Monday, May 07, 2007

"STOP! Check the patient's wristband."

There is an excellent paper involving six countries by the BEST Collaborative in the May 2007 issue of Transfusion:
The paper has much to recommend it:

1. The research concerns transfusion to the wrong patient, which is the most important serious avoidable hazard of transfusion. It's always nice to read research that tackles transfusion issues that are both serious and common, i.e., ones that are clearly and directly significant to transfusion practice and patient safety. And after complaining about the content of Transfusion and its relevance to practice for technologists and nurses in an earlier blog (Whither immunohematology in AABB's Transfusion? ), the May issue is loaded:

2. The paper reports on a simple low technology intervention for reducing patient identification errors, namely a sticker tag that visually reminds transfusionists to stop and check the patient's wristband and requires removal to spike the unit.

  • The sticker reads, "STOP: Check the patient's wristband."

Low tech processes that work are bound to have wider applicability in smaller rural transfusion services and in the developing world.

3. An accompanying editorial (Kaplan H. Safer design.Transfusion 2007 May; 47(5): 758-9) discusses the BEST low tech approach and a high tech approach using radio-frequency microchips, examples of which are in TraQ's technology clearinghouse.

4. The BEST paper can be used to teach how to analyse scientific papers. Besides having an interesting design (multicenter cluster-randomized controlled trial involving short-term and long-term follow-up), the authors discuss multiple weaknesses and strengths of the research and basic statistical concepts such as sample size and statistical power.

5. The paper is a rare example of a negative study that gets published - the stickers had no overall effect on improving compliance with the bedside wristband check. Indeed, the results were slightly worse in the intervention group at the late re-audit stage, 8 weeks after introducing the sticker tag.

The authors speculate that this may have been a chance finding or that the constant reminder may have irritated the nurses and/or added to the complexity, producing the opposite effect to the one intended.

One can speculate that, if one part of a process is stressed, other critical parts may be inadvertently de-emphasized and forgotten, such as forgetting to breath when learning a new exercise for the first time.

Interestingly, the Kaplan editorial notes that procedural strategies are perhaps the least reliable for managing risks, yet are the ones often employed in transfusion and nursing practice where redundancy is typically used to increase reliability, e.g., having a second person check the work of another as done with the 2-person nursing check of patient and donor identification at the bedside prior to transfusion; or the 2-person check done when issuing blood from the transfusion service.

It's the old dictum about catching an error on the first inspection. If you have just seen your colleague perform several checks, you may not be as rigorous when confirming their work.

This study reminds me of other studies that have shown that educational interventions sometimes have no effect on changing behavior. Such results seem counterintuitive because we all want to believe that education will produce a positive effect. The trick is in discovering the right intervention or combination of interventions to motivate the target audience to change.

It's hard to comprehend how a sticker-tag saying "STOP: Check the patient's wristband." could have the reverse effect on a heath provider's performance. Investigating why this occurred could have relevance for similar studies.

If you would care to speculate or discuss further, please leave a comment.