The Oregonian today posted an article detailing how Oregon hospitals are cutting the rates of infections for certain types of conditions.
The rate of infections associated with central-line catheters last year fell to less than half the rate in 2009, when public reporting began. Bypass surgery site infections fell 10 percent, and knee replacement surgery site infections fell 20 percent.
Still, sizable differences persist between the best and the worst performing hospitals in the state’s third annual report on hospital infections, released today. In the Portland-area, for instance, three hospitals reported zero surgical site infections after colon surgery, while three others reported infection rates more than double the state average.
The Oregon Legislature required hospitals to begin publicly reporting infection rates in 2009. The goal is to push hospitals to work harder to prevent infections and give consumers objective information to compare hospitals.
Stopping infections associated with central-line catheters has been a high priority. They kill thousands of intensive care patients each year in the U.S. and cost hundreds of millions of dollars to treat. Many doctors long assumed the risk was inevitable, but studies since 2006 have shown that rigorous attention to safety steps can practically eliminate central-line infections.
In Oregon last year, 65 percent of hospitals reported zero central-line infections, up from 52 percent in 2009. Oregon’s statewide rate of central-line infections dropped to less than a third the national rate: from 1.39 to 0.62 per 1,000 days patients spent with central lines.
The improvement may have prevented 21 deaths since 2009, Oregon health officials estimate, given that as many as one-in-four central-line infections are fatal. Prevention also saves money, about $7,300 to $29,000 for each central-line infection avoided.
Hospitals last year began reporting five more types of infection. Oregon’s reported rates are 44 percent lower than the national average for central-line infections among newborns in intensive care, 21 percent lower for hysterectomy surgery site infections, 20 percent lower for colon surgery site infections, and 31 percent lower for spine surgery site infections.
Hip replacement surgery site infections are a different story. Oregon’s rate is 13 percent higher than the national average.
Using the report to comparison shop for hospitals remains problematic.
For example, a hospital may look worse only because it happens to care for a larger share of severely ill patients vulnerable to infections.
“The best way to evaluate a hospital is to look at their measures over time,” said Jeanne Negley with the state Office for Oregon Health Policy and Research.
The state relies on hospitals to self-report infections, adding more uncertainty. Higher reported rates may reflect more accurate identification work by a given hospital, rather than more infections.
Even experienced infection control experts have trouble deciding when an infection has been caused by a central-line catheter. In an audit of hospital records by the Oregon Public Health Division, hospitals and state reviewers initially disagreed on the status of 35 possible hospital-acquired central-line infections. In the end, the hospital proved correct in 13 cases and reviewers correct in 22 cases.
Overall, the audit found that hospitals accurately reported about 80 percent of the confirmed central-line infections in 2009. Dr. Paul Cieslak, manager of the health division’s communicable disease section, and colleagues found 86 cases that met the criteria and should have been reported. Hospitals missed 16 confirmed infections. They reported 76 cases, 6 of which didn’t meet the criteria.
Audits in other states have found more sizable underreporting.
Connecticut officials found that hospitals reported less than half of central-line infections meeting criteria set by the U.S. Centers for Disease Control and Prevention. Tennessee’s Department of Health found that hospitals reported 46 of the 59 central-line associated blood stream infections they identified by reviewing medical records.
- Joe Rojas-Burke