Getting at the truth behind hospitals' published infection rates

CONSUMERS should be cautious in interpreting the recently published infection rates of many of Washington's hospitals.

These hospital-specific infection rates — including the rate of bloodstream infections caused by central venous catheters, often referred to as "central-line infections" — are published because of laws recently passed by Washington's Legislature. According to the state's hospital association, which supports this law and posted these data on its website, the public reporting of infection rates will assist "consumers in making good decisions about hospital care."

Consumers Union might agree. Its March issue of Consumer Reports listed the rates of central-line infections for hospitals in 10 states (not including Washington), urging consumers to use these hospital-specific data to "protect" themselves and (when given a choice) select a hospital it rated a "top performer."

But these published rates are not necessarily as reliable a tool to evaluate the safety of hospitals as the consumer might have hoped.

Close examination of the methods used to collect and report these performance data suggests that published infection rates, like Washington's, can misrepresent a hospital's safety, resulting in erroneous judgments about its quality. To be sure, a hospital reporting a relatively high infection rate may not be any less safe than a neighboring and competing hospital reporting a lower infection rate, but rather, in addition to some other considerations, may have a sicker patient population or use more-sensitive and less-subjective methods to detect and count its infections.

Indeed, the use of scientific and standardized techniques to measure and report infection rates is especially important, because, in addition to the use of these published rates by consumers, hospitals may be financially rewarded for, and reimbursements may be based on, reporting low infection rates.

Bringing the facts into focus, reports suggest that the public reporting of infection rates can unwittingly introduce a measure-ment bias, if not also pose a financial conflict of interest, that causes the published rates to underreport the true incidence of infection (as well as to over-exaggerate the effectiveness and cost-savings of a set of infection-control practices under evaluation).

Indeed, studies have identified significant discrepancies between the rate of central-line infections reported by the hospital and the rate determined during an independent review of the hospital's infection data.

These studies would suggest that the process of publishing hospital-specific infection rates might be more myopic, clever and geared toward the interests of the hospital than it is scientific, discerning and focused on the protection of the consumer. No doubt transparency can be an effective disinfectant, and Washington's mandated public reporting of hospital-specific infection rates is a most important step toward improved accountability and health care. But this admirable effort may fall unnecessarily short of its stated goal because of one most salient oversight.

Though admittedly less like a fox guarding a hen house than a school system that permits children to grade themselves before gleefully handing their report cards to their parents, Washington's published infection rates, like those in the majority of the states, are unaudited and reported by the hospitals themselves. And, having not been independently checked for accuracy (for example, by state health officials or a computer program), these infection data may be in error, incomplete and of limited usefulness, which many consumers may not realize.

Washington legislators are therefore respectfully requested to consider updating the state's laws to require that the accuracy of its hospital-specific infection rates be independently validated before their publication, like a public accountant might certify an individual's federal tax return before its filing.

Otherwise, the significance, usefulness and purpose of Washington's published infection rates may be questioned.

Lawrence F. Muscarella, director of R&D for Custom Ultrasonics Inc. and founder of an infection-control website (www.MyEndoSite.com) and of a newsletter ("The Q-Net Monthly").