Sunday, January 10, 2010

Checklists - Simply the best? (Musings on mistake proofing)

This blog muses on using checklists for preventing mistakes in transfusion medicine (TM). The title comes from the inimitable Tina Turner's 1989 mega-hit, Simply the best.

The blog was motivated by an interview on Charlie Rose with Dr. Atul Gawande (see short promotional clip) about his new book:

The book (hereafter called "The Checklist") is getting much play in the media and is being hailed as brilliant.

What struck me was that the simple checklist and equivalent low tech mistake-proofing tools have been used in the TM laboratory for some time and more recently applied to physician blood ordering practices and nursing blood administration processes and procedures

In a way SOPs can be considered a type of checklist. They involve procedural steps that must be completed before proceeding to the next one. For example, to check a patient history in the LIS prior to pretransfusion testing:
1. Check for previous records using current patient name. (Check!)
2. If record is found, compare the following information on the current request to the previous patient record… (Check!)
3. Investigate and resolve any discrepancies (Check!) …. etc.
SOPs gained widespread use in TM when government regulation came to blood centres and later hospital transfusion services (TS). As quality systems were adopted, the TS developed laboratory and nursing transfusion-specific SOPs.

However, one problem with SOPs is their format, which does not correlate well with the way people learn. See Designing SOPs for Learners (Rohse & Cameron-Choi)

As the authors note, when it comes to instructions, less is often more, which is one reason why job aids (problem solving flowcharts, visual aids, and checklists) have become important tools in the TS laboratory.

On the wards, tools such as pocket guides were developed for various nursing functions, e.g., REACT and RESPONSE (Sunnybrook & Woman’s Hospital, Toronto).
A related approach to making SOPs more effective was used by Berte in working with the BC PBCO to develop a provincial TMS manual, in which SOPs lack preambles and begin immediately with work instructions. (See Developmental Model)

Although SOPS, simple checklists and other mistake-proofing mechanisms are extensively used by technologists and nurses, for various reasons these tools have been slow to reach physician training and practice in which the historical training model has been, "See one, do one, teach one":

With this history, and even given recent advances in medical education, Gawande's case for physicians using brief checklists is compelling.

A PubMed and web search identified a few physician-related “hits,” e.g.,

One TS strategy targeted to physicians has been to build checklists into transfusion requisitions in the form of requiring doctors to check specified rationales for ordering particular blood components and products, as show on this order form.

But overall, SOPs, checklists, et al., have not been widely adopted by TM physicians so far as I can determine and would welcome feedback on this issue.


Unfortunately, checklists, job aids, action-focused SOPs, and integrated picture / word instructions that I have seen are often too long and complex compared to the 2-minute ones discussed by Gawande. Considerable thought and expertise is required to develop concise yet inclusive checklists and their equivalents. Also, the checklist must be applied at the right spot in a process.

Implementation problems may occur if checklists are seen as "make work" adding another layer of complexity to already stressed staff. Or they may be viewed as unneeded crutches denoting weakness, particularly by physicians as Gawande notes.

A risk exists that checklists may lead to complacency, the same way double checking can. For example, if a nurse picking up blood from the TS lab has just seen a technologist confirm patient and donor identity against the requisition, he or she may not be as diligent in performing an independent check. The same may happen with two RNs checking donor blood unit and patient identity at the bedside prior to transfusion, which is why it's prudent to ensure that the first check picks up any discrepancies.

So, as in the blog's title, is the simple checklist "simply the best” for preventing missed steps and increasing patient safety? As documented in The Checklist, concise checklists that include all critical steps can be lead to significant improvements.

Of course, my question is false, since it's unnecessary to designate one mechanism as "the best" when multiple approaches have merit.

For example, the 'cartoon' approach to SOPs with integrated text and diagrams can also help prevent errors, and seems particularly useful when equipment is involved. This is why so many "how to assemble" pamphlets use this approach, unfortunately often with too little or no text as in Ikea ("I'll keel ya!") assemblies.

Given that multiple mistake-proofing tools are needed to accommodate different work environments with diverse procedures of varying complexity, and given that resources are finite, which tools should receive priority for development and evaluation?

In transfusion medicine, much effort and expense has been devoted to developing technological solutions such as barcoding and radio frequency identification (RFID).
Somehow we have come to rely, almost blindly, on technology as a cure-all for errors, most of it directed at eliminating errors associated with technologist and nursing procedures.

Is the technology approach emphasized because we think that machines are foolproof and only humans make errors? This always amazes me because working with computers on the Internet all day, I routinely experience multiple errors made by computers, many seemingly inexplicable. Or do you believe that such errors are invariably operator error?

Investigating simple mechanisms like checklists and SOPs for physicians who order blood occurs but has not received prominence. Perhaps The Checklist will result in some new research in this area.

Bottom Line
Regardless, all such mistake-proofing tools are meant to supplement, not replace, the practical training and professional development that health professionals receive as undergraduates and practitioners from their educators and mentors.

For my money, the "simply the best, better than all the rest" influence on patient safety is the attitude of the individual health care professional. I’m biased. It’s not based on an RCT, just a lifetime of careful observation.

Food for Thought

Tell me, what is it you plan to do
with your one wild and precious life?

- from "The Summer Day" by Mary Oliver
In an irreverent vein, to counter this lovely thought, remember this one?
Comments are most welcome BUT, due to excessive spam,  please e-mail me personally or use the address in the newsletter notice.