Sunday, March 16, 2008

When it comes to Hospital Infection, Size Matters (sort of...)


The latest issue of Annals of Surgery plopped into my letterbox recently. Its cornflake-colored cover glowed invitingly on my desk for a good portion of the morning, successfully ignored. By mid afternoon, however, the intellectual temptress had successfully worn me down.

Grazing the Table of Contents, I came across a study from Anderson and colleagues at Duke in North Carolina. They reported a worrying vista -- later this year, Medicaid and Medicare in the US will cease reimbursement for what they deem additional care provided as a result of surgical site infections. This move, I predict, will add fuel to the bush-fire of public and political hysteria sweeping westward from Europe on the issue of hospital-acquired infection -- an oft-overstated, oversimplified "MRSA plague". I know that surgeon-rapporteur Atul Gawande has touched on the the issue of nosocomial infection (infection as a result of hospitalisation) in his latest book "Better", but I was sufficiently intrigued by the article as to attempt to flog this mortally-wounded horse.

There is some evidence in the cancer literature that "size matters;" namely, that the larger the institution and the greater the caseload, the better the outcomes. Analogous to this, Anderson et al. hypothesized that the larger the center, the lower the rate of surgical site infection. They undertook to examine this by polling a prospectively-accrued database of some 132,000 operations in large and small hospitals over a two-year period.


Their analysis of the resultant 1,434 cases of wound infection turned out less obvious (and arguably more interesting) results. Intuitively, low-volume centers (less than 1500 operations per year) per formed worse than the medium volumed hospitals. However, this trend did not translate to the "super sized" hospitals. Even after adjusting for surgeons experience, the number of high risk procedures and the presence of a training program for residents, the prevalence of surgical site infections for large hospitals (more than 4000 cases/year) was HIGHER than that in medium centers.

In the Irish context, where tertiary-referral centers are routinely forced to operate at-or-above bed capacity, this is intuitively true, but I was amazed this might be the case Stateside, where the wards, we in Europe fantasize, are paved with gold. I can only rationalise that there is a critical size where the budget, staff-levels and infrastructure are large enough to have the required expertise but not so large as to be an impediment to hygiene efforts.

Given the enthusiasm and political mileage attached to public "league tables" of hospitals' infection rates, and the looming reality of vanishing reimbursements, the "super hospitals" are going to have a long hard look at what this might mean going forward.