Archive for September 2009

Physician Accountability for Violation of Safety Rules: The Time For Excuses Has Passed

In this week’s New England Journal, Peter Pronovost and I make the case for striking a new balance between “no blame” and accountability. Come on folks, it’s time.At most hospitals, hand hygiene rates hover between 30-70%, and it’s a near-miracle when they top 80%. When I ask people how they’re working to improve their rates, the invariable answer is “we’re trying to fix the system.” Now, don’t get me wrong. I believe that our focus on dysfunctional systems is responsible for much of our progress in safety and quality over the past decade. We now understand that most errors are committed by good, well-intentioned caregivers, and that shaming, suing, or shooting them can’t fix the fallibility of the human condition.But not washing hands? When I hear, “It’s a systems problem,” my BS detector goes a little bit haywire, particularly after I walk around the hospital and see alcohol gel dispensers…

Selective outcome reporting in clinical trials

Researchers analyzed all randomized controlled trials listed in pubmed in 2008 in 3 clinical area (GI, cardiology, and rheumatology) in 10 high impact clinical journals. They found that of the 323 trials, only 46% were adequately registered (registered before the end of the trial with the primary outcome cleared stated). Among those adequately registered, 31% had a discrepancy in the registered primary outcome and the published primary outcome (including omission of the primary outcome, reporting a secondary outcome as primary, and reporting unregistered outcomes as primary). Despite registration requirements for most high impact journals, many trials are not publishing the intended primary outcomes measures (abstract)

Dialysis patients and recurrent GI bleeding

In this case-control study of dialysis and non-dialysis patients with UGI bleeding, all had endoscopic control of their bleeding, and received 3 days of IV PPI (40mg IV BID), then oral PPI (20mg qday) for 2 months. The rate of re-bleeding at 7 days was about the same in the 2 groups, but between 7-30 days, the dialysis group had significantly higher rates of re-bleeding (10% vs 0%). Higher PPI doses in dialysis patients at discharge may be necessary to reduce the intermediate risk of re-bleeding (abstract).

Predicting mortality and intubation risk in COPD flares

In this large retrospective cohort of 88,074 patients admitted with a COPD flare, researchers determined 3 variables (BUN>25, HR>109, and acute mental status change) predicted risk of mortality and intubation. In those with all 3 factors, mortality was 13%-15% (in the derivation and validation cohorts). In those with none of those factors and <age 65, mortality was 0.3% in both cohorts. The AUROC for mortality and intubation ranged from 0.71 to 0.77. This simple risk score can predict mortality and intubation risk in patients admitted with COPD flares (abstract).

An Overlooked but Dangerous Handoff… of One Million Patients at a Time

A quick heads up on an article written by a very talented UCSF psychiatrist named John Young, which I had the opportunity to co-author. John observed that, despite all the recent literature about handoffs (such as here and here), no one has given much thought to the Mother of all Handoffs: the transition of outpatient panels from graduating residents to brand new interns that happens around July 1st every year.In the article, in this week’s JAMA, we point out that the “year-end handoff” carries a number of additional risks above and beyond the usual threat of fumbles. For example,Most handoffs involve two providers of comparable knowledge and experience. The year-end handoff involves a handoff from a seasoned expert (the graduating resident) to a novice (the brand new intern, who was a med student a week earlier);The volume is huge: for an individual resident, the handoff may involve as many as…