Teams of hospitals control infection better

U. PITTSBURGH (US) — When hospitals work together, they do a better job of controlling deadly and super-contagious bacterial infections like MRSA, a new study shows.

An individual hospital’s infection control efforts have a ripple effect on the prevalence of a deadly and highly infectious bacterium in hospitals throughout its surrounding region, as a multi-center research group led by the University of Pittsburgh demonstrates in a computer simulation-based study.

Hospitals working alone to control Methicillin-resistant Staphylococcus aureus, or MRSA, don’t achieve the same level of infection control, according to the results of the study published in the October issue of the journal Health Affairs.


“Unless they are associated financially or legally, hospitals often have their own separate infection control programs and procedures,” says Bruce Lee, director of Pitt’s Public Health and Infectious Diseases Computational and Operations Research (PHICOR) group.

“However, hospitals are rarely isolated islands and instead share patients extensively with other hospitals in their area, which can facilitate the spread of MRSA infections.”

Lee and his colleagues obtained real-world data from all 29 hospitals in Orange County, California, and developed a computer simulation to assess a procedure called “contact isolation” to limit MRSA transmission.

The procedure involves testing all patients for MRSA upon admission to a hospital. When a patient tests positive, hospital staff must wear gloves and gowns whenever entering that patient’s room or interacting with that person.

The simulation explored scenarios in which different combinations of hospitals implemented contact isolation to varying degrees.

“The more that hospitals work together and coordinate infection control efforts, the more they all benefit,” says Lee, also an associate professor of medicine, epidemiology, and biomedical informatics at Pitt’s School of Medicine and Graduate School of Public Health.

“For example, doubling the number of hospitals that adopt contact isolation can more than double their improvement in infection control.”

The simulation also found that requiring contact isolation at one hospital not only decreased MRSA at that hospital as expected, but also did so in other nearby hospitals that had not implemented the intervention.

When the researchers ran the simulation with all the hospitals in Orange County implementing contact isolation simultaneously with a 75 percent compliance rate, MRSA prevalence decreased an additional 3.85 percent over what the hospitals could have achieved on their own. Long-term acute care facilities fared even better, with a 12.13 percent additional decrease.

The simulation relies extensively on data regarding infection rates, annual admissions, average patient length-of-stay, hospital transfers, readmissions, number of intensive care units, and bed capacity at the Orange County hospitals.

MRSA is widely prevalent in US hospitals. In 2006, the MRSA colonization rate, or detection of the bacteria on a patient’s skin or soft tissue, was 12 per 1,000 inpatients. In 2010, the rate nearly quadrupled to 41 per 1,000 inpatients despite an overall decrease in the rates of MRSA infection.

The researchers had shown previously that hospitals in Orange County, which has a population of 3 million, are highly interconnected through patient transfers and readmission of patients to different hospitals after an intervening stay at home or elsewhere.

Some hospitals have started cooperating to control hospital-acquired infections. In California, the Safety Net Initiative is building a learning collaborative among California public hospitals to reduce such infections. The Pittsburgh Regional Health Initiative has successfully created a culture of change to improve overall patient safety. Similar programs in Iowa, Michigan, Nebraska, New York, South Dakota, and Wisconsin also have successfully reduced hospital-acquired infections.

This study was supported by the National Institute of General Medical Sciences Models of Infectious Disease Agent Study (MIDAS) grants, National Institute of Allergy and Infectious Diseases, and the Pennsylvania Department of Health.

Co-authors contributed from University of Pittsburgh, Pittsburgh Supercomputing Center, Harvard University Medical School, and the University of California, Irvine.

Source: University of Pittsburgh