CDC data show decline in some hospital-acquired infections


Key clinical point: Selected strategies for preventing hospital-acquired infections appear to be working, based on a CDC comparison of 2015 and 2011 data.

Major finding: In two point prevalence surveys conducted in the same hospitals, the rate of HAI was 22% lower in 2015 (P = .001), compared with 2011.

Data source: CDC national surveys of HAIs in 148 hospitals in two different years (2011 and 2015) were compared.

Disclosures: Dr. Magill reported no financial relationships relevant to this study.



– There was an encouraging 22% reduction in hospital-acquired infections (HAIs) after adjustment for clinical variables when 2015 and 2011 data from national Centers for Disease Control and Prevention hospital surveys were compared.

“The data suggest that national efforts toward preventing HAIs are succeeding,” reported Shelley S. Magill, MD, PhD, a medical epidemiologist in the Division of Healthcare Quality Promotion at the CDC who summarized the data at an annual scientific meeting on infectious diseases .

Central Venous Catheter (CVC) MrArifnajafov/CC-BY-3.0
The reductions in HAIs were not evenly distributed. Rather, the biggest reductions were seen in the rates of surgical site and urinary tract HAIs. For other HAI categories, there were mixed results, with some modest increases as well as decreases.

The comparative data were drawn from point prevalence surveys conducted in 2011 and 2015 as part of the CDC’s Emerging Infections Program. In this type of survey, the data are collected over 1 day, providing a snapshot in time among selected hospitals. The analysis presented by Dr. Magill was restricted to the 148 hospitals that participated in both the 2011 and 2015 surveys, although the 2015 survey included a total of 199 hospitals, of which other data analyses are planned.

Due to the change in incidence, the rank order of HAIs was different in 2015 relative to 2011. While surgical site infections (SSIs) represented the most frequent HAI in 2011, they fell to the third most frequent HAI in 2015; pneumonia and gastrointestinal (GI) infections assumed the first and second spots, respectively. The GI HAI infection category includes Clostridium difficile infection.

The incidence of SSI HAI among all hospitalized patients in the survey fell by 41% between 2011 and 2015 (from 1.00% to 0.59%; P = .001). The other big contributor to the overall reduction in HAIs was the fall in the incidence of urinary tract infections, which fell 36% (from 0.55% to 0.35%; P = .04). The decrease in pneumonia (from 0.97% to 0.89%) was not significant, nor was the even more modest reduction in bloodstream HAI (from 0.45% to 0.43%). There was a modest increase in GI/Clostridium difficile infections (from 0.56% to 0.59%).

The surveys do not permit the reduction in HAI rates to be attributed to any specific prevention practices, but Dr. Magill pointed out that the overall reductions correlate with reduced use of urinary catheters and central lines; reductions of both have been advocated as a means for improved infection control. Of several factors that might contribute to a reduction in SSI HAI, Dr. Magill speculated that better adherence to guidelines and more rigorous steps at preoperative infection control strategies might be among them.

Detailed analyses of the data collected from all of the hospitals that participated in the 2015 survey are planned, including an evaluation of which antibiotics were used to treat the HAIs found in this survey. Although the findings so far encourage speculation that infection control practices, such as prudent use of urinary catheters, are having a positive effect, Dr. Magill said that the data also point out the challenges.

“Given that pneumonia continues to represent a large proportion of HAIs in hospitals, more work is needed to identify risk factors; understand the factors that are preventable, particularly in the nonventilated patients; and develop better preventive approaches,” Dr. Magill said.

Dr. Magill reported no financial relationships relevant to this study.