How can we help you?
Click on the section to view help questions and answers:
First Steps to Implementation:
The administrator is responsible for managing the relationship with ControlLab, keeping the registration data up to date, managing and delegating tasks to the Online System, receiving and distributing materials and correspondence, ensuring deadlines and analyzing results.
One of the main tasks is to promote the dissemination of information, starting with the purpose, operating rules and evaluation criteria of the PT. For this, it is important to ensure that all laboratory stakeholders read and understand the Gibi Quality Control information, the Participant Manual and the “Questions and Answers” available on the site.
To manage the day-to-day of the program there are the warning emails and the homepage of the Online System, which describes the rounds in progress with the deadline for reporting results and recently published evaluations.
For the monitoring the evaluations, there is the management report and the general accumulation. It is important to conduct a critical analysis of the results together with those involved and analyze the causes for inappropriate results immediately, based on the evaluation report, results profile and internal documents/records. To help you in this step there is the “Diary of Proficiency”, available in print format and use in the Online System (Education).
A new user always has doubts about the operation of the service and its responsibilities. For this reason, the first orientation is to read the Participant Manual, which explains how the program works, and the “Questions and Answers”, which clarify the participants’ most common doubts.
The main care is to run the control correctly so that the evaluation represents exclusively the performance of the routine and so that it is not contaminated with errors related to the quality control itself.
The main precautions include:
1. Read and follow the instructions for using the materials to avoid misunderstandings in the handling, since there are some procedures that differ from those adopted in routine patients;
2. Reconstitute freeze dried material with reagent water and with calibrated pipettor support, paying attention to the volume indicated on the label, since each material has a specific volume and errors in this step cause damage to the evaluation;
3. Use reconstituted material immediately to avoid variations resulting from improper storage;
4. Give special attention to dilutions and consequent calculations to avoid reporting erroneous results;
5. Check the need to convert results to the unit adopted by the program and perform the conversion correctly (see “Why do I have to convert my results to the unit adopted in the PT?”);
6. Identify the analytical system correctly (reagent, method, equipment, etc.) and report all requested information. According to the test, the absence of a data referring to the methodology may make its evaluation unfeasible;
7. Be aware of the response deadline and e-mail warnings (see “What happens when I stop responding to a round?”);
8. Contact Controllab immediately if you have any questions.
Although the modules are quarterly, participants with monthly payments split the amount. Thus, the invoice describes all the modules in which the participant is enrolled, and is charged the amount corresponding to the monthly apportionment. This is done so that the laboratory has a value for payment always equal, with greater ease of control.
It’s simple! Just do not answer. As the PT (proficiency testing) modules are composed of test groups, it is expected that one test or another will not be part of the participant’s routine. In this case, it is enough to never report any results, and this essay simply will not count points in your evaluation.
If you outsource, do the same. Do not submit results that were not obtained by your laboratory, or otherwise will appear on your evaluation and certificate as if it had been performed by your laboratory.
Each participant should only report results obtained in his routine, should not report tests performed by third parties or other technical unit of the laboratory. Just as you should not exchange information with other laboratories. This is because:
1. The objective of participating in the program is to have the opportunity to identify the good progress or not of each participant processes. When the results are obtained by another, there is no benefit.
2. When duplicating the result of a single laboratory, a trend is inserted in the result group that is not real, which may prejudice the statistical analysis and the evaluation of the other participants. The statistical model adopted presupposes only one result of each participant, which individually contributes to the determination of the target value used for the calculation of the evaluation range.
3. The certificate of proficiency and other documents proving participation in the program are issued to the enrollee in the program, which means that all the tests and results are executed by the enrollee.
The modules of the PT (proficiency testing) are quarterly to meet statistical requirements and ANVISA/REBLAS. The purpose of the proficiency testing is to identify errors, especially systemic ones. However, it is only possible to conclude whether an error is systemic (repetitive) or random, when at least two different items are tested.
With this, the first prerequisite is: never send a single sample per round. If we chose to submit monthly, we would have at least 24 items per year for each trial/module, and this would make the program much more costly.
On the other hand, there are no effective gains that justify such frequency and quantity of materials. Globally, 10 to 20 test items are adopted, with a common periodicity of 2 to 4 months.
But this rule presupposes that the laboratory adopt effective internal controls with more frequency in its routine (daily, between batches etc). Do not change internal control with external control. The first one is to verify its reproducibility and small deviations; already the proficiency test is mainly directed towards accuracy.
The rounds are sent according to the annual calendar, previously defined and distributed to the participants. The rounds are usually sent in the first half of the month, with some adaptations for the current year and holidays. Therefore, it is necessary the annual calendar available on the internet (website and online system).
Users of the Online System also receive an email notification of the round referral.
Not necessarily. Each module has its materials (test items), which can be:
• unique to more than one module (such as Biochemistry and Basic Hormones: all tests are dosed on the same item);
• unique for all tests of a module (as in Tumor Markers: an item for analyzing all markers);
• Individuals per test (as in Molecular Biology: one item for each type of test – HBV, HCV, and HIV);
• Different for test groups (such as in Cell Count: one material for global counting and one for differential);
• Items are labeled with an individual identification (e.g. BA01), and response forms (printed or online) repeat this identification for the tests to which the item corresponds.
This distribution is always the same and repeated every round, so that the participant becomes familiar with the system. Any change is communicated to the participants.
In order to make a statistical study and compare its results with those of the other participants of the PT (proficiency testing), it is necessary to adopt a single unit of measurement. Thus, we always seek to adopt the unit most used in the market.
To compare the results of different participants and to promote statistical calculations, it is necessary to set a default number of decimal places. Thus, a number of decimal places are chosen and/or relevant for the analysis of results.
The parameters calculated manually from multiple systems (using other parameters not directly related to the analysis at issue) are evaluated in an educational way in the program. These parameters do not present values obtained directly from the dosages, accumulate uncertainties due to the different variability contributions of the raw results and do not have the reliability required for an evaluation.
This “non-achievement – Na” account for your PT (proficiency testing) assessment and must be recovered at the end of the period with the special round. But attention! For tests that are not performed in their routine, the process is different. See the answer to “How should I proceed with tests that I do not perform/outsource.”
The rounds are quarterly and have a reduced number of test items (when compared to the volume of patients attended by the laboratory). Therefore, it is very important that the laboratory responds to all the rounds and tests of your routine.
A non-participation is only paid when the laboratory has stopped activities, such as instrument breaking. In this case, it is necessary to justify as soon as possible to Controllab. This allowance can only occur once a year, to maintain the annual representativeness of its results. See the answer to “If you cannot answer for lack of kits or broken equipment, what happens?”.
If, during the whole run-in period, the laboratory is stopped for operational problems (such as stopped equipment or lack of kit), the laboratory should contact Controllab and send a document proving their situation, such as the report or repair guide of the equipment in the period.
Thus, Controllab can discount this round of laboratory scores. However, this procedure may not occur more than once a year, since the sampling of the proficiency testing would no longer be representative for the certificate.
The evaluations today are released within two to ten business days after the deadline for sending results. New policies are being studied and implemented at all times, to increasingly reduce this evaluation period.
The evaluation process includes (1) receipt of the results, (2) statistical and technical analysis, (3) customer feedback analysis and (4) generation / access to evaluations.
In receipt of results, we are increasingly restricting the granting of increased response time to minimize the impact on the release of evaluations. In the analysis phase, the process has been optimized without giving up the analysis of qualified professionals and consulting the advisors in addition to the statistical treatment. These policies are to ensure a more thorough analysis and more reliable evaluations.
Customer feedback analysis occurs simultaneously with the statistical and technical analysis of the data and must be done completely for the release of the evaluation, since it may contain important information to define the evaluation. However, the volume of comments grows every day (ten to twenty times greater than when implementing the online system), with different information (see ‘When I comment on response forms do you respond?’) and often requires a lot of time in the evaluation process. Actions for the best use of the ‘feedback’ field and reduction of frequent questions have been adopted to minimize this impact.
The generation and access of the evaluation are already immediate since 2004. Once released, it is available to consultation in the Online System.
A delay in the evaluation is usually associated with some atypical performance in the data that requires a more complex analysis, such as material reanalysis (material quality control), contact with users / manufacturers or conducting research and consultation with third parties (other advisors).
Quality control materials are previously analyzed. They go through a quality control that aims to approve them regarding homogeneity and stability. Meanwhile, the evaluation of the participants’ results is the statistical model most used by the proficiency testing providers. This is because it has a consistent volume of results to evaluate accuracy and allows its comparison in a similar analytical system, since the control made by the provider does not cover all existing analytical systems.
However, it is important to note that a comparison is made between participants’ results and quality control to release an evaluation. Controllab always seeks consistency of these results so that there are no mistaken evaluations.
A result may not be evaluated because:
(a) didn’t form an evaluation group (less than 5 results);
(b) the evaluation group presented high variation;
(c) by determination of the advisory group, this decision is described in the document “Results Profile”.
In some specific situations, it is possible the evaluation in groups with less than 5 results, as statistical consistency of the data is verified.
There are a few possibilities:
1. Test with high variation: the results of the participants do not form a consistent statistic, presenting high variation. The option is to not evaluate based on this data.
2. Tests with predictable results: some analytes are only available (human matrix etc) in normal concentrations, and it is not possible to change them synthetically. In this case, they do not meet the requirement of proficiency test of “materials with unknown results” and no evaluation is done.
3. Insufficient number of evaluations: 8 to 16 items per year are sent according to the test, if more than 50% of the tests have not been evaluated (see the question “Why sometimes my results are not evaluated?”), the option is to not evaluate because the volume of evaluations is not representative for the year.
Certificate and related:
The certificate lists all tests with performance higher than or equal to that defined by ANVISA, in procedure GGLAS 02/43. We define these tests as “adequate” because they have reached the minimum acceptable level of performance.
The certificate also presents educational tests (see ‘Why are some tests classified as EDU?’). Inadequate tests, with less than acceptable performance, is not listed on the certificate. The “Management Report”, available for consultation in the Online System, maintains an updated list of inadequate tests.
Still having doubts?
Contact us and tell us how we can help you.