Enterobacter spp.

After multiple adjustments, it was determined that the risk of acquiring an HCAI was 16% lower in 2015 than it had been in 2011 (risk ratio, 0.84; 95% CI, 0.74-0.95; P = 0.005). Without adjustment for the presence of a ventilator, central venous catheter, or urinary catheter, there was a 24% reduction in HCAI during the interval.


This study demonstrates that national activities, locally applied, have had great success in reducing the prevalence of HCAI in the United States. The investigators estimated that in 2015 there were 633,300 (95% CI, 216,000 to 1,912,700) patients with an HCAI, a number that translates into a reduced prevalence since 2011.The largest improvement was seen in surgical site and urinary tract infections. The authors suggested that the former may be related to improved surgical antibiotic practices and, possibly, MRSA decolonization procedures. The decreased use of bladder catheters may have been responsible, at least in part, for a corresponding decrease in urinary tract infections despite adjustments in the multivariate analysis. In contrast, there was no evidence of a decrease in the prevalence of healthcare-associated pneumonia cases, most of which were not associated with mechanical ventilation, or of C. difficile infection. The latter presents some difficult issues in interpretation because of the introduction of PCR testing for the presence of the toxinB gene, which, if used alone, may overestimate the frequency of infection by this organism, possibly by a factor of 2. Although the diagnostic methods were not examined, such confounding may have masked an actual decrease in C. difficile infection.

Overall, these results point to the remarkable progress that has been made in reducing HCAI in the United States, but they also demonstrate that there is a long way to go.