College of American Pathologists
Printable Version






cap today



May 2006

bullet Middleware

I enjoyed your article, "Smooth Operator in the Lab: Middleware" (February 2006) but wanted to raise several points not addressed in the piece.

First, the article noted that middleware is offered by both instrument/ automation vendors and "pure middleware" vendors, but didn’t suggest which option was preferable. In my experience, middleware from instrumentation/automation vendors tends to be a better choice, primarily because satisfactory performance is guaranteed—that is, backed by an instrument/ automation vendor warranty.

Keep in mind that the primary goal with middleware is to improve the efficiency of lab operations by adding functionality not readily available on legacy laboratory information systems. You will get far more efficiency—performance-wise and financially—from your investment in instrumentation and automation if you take the approach we did at Oklahoma University Medical Center: a totally integrated solution provided by essentially one major vendor with an established track record in the field.

The middleware we acquired was developed to conform to the specifications of our instruments and automation system and to optimize their performance. This approach provided an excellent fit from the standpoint of functionality and flexibility, in my view.

Another reason for this approach is user-friendliness. Our instruments, automation system, and middleware all have the same technologist user interface, so these different middleware applications all look and feel like one application to the user. This, too, contributes greatly to work efficiency at our institution.

Finally, our middleware has valuable features that are essential—in particular, critical value alerts and exceptional data event alerts. The former signals technologists that a critical value has occurred, with brightly colored pop-up screens. This system can then communicate the protocol to be followed, including contact information for the proper clinician to be notified. It even provides a log so the technologist can enter the time the message was delivered and the name of the person who received it. Timely critical values reporting is highly valued by JCAHO, as detailed in its National Patient Safety Goals. JCAHO doesn’t specify middleware as the preferred response to that challenge, but it’s hard to imagine a better one.

As for exceptional data events, these occur so infrequently in labs that they rarely get handled well. Our middleware helps correct this problem by, again, providing management-defined protocols to be followed for each discrete event type. Thus, all problems get handled by technologists consistently—between shifts, on weekends, and even on holidays.

Our lab’s goal at Oklahoma University Medical Center is a continuous, real-time flow of information from the vein of the patient to the brain of the physician. That means no batch processing to slow the testing process at any stage, from the physician order all the way to the real-time reporting of results. Indeed, we now process samples so smoothly and quickly that tests only rarely get ordered as stats. Without our instruments, automation, and middleware working in the well-orchestrated manner they do, that level of performance would be virtually impossible.

Kenneth E. Blick, PhD, ABCC, FACB
Professor and Director of Clinical Chemistry
Oklahoma University Medical Center
Oklahoma City

bullet Articles as ads

I usually enjoy the articles in CAP TODAY, finding them informative and useful. However, some of the articles are nothing more than advertisements for particular businesses. The March 2006 issue was particularly annoying. The article "Owning, Rather Than Buying, Reference Lab Services" was excessively positive about Warde Medical Laboratory, and at the end of the article a traditional advertisement for the same company was juxtaposed!

I would appreciate a disclosure at the beginning of such articles that a financial connection exists between the writer, publication, and subject of th article.

Frederick E. Dennstedt, MD
Parrish Medical Center
Titusville, Fla.

    arrow I was sorry to learn that you considered the article little more than an advertisement for Warde Medical Laboratory and that it appeared that the writer, CAP TODAY, and Warde had a financial connection. I can assure you no such financial connection exists in this case or in the case of any other CAP TODAY article. As editor, I assign a story to be written only if I think it will be of interest and useful to CAP TODAY’s readers.

The co-tenancy concept was one that I felt met that criterion. I knew that the story of the Michigan Co-Tenancy Laboratory had been told in presentations at meetings and that the cost of reference laboratory tests is a problem for many labs. With co-tenancy a solution to a problem for some, I thought that others might benefit from learning about it. It falls into the category of business models, solutions, and other innovations laboratories might want to read about—which we consider worthwhile editorial for our magazine.

MCL-Warde decided to purchase an advertisement to accompany the editorial coverage after I decided to publish the article and accompanying sidebar about Hurley Medical Center. I did not, I assure you, decide to publish an article about MCL-Warde because they decided to purchase an ad, though I realize it could appear that way because of where the ad was placed. Fact is, advertisers like to publish their ads in issues that contain editorial related to their tests and instruments (and some request placement near the related editorial, a request we may think is unwise but grant when we can). But in every case in CAP TODAY, it’s the advertiser making its decision based on it being informed about related editorial, not the editor making her decision based on what company is buying ad space.

I was pleased to hear that you generally do find our articles informative and useful. That is what I aim for with every story, though I clearly missed my target in this case with at least one reader (and possibly others).