MCC Metric of the Month Blog: Central Laboratory Sample Analysis

Central laboratory analysis of samples collected by the investigational sites is an important source of safety and efficacy endpoint data. When labs are unable to generate reports due to sample acquisition and/or stability problems, the sponsor runs the risk of potential safety issues and costly delays within a study. This month, let’s look at a key performance metric collected and reported by central labs – MCC Lab Performance Metric v2.1 #7: the percentage of “reportable” tests (e.g. percentage of tests not cancelled due to sample issues).

Why this metric is important:  This metric enables sponsors and labs to identify when there are data quality problems (missing lab data due to missing reports) and take early action to address sample acquisition quality and/or sample stability problems. Monitoring this metric on an ongoing basis and addressing issues should prevent additional data quality problems. Site monitors should be made aware of regional or site – level sample quality problems prior to conducting remote and/or onsite visits.

Definition: To identify the percentage of test that are able to be analyzed, compare the number of tests the central lab reports to sites against the number that should have been reported during a given time period. A test is defined as reportable if it has not been cancelled for a particular reason (e.g. QNS, stability, etc.).

How to calculate this metric:

The formula is the (number of tests which were not cancelled for a reason or "general areas of concern") divided by the (total number of tests received for testing) multiplied by one hundred. Typically, this metric is calculated on a quarterly basis.

Example: 

In this example, we have a study where 50 samples were received by the lab for analysis for 100 tests (2 tests per sample) within a quarter. From those 50 samples, 5 were unable to be analyzed due to stability problems. Thus, the remaining 45 samples had 2 tests performed for a total of 90 tests. If we follow the formula, the percentage of reportable tests is 90 percent.

95% is a reasonable target for this metric.

What you need in order to measure this:  All you need to calculate this metric is the number tests that were not cancelled for a given reason (e.g. QNS, stability, etc.) and the number of total tests that should have been performed if all of the samples received were able to be used for testing within a certain time period (i.e. quarter)

Things that you can do to improve performance:  Once you are tracking this metric, if the percentage of tests provided are not meeting the target of 95%, you should review processes and procedures as well as consider the following: 

• Volume of samples (is it a small number of samples so the impact is minimal?)

• General reasons and trends for the loss (e.g. clotting, mishandled, hemolysis, etc.)

• Geographic region trends within and/or across studies which could be related to a regional transportation/courier problem or a regional training issue

• Site-specific issues which may require retraining

• Sample mislabeled

• Sample acquisition quality problem   

• Specific/primary markers to determine if lose rate may need to be reviewed in depth for a specific protocol

Companion metrics:  Another metric that you should consider in tandem with this metric is MCC Lab Metric v2.0 (6) lab tests reported within the expected turnaround time.

Linda Sullivan, CEO, Metrics Champion Consortium, [email protected]