A pathologist at a small community hospital in New Jersey reports that her group types up their own free text transfusion reaction and antibody identification reports. Currently they are in the process of changing their lab-system so that they can use 'canned comments' for each type of transfusion reaction, antibody identification, and associated recommendations. The inquiring pathologist would like some feed-back as to what others are doing with regards to using canned reports, and what resources are available for such comments.

The following comments have been received.

ADDENDA Oct. 16, 2006

1. Colleagues at the University of Michigan have graciously shared a series of basic templates (see attached) which the transfusion medicine physicians (and residents in training) use for each of the major transfusion reactions (hemolytic reaction, TRALI, sepsis, etc), including workups for platelet refractoriness. These transfusion reaction templates provide a systematic guide for the collection and incorporation of patient history, vital signs, radiology and laboratory results into a report. Their physicians do not routinely signout all antibody workups, however, they use the Delayed Hemolytic Transfusion Reaction template as a 'guide' for signout of complex antibody identification problems.

ADDENDA Nov. 1, 2006

2. A transfusion medicine physician in the greater Chicagoland area reports that he uses canned text fragments extensively to generate comments on antibody identification results and transfusion reactions. In addition, he adds a somewhat shorter canned comment to the "Allergies" function of their electronic medical record. This provides a warning flag and information on the compatibility problem in a longitudinal fashion rather than in an encounter or test-based format. A typical antibody comment attached to an antibody identification result has one paragraph summarizing the serologic findings, a paragraph discussing their implications for the patient including suggested further testing, and his name and title. The responding physician is curious what criteria others use for performing such "consults". He writes one for each transfusion reaction. But for blood group antibodies he only writes one for the first antibody identification result, or for subsequent work-ups in which there is a new antibody. What do others do? How widespread is the practice of writing any comment to these results? Do others write comments on insignificant cold auto-antibodies, etc?

ADDENDA Nov. 7, 2006

3. A colleague affiliated with the National Blood Service in the UK reports that the Red Cell Immunohaematolgy arm of the NBS makes extensive use of canned/coded comments, via their national IT system. He cautions that these opinions below are his own, and do not necessarily reflect those of the NHSBT/NBS.

The wording used in the canned comments is based on information found in BCSH guidelines, NBS policy documents, with additional comments based on experience of managing problem cases. From personal experience he reports that there are two, main, day to day problems with using canned comments;

1. how the comments are coded
2. how specific the comments are

He states (verbatim): "With respect to #1, for example; if your code for 'Give Jka negative blood' is JKA, you'll remember it, if it's F34 you'll probably have to look it up every time you want to use it (you will have lots of other, similar, codes). Using ISBT codes can seem attractive but when you've got around 50 for RH you can struggle to remember that anti-Cw is RH8, let alone that anti-cE is RH27. So keep your codes memorable, provided they print out in the 'correct' way, who cares what you call 'em."

"With respect to 2) you can easily design some very specific text to go with a code. For example, we have around 60 individual codes associated with reporting and managing antenatal patients with red cell antibodies (out of a total of about 240 'reporting' codes, but there a plenty more available for other reasons). This means you have little chance of remembering the code for a given text string and require code sheets, long drop down menus and/or flow charts readily to hand. I think it would be easier to 'build' comments from a small core of codes, which you are much more likely to remember. Of course you could commission a 'rules based' system that automatically inserts text into reports, depending on the results you put in. We've got a bit of that too. Seems like a good idea, but try building them - now you see the problem."

ADDENDA Nov. 8, 2006

4. A transfusion medicine colleague who works in a hospital in Washington, DC reports that their transfusion service uses templates to report the following:

1. antibody work-up reviews- which are done for all newly identified antibodies
2. transfusion reaction reports- which are done in all initiated cases; they have templates for febrile and allergic reactions, and they are writing one for delayed serologic reactions.
3. 'transfusion reviews' for special products requests; they use templates for when they recommend the transfusion of leukoreduced, irradiated, and/or CMV seronegative blood products.

If a case report falls outside the templates, the report is free-texted. The reports are all done by blood bank physicians. Residents on service write the draft, attendings sign the cases out. Attending physicians "do it alone" when there is no resident on service.

Please submit comments to the e-Network Forum.

Ira A. Shulman, MD
CBBS e-Network Forum Editor & Moderator

W. Tait Stevens, MD
CBBS e-Network Forum Assistant Editor & Moderator