When MTs mind the Paps: QA for send-out tests.

When MTs mind the Paps: QA for send-out tests

I would like to relate to you the dramatic results of an 18-month study conducted by medical technologists who decided it was time to "mind the Paps." The results of our intervention convinced us that it's time medical technologists moved beyond tradition and did more with Pap slides than pass them along to the cytology laboratory. Five thousand slides for the better, we are proud of our accomplishments.

We all know that the heat is on cytologists and pathologists. False-negative Pap reports have triggered an intense scrutiny of cytology laboratories and the qualifications of those who screen Pap smears. Strict government standards were implemented with CLIA '67 on Sept. 10, 1990. Unfortunately, even these will not wholly assure accuracy. Current literature states that cytology lab errors account for just 50% of false-negative reports. The other 50% result from specimen compromise within our clinics! [1]

At the primary-care facility where I work, 250 to 350 Pap smear slides are collected each month by non-ob/gyn providers and sent to a reference laboratory for screening. Each test is logged and dispatched by technologists on our staff. We handle the slides, see the requisitions, and receive the reports. The providers follow up on abnormal results. But what about false negatives? Were we sending out smears of high quality? Were our Pap tests 50% compromised? In the fall of 1988, I targeted Paps for a QA review.

* Setting the stage. To do first things first, I sought the support of our medical director and then my staff. There was no way I could begin a targeted review without their cooperation, assistance, and ideas. For clinical guidance, I turned to Novak's "Textbook of Gynecology," a highly respected resource. [2]

Novak's states that accuracy in cervical cytology depends on two factors: the technique used to obtain the smear and the quality of the laboratory that will read and interpret the slides. Methods for reducing potential sources of error include scraping the squamocolumnar junction properly, obtaining a sufficient endocervical sample, making a smear that is thin but contains enough material for assessment, fixing the slide quickly to prevent air drying artifact, and selecting a good cytology laboratory. Regardless of the sampling procedure used, according to Novak's, the cytopathologist must be informed of any unusual clinical finding or history.

I chose to focus our QA review on aspects of the collection technique that we could easily monitor from the cytology reports we received daily. To facilitate data recovery, I designed a simple log book that has become an invaluable source of statistics. It reveals not only specimen adequacy but also report turnaround time, unsatisfactory and abnormal smears, and the name of the provider who is doing the Paps. The book contains a computer-generated patient label and four columns of information (Figure I). The last column contains the initials of the technologist who has received the report and will assume responsibility for logging and charting.

* The review begins. Initially, we used the presence of endocervical cells as evidence of smear adequacy. The endocervical component was noted to be present or absent on each Pap report that we received. When we received a report, placing a check under the "normal" column in our Pap log indicated that there was no sign of malignancy and that an endocervical component was present. Writing "EC" under "Other" indicated that no endocervical component had been recovered.

In just a few weeks, it became apparent that our clinic had a poor endocervical recovery rate. Our medical director took action immediately. After holding a meeting with staff physicians and a physician's assistant, she suggested that we begin a trial with endocervical brushes. New clinic policy would dictate that patients whose smears revealed no endocervical cells but who had not undergone a hysterectomy would be recalled for a repeat Pap. Suddenly everyone's attention was riveted on collection technique.

Investigators have shown that when the presence of endocervical cells is used as a criterion, up to 30% of all gynecology specimens sent to cytology laboratories may be considered less than optimal. [3] Other studies have documented that endocervical cell recovery was enhanced when an endocervical brush was used to collect a specimen as contrasted with the method we had been using, the Ayre spatula and cotton swab.

In one study of 5,716 slides obtained by spatula and brush combined, 98% contained endocervical cells, while only 84% of 24,496 slides obtained by spatula alone showed endocervical cells to be present. [4] We hoped that we, too, would see improvement in our trial with the brushes.

Initially, only one provider was asked to use the brush for obtaining specimens. Her recovery rate increased by 15% in one month. Based on her positive evaluation, an in-service presentation was conducted and our approximately 15 full and part-time providers then began to use the brushes. Within four weeks, the clinic's overall endocervical recovery rate had improved markedly. Recovery jumped to 92% within eight weeks and contniued at that rate or higher ever since.

* Expanding scope of review.

With the right tool in hand and our hands feeling more secure in their technique, it was time to review other indicators of smear quality. Rarely did we see a report pointing to the presence of air drying artifact, yet each month three or four of our reports stated that excessive blood or scanty epithelial material had made one of the slides we had submitted unsatisfactory. We noted such findings under the "Other" column in our Pap log book. Then we let the numbers talk to us.

A quarterly review established that on average, zero to one unsatisfactory specimen per provider could be expected in a three-month period. When a new provider on our staff was noted to have submitted 17 unsatisfactory slides in a single quarter, we alerted our medical director. She worked with the new staff member to improve her collection technique. This is what QA is all about. We were making a difference!

Meanwhile, the lab staff decided to turn their attention to Pap requisitions. Were we giving the cytologist and pathologist all the information they needed to evaluate the smears that we were sending out?

We noticed that prior surgical history had been omitted on a few of the reports we had received. On some others, the exam date was missing or the name of the requesting physician was mis-spelled. Occasionally a patient's name or ID number would contain an error. We decided to verify the information on all Pap requests before sending them. We believed that our diligence early in the process would improve the quality of the report we would ultimately receive.

I sent a memo to all our providers with a sample Pap requisition form attached. I highlighted the areas on the form that they would be asked to fill in. Whenever we received an incomplete form after that, we returned it to the provider--assuming we could catch him or her--with a reminder that it was important to provide all pertinent data. If the doctor was running too fast for us, the lab staff would pull patient charts to obtain missing data or print clearly over an illegible entry. Our efforts paid off when patient information was correct on virtually every report.

To enhance legibility, we asked all providers to use their signature stamps rather than signing the forms. When logging the order into the computer, we verified the patient's name and ID number on both the form and the slide.

Although the providers were not 100% compliant with our request to fill in all the blanks on every Pap request, the improvement was noticeable. The clinic nurses assumed responsibility for filling in some portions of the request at the time of patient registration.

Then it happened: Our cytology reference lab announced an unreasonable increase in the screening cost per slide. A corporate decision was made to use a different reference lab.

* Focus of review changes.

When we switched cytology reference labs, it seemed appropriate to change the focus of our review. We would no longer concentrate on aspects of collection technique. Now we would monitor the reports generated by the new lab as we learned to deal with a different reporting systems and format.

Fortunately, our clinic staff understood the motivating factor behind this corporate decision. While we all regretted the need for change, we accepted it. Unfortunately, the transition proved to be less smooth than I had hoped. The new lab had a lousy reporting system.

Historically there have been several methods of reporting cervical cytology. These include the Papanicolaou class system, a World Health Organization system, a cervical intraepithelial neoplasia system, and other methods that combine any or all of these or represent variants of them. As a result, many Pap reports are ambiguous and confusing. Frankly, the reports from our new lab fell into that category.

One example: We frequently found the entry "Findings: none" on a Pap report. Our physicians assumed this meant the smear had been adequate and normal--but such was not the case. I called to inquire and learned that the designation "Findings: none" merely meant that no endocervical cells had been seen. This, combined with the reference lab's failure to note a surgical history or previous Pap abnormality on the reports, made it impossible to interpret test findings without reviewing the patient's chart.

Our medical director wrote the new lab a letter communicating our concerns regarding the ambiguity of the reports. We had a favorable response in the spring of 1990 when the reference lab "went Bethesda."

* The Bethesda system. At a workshop sponsored by the National Cancer Institute (NCI) in December 1988, participants sought to develop a uniform cytopathology reproting system that would eliminate confusion in Pap reports while facilitating research. Cytopathologists and cytotechnologists in attendane addressed the format of the Pap report as well as the categories and descriptive diagnoses typically used in Pap result reporting. (4) From this group emerged a simplified system of "communicating cytopathology findings to the referring physician in unambiguous diagnostic terms that have clinical relevance." (5) This system is named for Bethesda, Md., where the NCI workshop was held.

In May 1990, the Bethesda system of classification was formally adopted by the NCI, the College of American Pathologists, and the American College of Obstetricians and Gynecologists. The Department of Health and Human Services now requires the roughly 12,000 clinical laboratories that receive Medicare and Medicaid reimbursement to use the procedure.

When our reference laboratory began using the Bethesda system of reporting, the implications immediately became evident. Each Pap report became, in effect, a complete medical consultation (Figure II). We needed no hospital chart to scan a patient's surgical history, nor did we have to read between the lines to determine whether the specimen submitted for analysis was adequate. Statements on the report that facilitate review include "satisfactory for interpretation," "less than optimal," and "unsatisfactory." Being able to log each report as "within normal limts" or "other," and to enter a more detailed descriptive diagnosis and recommendations, takes the guesswork out of follow-up.

Descriptive diagnoses are simple and clear. Gone are the confusing classes of Pap results that we had to deal with before. Terminology says it all. We no longer have to keep a guide to interpretation in the lab for our clinic's physicians. The Bethesda system represents a substantial advance in cervical cytology reporting. Combined with improved sampling technique, the system has greatly enhanced the quality and reliability of the Pap smear.

* Relevance factor. When I targeted the Pap test for review, I wanted more than good reports; I wanted better results for our patinets. Effective collection technique and accurate screening will lead to an acceptable report. A good result--high-quality health care and early intervention when needed--is achieved when appropriate follow-up care is assured and given.

Our laboratory staff has developed a system to monitor the follow-up of every abnormal test. For Paps, this process begins with a review of the Pap log to insure that all Pap reports are received in good time.

Turnaround time is an essential component of a QA program. Our Pap TAT averaged 14 to 18 days until September 1990 and now averages 24 to 28 days. This delay is due to a new Federal law that limits cytotechnologists to reading 120 slides in 24 hours. Because 40 of these may be for quality control purposes only, however, the law effectively restricts them to 80 patient slides per day. Experienced CTs used to read many more. At this point, some reference labs have a Pap turnaround time as long as three months.

Any abnormal Pap report is listed as "pending" in our computer until the patient has been contacted and the communication documented. The lab pulls the charts of patients whose findings are not normal. Such charts are delivered directly to the provider, who assumes responsibility for appropriate follow-up care. Women referred for colposcopy receive a copy of their Pap report along with an informative booklet about the Pap test. The booklet discusses possible findings and types of exam or treatment that may be indicated. Patients whose findings are normal receive notification cards that they addressed to themselves when visiting our clinic.

* Helpful tips. Worth mentioning are a few tips and observations I noted during our review. Medical technologists must interact with the clinic staff an d sustain a positive, cooperative attitude. It is a mistake to rely on the reference laboratory to keep a lab's records, which should instead be carefully maintained as part of the in-house QA program. Comparing the two can produce surprising results.

Reading the Pap reports closely may reveal suspicious findings. Trichomonas found in an asymptomatic clinically normal 60-year-old woman or endocervical cells found in a specimen taken from a woman who has had a hysterectomy suggest inaccuracy.

In reviewing statistics from our reference lab on their endocervical recovery rate for the specimens we sent them, I learned that they were including all slides, even those of patients who had undergone hysterectomies and no longer had such cells. These patients' results should not be included in the recovery rate data.

Technologists should never hesitate to pick up the phone and ask for a rescreen or to borrow a slide for a pathologist to review. That's called participating in patient care.

It's wise to discuss trends with the medical director. Just as important is to put any concerns in writing and document them.

Asking many questions is crucial when shopping for a reference lab; the choice is at least as significant a decision as one to purchase an expensive piece of laboratory equipment and deserves ample time and thought.

Everyone in our clinic has benefited from the technologists' ongoing review of Pap tests. We have all learned a great deal and feel that we have dramatically improved the quality of every Pap we collect. Repeat Paps are fewer, resulting in a more cost-efficient medical practice. The time of our providers and nurses is better utilized.

Patients' comment cards reflect satisfaction. They appreciate our policy of open communication. Part of our quality assurance effort includes reviewing such cards and responding to their concerns in an appropriate manner.

When medical technologists mind the Paps...and we've only just begun!

(1) "Cervical Cytology: Evaluation and Management of Abnormalities." Technical Bulletin No. 81, Washington, D.C., American College of Obstetricians and Gynecologists, October 1984.

(2) Jones, H.; Wentz, A.; and Burnett, L. "Novak's Textbook of Gynecology," 11th ed., chap. 26. Baltimore, William & Wilkins, 1988.

(3) AMA Council on Scientific Affairs. Quality assurance in cervical cytology: The Papanicolaou smear. JAMA 262: 12, 1989.

(4) "The Proposed 'Besthesda System' for Reporting Cervical/Vaginal Cytologic Diagnoses." ASCP News, pp. 6-7, September 1989.

(5) The 1988 Bethesda System for reporting cervical/vaginal cytological diagnosis. JAMA 262: 7, 1989.