June 2009
Feature Story

Anne Paxton

The CAP’s Laboratory Accreditation Program checklists have done something non-traditional for more than 20 months: They haven’t changed. While the program was under review by the Centers for Med­icare and Medicaid Services for deemed status under CLIA—in response to the CAP’s September 2007 application for renewal of deemed status—the LAP checklist elements were frozen and could not be altered.

But in March, the accreditation program once again received the maximum six years of approval from the CMS for all its specialties and subspecialties, and a whole slate of checklist enhancements, which take effect this month, can be unveiled. “We’re very happy to have received CMS’ approval and proud of the program that continues to be the gold standard,” says R. Bruce Williams, MD, chairman of the CAP Commission on Laboratory Accreditation. “Our changes were delayed somewhat during the CMS approval process, but behind the scenes we’ve been very diligently updating the checklist and adding new sections.”

Along with clarifications and revisions throughout the checklist, some of the most significant additions are in the anatomic path­ology area. An entirely new section that has been in development for the past four or five years will be devoted to digital imaging. “It’s a very new area,” says Stephen Sarewitz, MD, chair of the program’s Check­list Committee, “and we’re feeling our way along with it. We spent a lot of time getting information from people in the field about what these checklist requirements should include.”

The initial draft of this section tried to graft the requirements from clinical lab testing onto digital image analysis. “And it didn’t really work; it didn’t fit, so we made a lot of changes and revamped it. So this version is out there to see how it works and whether labs and inspectors find it useful in helping us to assure quality.”

It covers all the basics, he says. “We have a section on validating the system, on calibration, on quality control specimens, thresholds for calling something positive, and acceptable controls before reporting results.” The section on DNA staining requires criteria for acceptability of histograms, DNA distribution, appropriate controls, and identification of an aneuploid cell population.” Personnel qualifications and items that have to be included in reports are also listed.

The checklist already includes elements for synoptic reporting of cancer cases, with the requirement that for cancer specimens, all standard elements needed to stage and grade a case be included in the report. But the American College of Surgeons’ Commission on Cancer brought to the CAP’s attention published data indicating that pathology reports that use the synoptic format more often contain all the required elements for cancer diagnosis than do narrative reports, Dr. Williams says. “Many pathology labs already use synoptic reporting, but for those that don’t, we thought it was important to make them aware that synoptic reporting appears to be a better means of communicating required elements to clinicians.”

“Until now, the narrative form has been acceptable—for example, in a paragraph the pathologist might say the tumor was found to be invading to a depth of x, the grade is y, vascular invasion is or is not seen, and so on. A synoptic format gives bullets with elements lined up: Vascular invasion: Yes or No. Depth. Size of tumor. All are presented in list form.” It helps remind pathologists on every case what they’re supposed to include, Dr. Williams says, and it helps physicians assimilate the information quickly. “It gives them a template they can always rely on being there to check elements they are interested in,” he says.

“What’s being added are that the lab actually has to audit the reports every year to make sure all these elements are present,” Dr. Sarewitz says, “and at least 90 percent of reports must contain all appropriate elements. The requirement for having the elements is a phase two requirement, while the auditing requirement is a phase one.”

The committee decided to revive use of a “phase 0” status for the requirement that data elements be reported in a synoptic format—that is, not in a narrative. Phase 0 is “kind of a pre-requirement,” Dr. Sarewitz says. Though it is a deficiency, the lab will not have to respond nor will the phase 0 count against the lab in the calculation of percentage of deficiencies. “We want labs to get used to moving into it,” he explains, “so that in a later checklist edition it would move into becoming a phase one requirement.” Assuming nothing unusual happens, the next checklist edition or the one after that will turn it into a phase one.

In the clinical pathology area, another new section of the checklist is devoted to direct-to-consumer testing. This section has been in development for quite a w­hile, says Dr. Wil­liams. “We felt it was important to be patient advocates in this area.”

“We know it’s out there and becoming very pervasive,” Dr. Sarewitz says, “and regardless of one’s philosophy about that kind of testing, we know our accredited labs are offering it, and we want to be certain the appropriate quality practices are being followed. If you’re going to do it, you need to do it right.”

An enhanced introduction to the test validation section of the laboratory general checklist will offer better guidance to laboratories. The Checklist Committee wanted to clarify that though the test method validation section is aimed at blood, “a body fluid like pleural fluid has a different matrix than blood, and that may affect the test method or instrument, so you don’t get the same results if you’re testing for the same analyte,” Dr. Sarewitz says. “With these other kinds of body fluids being tested, you have to establish method performance specifications as well.”

“So what the section now says is if the lab can show the instrument gives the same results for the fluid as for blood, or it can show there’s a reproducible difference—such as always getting a count that’s 10 percent higher with fluid, which can be adjusted accordingly—then that’s acceptable.”

“If not, the lab has two choices. It can do all the studies of accuracy, precision, and other specifications for the fluid separately. Or, in cases where that’s not practical—such as where there are not enough fluid specimens to do the study—the lab should put a disclaimer on the report saying the method performance specifications have not been verified for this type of specimen.”

An example is in the case where no reference ranges exist, such as with pleural or peritoneal fluid. “We’re not saying they have to do the study, but there needs to be an indication that it wasn’t done.” The checklist now says the lab director must approve all method performance specifications.

Though not all quantitative meth­od performance specifications apply to qualitative testing, “we wanted to put in something about their having to be done as well,” Dr. Sarewitz says. “So it’s up to the lab to do what’s needed to validate qualitative tests.”

Revisions have been made to the checklist requirements on referral of specimens to reference laboratories. “There is really only one discipline, histology, that is not covered under CLIA, so we added that if a lab is going to send anything out for histologic processing, it has to be to an organization accredited by CAP, or a CAP-accepted accreditor,” Dr. Sarewitz says. It’s somewhat of an anomaly, he explains, because CLIA does cover histopathology—the microscopic interpretation of slides—but CLIA was never extended to histology because it was felt it involved purely processing of tissue specimens, not patient care.

The Checklist Committee also decided it was time to pin down the term “periodic.” There have traditionally been two diametrically opposite philosophies about whether the term should be retained, Dr. Sarewitz says. “We didn’t want to be prescriptive; we wanted labs to be able to adjust their processes to their own environment. But I think that argument lost out. We’re in a more regulatory environment now, and labs are less comfortable with having that kind of leeway. There were many questions from the field about what we mean by ‘periodic,’ and we want to be sure labs are doing these important processes at reasonable time intervals.”

In most cases, the term has been clarified to “at least annually,” but in a few it was changed to monthly or quarterly. “One section where we’re not changing it—because basically we don’t know how to change it yet—is the computer section.” Experts in the field, he says, are uncertain how often some of these checks relating to lab information systems should be done, and since it’s not a CLIA issue in any case, “it’s been left for the time being.” But the statistical review of quality control data—for example in CHM.14500 of the chemistry checklist—has been changed from “at specified intervals” to monthly. “‘Specified’ was not being interpreted in a standard way,” Dr. Sarewitz notes.

The microbiology checklist contains a number of revisions. The first change covers microbial biochemical identification systems. These systems may have numerous biochemical tests and for each new lot of test reagents the lab had to run a test with a known positive and negative organism. Though this is a CLIA requirement, “that was recognized to be overkill because these are very robust systems,” Dr. Sarewitz says.

The new checklist item comes from revised CLSI document M50, developed with input from the CMS, on an innovation called streamlined quality control. “It’s a way of using just a few indicator organisms to validate a new lot, and it can be used only if the manufacturer has developed and validated a streamlined quality control. If it has, then the lab can basically use this shortcut and not have to do extensive testing.”

Test kits for testing for bacterial antigens have until now been subject to a requirement that if the kit had an extraction phase, the lab had to use daily external QC. “That has been changed so that those kits will not be required to use external QC—providing they’re not high complexity,” Dr. Sarewitz says. “For high complexity, all kits have to have external QC runs daily; for all other types they can rely on internal controls for daily QC.”

The entire molecular microbiology section in the checklist has been expanded and reorganized because of the increased use of molecular techniques in microbiology labs. “We’re very excited about these changes in the checklist,” Dr. Williams says. Previously, the molecular section of the microbiology checklist was devoted to molecular infectious disease testing by FDA-approved tests only, but it’s been rewritten to cover laboratory-developed or laboratory-modified microbiology tests as well. So now labs won’t have to use the molecular pathology checklist for such tests. “This will save time for inspectors as well as labs,” Dr. Sarewitz says, because this portion of the checklist is divided into sections according to the complexity of the test involved and labs can use just what applies. The molecular microbiology section does not have to be used for nonamplified molecular tests that are run directly on positive cultures, he points out; for those, just the regular microbiology checklist can be used.

In the molecular microbiology section only, FDA-approved high-complexity tests with internal controls will no longer need to be subject to external QC each day, once the laboratory has validated the internal controls against external controls. “That’s because there are really only one or two tests like that and they’re very robust,” Dr. Sarewitz explains. “The cost of doing external QC every day might be prohibitive and might actually make the test not available for patients. We didn’t want to have that happen, so we made this exception.“

Another change involves a CLIA ‘88 requirement specifying that if nonwaived tests are run by two different methods or instruments within one lab, the lab needs to check those methods against one another every six months.” They don’t have to be the same, but the lab has to know the relationship between the methods and the results you would get,” Dr. Sarewitz says. “We had in the checklist that this applied only to quantitative testing, but in response to a CMS comment, it now applies to qualitative tests as well.”

Also worth noting is an added checklist requirement that specimen containers delivered to the laboratory from the outside include two patient identifiers. “This has been the standard in blood banking for years,” Dr. Sarewitz says, and it reduces the already small but real probability of specimen mixup. The committee felt it was appropriate to apply this requirement to all specimens.

And the requirement that the laboratory notify the CAP of an investigation by a government entity now includes the following addition: “This notification must include any complaint investigations conducted or warning letters issued by any oversight agency (i.e. CMS, State Department of Health, The Joint Commission, FDA, OSHA).”

Collectively, the checklist revisions reflect one of the consistent strengths of the CAP’s Laboratory Accreditation Program, Dr. Williams says: “We are continually updating the checklist to keep it current with actual pathology practice.”

For full details of the updates, see Accreditation Checklists under Accreditation and Laboratory Improvement at www.cap.org. Later this summer the College will issue a new Standards for Laboratory Accreditation and a new Laboratory Accreditation ­Manual.