Pennsylvania has issued hospital-specific data on infections among 1.6 million patients treated at 168 facilities statewide in 2005. The report, released by the Pennsylvania Health Care Cost Containment Council, marks the first time any state has issued data on individual hospitals, and is expected to establish a baseline for future performance and quality improvement. It also overturns a lot of conventional wisdom about how infections occur.
“It's a breakdown in processes that creates infections,” said Marc P. Volavka, executive director of the council, in an interview.
The detailed report, available at the council's Web site (www.phc4.org
Of 1.6 million patients treated at the 168 facilities, 19,154 had a hospital-acquired infection, for a rate of 12 per 1,000 cases. The infections accounted for 394,129 hospital days and $3.5 billion in charges.
The average length of stay was 20.6 days for those with an infection and 4.5 days for those without. Charges were higher for those with infections than for those without, averaging $185,260 and $31,389, respectively. Similarly, mortality was 13% and 2%, respectively.
Most cases were covered by Medicare or Medicaid. Only 276,523 of the patients had commercial insurance; among them, 1,522 acquired an infection in the hospital. Private payers covered only about $53,000 of an infection-related stay, but the total payout was $82 million.
Even though all hospitals are reporting, it is likely that the data hugely underestimate what actually occurs, said Mr. Volavka. He noted that the council has not asked hospitals to track infections subsequent to discharge, which may be when most surgical site infections develop.
The data collection began in 2004, when hospitals were required to report on surgical site infections for circulatory, neurologic, and orthopedic procedures; indwelling catheter-associated urinary tract infections; ventilator-associated pneumonia; and central-line-associated bloodstream infections. In the third and fourth quarters of 2005, hospitals had to expand reporting to include all surgical site infections. In the fourth quarter of 2005, pneumonia, bloodstream, and urinary tract infections not related to devices were added.
Urinary tract infections (UTIs) were the most common, affecting 11,265 patients, for an infection rate of 7.2 per 1,000. Those infections were particularly common in heart failure patients, followed by those admitted for other cardiac conditions.
Surgical site infections had the second-highest incidence rate, at 5.2 per 1,000, affecting 1,615 patients. Intestinal surgery accounted for the highest percentage of surgical site infections (9%), closely followed by angioplasty and surgery for osteoarthritis and leg fractures.
These surgical infections accounted for most of the infections in each age group, except for those patients older than 60 years, in whom UTIs were most common.
Aside from UTIs, the number of infections actually declines as patients age, a fact that runs counter to prevailing theories about older patients' being more vulnerable to infection, Mr. Volavka said. He added that more UTIs occur in the over-60 group because it comprises a preponderance of people who age in state hospitals, where they are catheterized instead of helped to the bathroom.
“It's not because the patients are by definition more at risk. It's the behavior of the hospitals that puts them at risk,” he said.
Several recently published studies appear to support Mr. Volavka's assertions. Researchers at Allegheny General Hospital in Pittsburgh found that severity of illness did not predict central-line bloodstream infections, and that the most common primary diagnoses among those infected—acute myocardial infarction, heart failure, respiratory failure, and deep venous thrombosis—were not usually considered risk factors (Am. J. Med. Qual. 2006;21[suppl]:7S-16S). A group at a clinical research organization had similar findings, concluding that sicker patients were not necessarily at higher risk for infections (Am. J. Med. Qual. 2006;21[suppl]:17S-28S). Finally, a third study found that hospital practices—such as method of hair removal, hand-washing, and operating room traffic flow—played an important role in predicting which patients were at risk for surgical site infections (Am. J. Med. Qual. 2006;21[suppl]:29S-34S).
The studies “make it clear that it is the process of care, not the underlying clinical condition of the patient, that drives the current epidemic of hospital-acquired infection,” said Dr. David B. Nash, chairman of the department of health policy at Jefferson Medical College in Philadelphia.