Essays

The final, final, final, final, final CLIA Rule was published on January 24, 2003. After 15 years, CLIA-88 is finally final. The last rules contain some big changes in terminology, organization, and regulation. It promises to have a huge impact on the "compliance" minimums for QC, on the responsibility for QC in the laboratory, and on the requirements made on manufacturers to provide quality instruments and methods. Read this essay, first of a series, for in-depth and comprehensive coverage of these new rules.

• Key changes in organization and terminology
• Key changes in regulation
• Still some confusion!
• We'll try to help!

This is the first in a series of discussions of the final, final, final, final, final CLIA regulations that were published January 24, 2003 in the Federal Register. That's the meaning of "Final5" in the title. This is the fifth revision of the Final Rule and finally, in the year 2003, CLIA-88 is final.

Readers can obtain a copy of the Final Rule by looking for a link "Requirements Relating to Quality Systems and Certain Personnel Qualifications; Final Rule published 01/24/2003" at this URL: http://www.cms.hhs.gov/CLIA/

The Final Rule is covered in pages 3640 through 3714, but many of these pages are background material. There is an extensive "crosswalk" table (9 pages) to help readers locate the requirements in the reorganized document, many pages of comments and responses that led to the final rule (23 pages), a summary of changes (4 pages), estimates of financial impact (22 pages), and the final rule itself (13 pages). Out of 54 pages, you should first read the 13 pages from 3702 to 3714 to see what is new and different in the Final Rule. Read the other materials if you have time and want some insight into why changes were made.

This discussion will focus on key changes in organization, terminology, and regulations that will have major impact in the application of the Final CLIA Rule. Later discussions will get into more detail about specific parts of the Final Rule and ideas on how to address and implement those provisions.

Key changes in organization and terminology

For those of you who are familiar with earlier drafts of the CLIA regulations, there are some key changes that will help you understand the Final Rule. For a complete and detailed list of all the changes, refer to pages 3674 to 3678 of the Federal Register.

• Organization follows the flow of a patient specimen through the laboratory, i.e., pre-analytical, analytical, and post-analytical systems. This reorganization has been hyped as a big change to make the document consistent with a "quality systems" approach, as described in the NCCLS GP26-A2 and HS1-A documents. It's not such a big deal since there are only two columns for "Pre-analytical Systems" [section 493.1240, page 3705-6] and one paragraph for "Post-analytic Systems Assessment" [section 493.1299, page 3713], with 8 pages in-between for "Analytic Systems." Most of the Final Rule focuses on the quality of the analytic phase and there is much more detail in this area than in the pre- and post- analytic areas.
• "Non-waived Testing" replaces High and Moderate Complexity. Because of the large increase in the number of test methods that have been classified as "waived" testing, the methods remaining tend to be mainly those of high complexity, thus the original classifications of moderate complexity and high complexity are now lumped together under the classification of "non-waived testing." This is actually the organizational change that contributes most to the simplification of the Final Rule. No more switching back and forth between high and moderate complexity due to different requirements - it's just one set of requirements for non-waived tests - except for personnel, where the distinction between moderate and high complexity tests still affects the qualifications for personnel performing those tests.
• QA now means quality assessment rather than assurance. This change is supposed to focus the quality effort on assessment of performance in all phases of the testing process and also place new importance on establishing and following written polices and procedures, taking corrective action, and making changes in policies and procedures to prevent problems from reoccurring. In my view, assessment actually leads to assurance when coupled with these steps to make corrective actions permanent by changes in policies and procedures that prevent problems from occurring in the future. This change in terminology adds little to the understanding of quality management and the quality improvement process, but you need to be aware of it anyway. If the inspector asks about your quality assessment procedures, you need to recognize the new terminology. Everything that was formerly quality assurance is now part of the section on analytic system.
• Definitions added for calibration, calibration verification, FDA-cleared or approved test system, reportable range, and test system. Evidently there has been so much confusion about the exact meaning of these terms that new definitions have been added to the Final Rule. See page 3702. These clarifications may be driven, in part, by CAP's use of different terms for analytical measurement range (AMR) and clinically reportable range (CRR). There are a lot of questions about this and we will take up some of the specific issues in this area in a later discussion.

Key changes in regulations

• NONE FOR WAIVED TESTING!
• Final Rule becomes effective April 24, 2003. All laboratories must now meet and follow the QC requirements as of April 24, 2003.
• Laboratories must validate the performance of all methods introduced after April 24, 2003. For unmodified FDA-cleared or approved systems, this includes the following performance characteristics: accuracy, precision, reportable range, and reference intervals (normal values). For modified or non FDA-cleared systems, additional characteristics include analytical sensitivity (for detection limit), analytical specificity and interference, reference intervals, and any other performance characteristics required for the test. CAP has consistently required validation of all methods, so if you're a CAP approved laboratory you've been doing this all along. JCAHO has been a little more selective, but basically if you're JCAHO approved, you've also been doing this. The main impact will be on small labs that up till now have only been following the manufacturer's directions, have a procedure manual, analyze two controls per assay, and document corrective actions.
• Laboratories must determine calibration and control procedures for new methods. This requirement is found in section 493.1253 (3) on page 3707. To my knowledge, this is a new statement that appears for the first time in this Final Rule. It is explained as follows: "The laboratory must determine the test system's calibration procedures and control procedures based upon the performance specifications verified or established." The reason for this is probably due to the two items below - the lack of FDA clearance of QC instructions and the transfer of QC responsibility to laboratories.
• FDA clearance of manufacturer's QC instructions has been eliminated. This is the part of the original CLIA rule that has been postponed and postponed and postponed and postponed. At long last, they have just killed it because of the government's inability to implement the process. FDA clearance of QC was a key feature in CLIA because it provided the basis for allowing laboratories to follow the manufacturer's directions for QC. If those QC directions were reliable, then the laboratory should be able to produce quality results. If not reliable, there would be no guarantee that quality would be okay. The responsibility for quality now rests with the laboratory because the manufacturer's directions will not be validated by FDA.
• Laboratories are responsible for having control procedures that monitor the complete analytical process, take into account the performance specifications of the method, detect immediate errors, and monitor long-term precision and accuracy. This is where the responsibility for QC has been re-assigned to the laboratory because the manufacturer and FDA are now off the hook. The laboratory and the laboratory director are responsible for doing the right QC right. This change in responsibility is actually good because neither the FDA nor the manufacturer are able to determine what QC is right in your laboratory. Only you can do that, based on the quality needed for your patient population, the performance observed for the method as operated by the people in your laboratory, and the QC procedures that you are able to implement in your laboratory. The regulations don't provide any guidance on how to do that, but we will try to help you by providing training materials and tools on this website. We will cover this in more detail in later essays.
• Minimum QC now two levels per day rather than per run. In the absence of doing what is required in the item above, the default requirement is a minimum of two levels of control per day. Note the change from per run to per day in the Final Rule. This really isn't such a big change since a run was earlier defined as being a maximum of 24 hours and many laboratories already equate a run with a 24 hour period. There are some additional details that apply to certain kinds of tests, e.g., blood gases where the requirement is still one control every eight hours or one control each time specimens are tests unless automatic instrumentation internally verifies calibration at least every 30 minutes" (see specialty areas on pages 3708-3712).
• Reductions in QC requirements in specialty and subspecialty areas. In general, the amount of QC has also been reduced in many (but not all) of the standards for specialty areas (pages 3708-3712). These changes to reduce the amount of QC are the basis for claiming the Final Rule reduces the financial burden of QC. This will receive a lot of attention because it shows the government's willingness and interest in containing the cost of laboratory testing. For example, the standard for control procedures in hematology now reads "one control material must be tested each 8 hours of operation," whereas two were required before the Final Rule. Before reducing QC, laboratories should consider the implications of the requirements to "detect immediate errors" and "monitor over time the accuracy and precision of test performance that may be influenced by changes in test system performance and variance in operator performance." These two requirements precede the specialty requirements and still must be satisfied.
• Proficiency Testing (PT) requirement for 80% agreement has been restored. I can't say that I spotted this when reading the Final Rule, but it is clearly stated in the discussion of "highlights and organization" on page 3641. The main impact of this requirement is on the scoring of PT surveys and samples that might not be graded if a 90% criterion for agreement is applied. Many PT samples were not getting graded when the 90% criterion was applied and a non-graded sample actually benefited the laboratory (was as good as getting an answer that was correct within the PT criterion for acceptable performance).
• Non-PT'd tests must have accuracy verified by the laboratory. Any tests performed by the laboratory, but not included in a PT program, must have their accuracy verified by the laboratory at least twice per year. This is not really new, but is being re-stated more clearly. It means the laboratory will need to define some procedure to periodically compare performance on patient specimens over time, compare performance with another laboratory, compare performance with a reference laboratory, or test reference materials that have known or assigned values.
• Laboratory director qualifications for high complexity testing clarified and implemented as of February 24, 2003. This affects laboratories that formerly performed the high complexity tests that are now part of non-waived testing. The laboratory director must hold an earned doctoral degree in a chemical, physical, or biological, or clinical laboratory science from an accredited institution and be certified by a board approved by HHS. Alternative qualifications include 2 years of laboratory training or experience and 2 years of laboratory experience directing or supervising high complexity testing prior to February 24, 2003.

Still some confusion!

At least, I am still confused about some issues and applications of the regulations. I suppose it might be considered a good measure of a regulatory document that the end result is some confusion. That leaves some room for interpretation and flexibility in implementation. CMS often uses the State Operations Manual (CMS Pub. 7) to provide clarifications, so we will have to wait for the new edition that goes along with the Final Rule. Here are some of the questions that laboratories will be asking:

• Can laboratories just run two controls per day and get by? Probably! In spite of the language that implies the need for control procedures to detect immediate errors and monitor the long-term precision and accuracy of the method, a simple default remains - just run two controls per day.
• Can laboratories deviate from the practice of running two controls per day? Probably, as long as the laboratory is able to defend their alternate practice. That defense will need to consider the performance specifications verified or established by the laboratory, the control procedure's ability to detect immediate errors and monitor long-term precision and accuracy of the method, and the establishment of criteria for acceptability for all control materials.
• Is Electronic QC (EQC) still okay? Probably not! EQC had a temporary allowance, pending publication of the Final Rule. It would be hard to see how EQC could be approved when the new language states that QC should monitor the whole analytical process. However, the State Operations Manual may provide some wiggle room. There certainly will be pressure by manufacturers to keep EQC as an alternative to the use of liquid controls. This should be counter-balanced by the responsibility placed on the laboratory and laboratory director to perform appropriate QC. Most likely laboratories will need to perform liquid QC as a complement to electronic QC.
• Should laboratories still follow the manufacturer's directions? Definitely, but that doesn't eliminate the laboratory's responsibility to verify the manufacturer's performance claims. In addition, the lack of any specific directions on QC doesn't let the laboratory off the hook on QC. The laboratory should establish specific directions and, if conditions warrant, do more and better QC than recommended by the manufacturer. Also, if the manufacturer's QC directions require more QC than the minimum required by CLIA, the laboratory must follow the manufacturer's directions.
• Do laboratories need to gear up to validate new methods? Definitely, that appears to be the major impact of the Final Rule for many small laboratories that have been inspected for compliance to CLIA, rather than inspected for CAP or JCAHO accreditation. Any laboratory that performs non-waived testing will have to perform validation studies for any new methods introduced after April 24, 2003.

We'll try to help!

Westgard Web is enlisting the help of Dr. Sharon Ehrmeyer, who is the most noted authority on the CLIA regulations. To be sure we stay grounded with information that will be useful in the laboratory, Elsa Quam and Patricia Barry (Quality Specialists in the Clinical Laboratories at University of Wisconsin Hospital and Clinics, UWHC) will review these materials and help us provide practical details. Dr. Teresa Darcy (Medical Director at UWHC Clinical Laboratories) will work with us to add knowledge of the CAP and JC inspection guidelines and processes.

Regardless of any ambiguity in the regulations, we will provide advice on what is right and best for the laboratory and the patients you serve. "Doing the Right QC Right" is our guiding principle and continues to be the best approach for defining quality management practices.

See our CLIA Final Rule Section for more

James O. Westgard, PhD, is a professor of pathology and laboratory medicine at the University of Wisconsin Medical School, Madison. He also is president of Westgard QC, Inc., (Madison, Wis.) which provides tools, technology, and training for laboratory quality management.