High compliments on the article, “A1c: added oomph to diagnosing diabetes” (July 2010).
Although the article was highly informative technically, there was no discussion of the practical side—how many over what period. As a chemistry supervisor at one of the veterans hospitals affiliated with a large teaching university, I’ve seen as many as six A1c’s ordered on a single patient in a period as short as 10 days. Many rotating residents do not review previous orders and override the computer “gate” designed to remind them they are exceeding standards. It’s not a lack of peer review. In our highly litigious society, many physicians protect themselves by over-ordering. They fail to review patient history or are just not concerned about the expense involved in offering tests such as A1c’s, or both.
Our VA system allows four A1c’s per year on patients known to have diabetes, two per year on patients who are borderline, and one per year in cases of previous “non-diabetic” results. Is this a fair standard?
Richard B. Lorenz, MLS(ASCP)
Laboratory Scientist, Chemistry,
Loma Linda (Calif.) VA Medical Center
• David B. Sacks, MB, ChB, medical director of clinical chemistry at Brigham and Women’s Hospital and associate professor of pathology, Harvard Medical School, replies: The American Diabetes Association recommends four times per year in patients not achieving glycemic goals and two times per year in patients with “stable” diabetes. The system described seems to be a fair standard. The only caveat is that the new criteria for diagnosis require two A1c tests, so the number may need to be increased in some of those patients in the first year only.
Specimen ID errors
I applaud and appreciate the attention your article, “Unmasking specimen ID errors, every step of the way” (June 2010), has focused upon the identification of errors in the anatomic pathology department. While the clinical laboratory may deal with greater volumes, I would argue that anatomic pathology departments work with more irreplaceable specimens and, as the article pointed out, one case can lead to several specimens, which can lead to several blocks per specimen, which can then lead to several slides per block. Added to this volume is a handoff of materials from accessioning to grossing to cutting to pathologist review. There is no wonder that specimen identification problems are such a significant issue.
However, I feel compelled to correct the misperception that a commercial system that addresses these error-sensitive workflow issues does not exist. PathView Systems’ Progeny was first installed more than 4.5 years ago to help resolve exactly the issues described. In Progeny, bar-coded requisitions are scanned individually and the order is displayed automatically at accessioning. The user can create the accession and immediately note and resolve any identification problems with the specimen as it arrives from any of the collection sites before an actual accession is created. Specimens are scanned individually to create the blocks just for the scanned specimen. Blocks are scanned individually, and only the applicable slide labels for the scanned block are generated.
The article is correct in that the software that takes advantage of the latest bar-code technology is not commonly available in the commercial world, but it does indeed exist. We invite anyone who might be interested to ask for a demonstration.
Vice President, Marketing and Sales