Basic QC Practices

A new abstract for the IFCC meeting in Istanbul returns to Italy to study the frequency of errors in laboratory processes across the Total Testing Process. Italy was where the study of errors rates really began back in 1997. Has Italy improved since then?

Another look at Italian Laboratory Error Rates, 2014

Sten Westgard, MS
JUNE 2014

Quality Indicators in Laboratory Medicine: The Experience of a Large Laboratory, L. Sciacoelli, A. Aita, A. Padoan, M. Plebani. Abstract 0962, IFCC 2014 WorldLab Istanbul. Abstract and presentation.

The landmark study of Mistakes in an Italian Stat Laboratory in 1997 set the dominant framework for our thinking about error rates in the total testing process of the laboratory. In that study, the pre-analytical phase of the total testing phase had the most errors. By transference, laboratories all over the world began to assume that their problems were primarily pre-analytical, and that analytical problems were negligible and could be ignored.

Nearly 20 years later, are laboratories in Italy (and thus, we seem to project, the world) still predominantly plagued by pre-analytical problems?

A new abstract of an Italian laboratory seems to suggest that dramatic improvements have been made in the pre-analytical phase, while analytical problems are strikingly significant.

Do you notice a difference between the pre-analytical and analytical metrics?

Now, let's add the defects-per-million rates to make this a little clearer:

Thus, we see that the errors and defects associated with the pre-analytical phase total to 18,977, while the errors and defects associated with analytical problems sum to 31,767. So tell us again, what’s generating more problems in the laboratory?

This may in fact be an underestimation of the impact of analytical problems, since each IQC failure and EQA failure impacts not just one patient sample, but perhaps hundreds or thousands of patient samples. If an IQC result is unacceptable, that may impact an entire shift, run, or day of testing patient samples. If an EQA result is unacceptable, that might affect weeks or months of patient samples. It’s hard to estimate the true impact of these analytical errors because we presume that the IQC is properly designed (if it’s an inappropriately strict IQC system, that may generate too many false rejections; if the system is inappropriately wide, then the system might actually be missing still more errors).

Now we should also note that not every pre-analytical Quality Indicator is listed in the abstract, and there are many more QIs that are currently tracked in that phase. So this narrow sample might not be capturing the whole scope of the errors occurring in the pre-analytical phase. Still, each of those pre-analytical problems tends to impact only one patient sample at a time, so the analytical problems might remain the bigger issue.

As with the previous study of pre-analytical and anaytical errors in a Romanian laboratory, ultimately this is not a contest to see which testing phase is worse. The total testing process is a three-legged stool, if any of the phases breaks down, the impact to patients is always bad. These recent studies show that the focus of the last nearly 20 years has had a real impact, improving pre-analytical processes to an good-to-excellent level of performance. Compare these rates to earlier studies of laboratory error rates.

The bad news is that the analytical phase continues to be a problem, and in fact may now be the major problem labs face. We need to make sure we're not exclusively paying attention to the pre-analytical phase with our efforts.The imprecision and bias of our methods are often overlooked, and they can generate invisible defects, causing thousands of results to be distorted from their true clinical value.