Patient misidentification in Papanicolaou tests: a systems-based approach to reducing errors.
Abstract: * Context.--Patient safety is of prime concern in every laboratory. Double labeling of glass slides is performed in many cytology laboratories where handwritten patient information on the frosted portion of the glass is overlaid with paper labels (sometimes containing bar codes). The cytotechnologists match this information by turning slides over. We use SurePath liquid-based cytology for Papanicolaou tests in our laboratory and noticed patient misidentification because of slide labeling errors, a problem that has not been addressed in the literature.

Objective.--To reduce slide labeling errors without increasing costs, using a systems-based approach.

Design.--All errors from mislabeled slides during November 2006 were documented, and an informal rootcause analysis was performed. Slides were labeled on opposite ends and monitored for errors.

Results.--Labeling on different ends of the glass slide reduced our error rate from 0.59% to 0%, made visual matching easy, and did not alter costs.

Conclusions.--The practice of overlaying handwritten information with printed labels for liquid-based Papanicolaou tests should be strongly discouraged and replaced with placing patient information in separate portions of the glass slide so that crucial patient identification is not hidden and visual matching is easy.

(Arch Pathol Lab Med. 2009;133:1297-1300)
Article Type: Report
Subject: Pap test (Methods)
Pap test (Equipment and supplies)
Pap test (Standards)
Stains and staining (Microscopy) (Labeling)
Pathological laboratories (Management)
Patients (Care and treatment)
Patients (Management)
Patients (Standards)
Authors: Meyer, Erin
Underwood, R. Shawn
Padmanabhan, Vijayalakshmi
Pub Date: 08/01/2009
Publication: Name: Archives of Pathology & Laboratory Medicine Publisher: College of American Pathologists Audience: Academic; Professional Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2009 College of American Pathologists ISSN: 1543-2165
Issue: Date: August, 2009 Source Volume: 133 Source Issue: 8
Topic: Event Code: 440 Facilities & equipment; 350 Product standards, safety, & recalls; 200 Management dynamics Computer Subject: Company business management
Geographic: Geographic Scope: United States Geographic Code: 1USA United States
Accession Number: 230247121
Full Text: Accurate specimen identification is crucial to patient safety and care in every laboratory. (1-9) Wagar et al, (8) in a longitudinal analysis of specimen identification errors in the clinical laboratory, found a total of 16 632 specimen errors because of mislabeled specimens (1%), requisition mismatches (6.3%), and unlabelled specimens (4.6%). We did not find any data on specimen misidentification related specifically to slide mislabeling in the cytology laboratory, which is the focus of this study, in a review of literature.

Ideally, vital patient/specimen information should be printed and remain on the glass slide throughout the laboratory process. This is especially important in today's world of automated screening, where slides can be sent to file without manual rescreening. At our institution we process more than 35 000 Papanicolaou tests per year using SurePath liquid-based Papanicolaou tests, and more than 90% of these tests are screened by TriPath's FocalPoint slide profiler, a US Food and Drug Administration-approved device for primary screening on both conventional and SurePath liquid-based preparations. Many histology and surgical pathology laboratories (ours included) have gone on to using direct label printers on the slides or bar-code systems that are integrated with the laboratory information system. In some cytology laboratories, however, labeling of slides is still a 2-step process, and double labeling of slides is a routine practice (S. Harrison, CT(ASCP), and K. Schofield, BS, CT(ASCP), unpublished data, December 2007). A laboratory technician first writes patient information (unique identifiers like name and laboratory accession number) by hand on the frosted portion of the glass slide with ink that resists the chemicals during processing and staining. This is followed by superimposition of a printed label with patient information including bar code or other identifiers like patient name, medical record number, and the laboratory accession numbers. The bar code is read by the automated screening device. The cytotechnologist matches the information on the printed label by turning the slide over and reading the handwritten information from the underside of the slide. We noticed a few instances where the handwritten information did not match the patient information on the printed label on the slide when slides came to the cytotechnologist for review. We assumed that the errors resulted from faulty system design. This prospective study is a system/process review with an informal root-cause analysis for process improvement aimed at reducing patient misidentification.

MATERIALS AND METHODS

The process of taking a specimen from its receipt in the laboratory to the time it is reviewed by the pathologist was performed independently by 2 observers (Figure 1). Using the existing slide-labeling technique, whereby the handwritten information was overlaid by printed information (Figure 2, A), cytotechnologists were asked to collect data from on all mislabeled SurePath liquid-based Papanicolaou test slides that came to them during November 2006 for sign-out or 10% quality-control review. This included reporting all cases where the identifiers (patient name, cytology accession number) on the printed label did not match the corresponding identifiers on the handwritten, frosted portion of the glass slide. Mislabeled cases were sent back to the laboratory technician for correction. Under the laboratory quality assurance program, data were collected prospectively and tabulated to include the number of cases mislabeled at any one time (A1-A4, Table), date and day of occurrence, the laboratory technician (LT1 and LT2) who performed that run, and the shift (am or pm) worked by the laboratory technician (Table). Case numbers and other unique patient identifiers were not collected. Figure 2, B, shows the new technique whereby the printed label was placed on the side opposite the handwritten label using Fisherbrand 24 x 30 coverslips. Cytotechnologists were again asked to collect data on mislabeled slides for 1 year, between December 1, 2006, and December 1, 2007. An informal root-cause analysis was performed when the errors were reported regarding the laboratory technician (shifts worked) who was on service, the number of specimens processed that day, and any other factors that may have played a role in the sequence of events leading to mislabeling of slides.

[FIGURE 1 OMITTED]

RESULTS

There were 2 laboratory technicians (LT1 and LT2) who worked the evening and day shifts, respectively. Between October 31, 2006, and November 21, 2006, there were 3076 cases processed, resulting in 3517.5 slides. Of these, 2844 were liquid-based Papanicolaou smears. Seventeen mislabeled cases/errors were identified during this time (0.59%;result A, Table). Between December 1, 2006, and December 1, 2007, following the change in labeling, 34335 liquid-based Papanicolaou smears were processed. No errors were identified (result B, Table). The total number of specimens processed on the days of the occurrence did not differ significantly from other days in the laboratory. We noted that the errors were not restricted to any one technical staff but made by different laboratory technicians who worked different shifts (Table). A detailed description of specimen processing in the cytology laboratory by the laboratory technicians is shown in Figure 1.

[FIGURE 2 OMITTED]

An informal root-cause analysis was performed soon after each error was reported. Slide-labeling was done as a batch process. Distraction during labeling in the form of conversations and overhead paging systems were noted. A few times, the process was interrupted, when the timer for the nongynecologic specimen stainer went off, and was resumed later. One laboratory technician suggested increasing the font size on labels to make visual matching easier.

COMMENT

The Institute of Medicine report To Err Is Human: Building a Safer Health System1 has brought the issue of patient safety to the forefront. Extensive databases describing error rates in pathology were developed and maintained by the College of American Pathologists and include the Q-Probes and Q-Tracks, which provide information on error rates from more than 130 interlaboratory studies. There are many areas for potential misidentification of a specimen when we track it through the process of collection to reporting out a case (Figure 1). Mix-ups may also occur when results are presented to the patient. This study focused only on the slide-labeling process. The laboratory technician is supposed to match the printed label before it is placed on the slide, overlaying the handwritten label. This is done as a batch process, where the slides are lined up, and the labels are pasted on. Distractions in the work area, including noise, interruptions, timers, overhead pages, and the like, were noted. Although the environment was the same after we instituted the change in labeling process, the error rate came down to 0, suggesting that not covering handwritten patient information and placing the printed label on the opposite end of the glass slide is a more active process, requiring more concentration, and is less prone to errors. This process facilitates checking that the 2 labels match by multiple personnel, such as the laboratory technician and the cytotechnologist or the cytopathologist.

Sirota (7) reviewed pathology errors in detail. Errors such as the one addressed in this study are considered slips or lapses, whereby the wrong automatic action is used to achieve a goal. Because these occur in situations that are not perceived as dangerous or problem prone, these errors often elude most quality assurance and oversight efforts. Cytotechnologists are trained to turn slides over to ensure that the patient identification matches. This grossly inefficient system assumes that the cytotechnologist is adept at reading mirror images of the handwritten information. It is also prone to serious patient misidentification errors that may never come to light if the slides go directly to file from TriPath's FocalPoint slide profiler and are not reviewed by the cytotechnologist. Discovering that the slides have been mislabeled also wastes significant time, and sometimes, the specimens have to be reprocessed. Ideally, we aimed for a labeling device (thermal labeling) that would label the slides once and would contain the bar code necessary for the TriPath's FocalPoint slide profiler. Integration of various technologies did not work. Using smaller coverslips (Fisherbrand 24 x 30) gave us enough room on the slide to label 2 different portions without the printed label touching the coverslip and the mounting medium. It is also now mandatory for the laboratory technician to visually check the labels before they put the slides on the TriPath's FocalPoint slide profiler. In this series, the error rate using the traditional labeling method was 0.5%, which was reduced to 0. Although errors in slide labeling are not well recognized in the literature, this study shows that it is a potential source for patient misidentification. Using a systems-based approach to process improvement, instead of an individual approach of blaming any one laboratory technician, can help resolve this issue, which is critical to patient safety.

In summary, the traditional system of overlaying handwritten information with the printed label in a high-volume area like the Papanicolaou test should be replaced by a more transparent system of placing both labels on 2 ends of the slide, so that crucial information is not hidden and visual matching is easy.

References

(1.) Kohn LT, Corrigan JM, Donaldson MS, eds. To Err Is Human: Building a Safer Health System. Washington, DC: National Academy Press; 1999.

(2.) Zarbo RJ. Monitoring anatomic pathology practice through quality assurance measures. Clin Lab Med. 1999;19:713-742.

(3.) Raab SS. Improving patient safety by examining pathology errors. Clin Lab Med. 2004;24:849-863.

(4.) Valenstein PN, Sirota RL. Identification errors in pathology and laboratory medicine. Clin Lab Med. 2004;24:979-996.

(5.) Condel JL, Sharbaugh DT, Raab SS. Error-free pathology: applying lean production methods to anatomic pathology. Clin Lab Med. 2004;24:865-899.

(6.) Howanitz PJ. Errors in laboratory medicine: practical lessons to improve patient safety. Arch Pathol Lab Med. 2005;129:1252-1261.

(7.) Sirota RL. Error and error reduction in pathology. Arch Pathol Lab Med. 2005;129:1228-1233.

(8.) Wagar EA, Tamashiro L, Yasin B, Hilborne L, Bruckner DA. Patient safety in the clinical laboratory: a longitudinal analysis of specimen identification errors. Arch Pathol Lab Med. 2006;130:1662-1668.

(9.) Quillen K, Murphy K. Quality improvement to decrease specimen mislabeling in transfusion medicine. Arch Pathol Lab Med. 2006;130:1196-1198.

Erin Meyer, DO; R. Shawn Underwood, CT(ASCP); Vijayalakshmi Padmanabhan, MBBS, MD

Accepted for publication October 16, 2008.

From the Department of Pathology, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire.

The authors have no relevant financial interest in the products or companies described in this article.

Presented in part at the annual meeting of the American Society of Cytology, Houston, Texas, November 2007.

Reprints: Vijayalakshmi Padmanabhan, MBBS, MD, Department of Pathology, Dartmouth Hitchcock Medical Center, One Medical Center Dr, Lebanon, NH 03756 (e-mail: viju@hitchcock.org).
Occurrence of Labeling Errors (a)

                                          No. of     Technician
 Occur-                                 Cases With    Involved
 rence                                   Labeling     (Shift,
of Error   Date of Occurrence             Error       AM, PM)

  A1       October 31, 2006, Tuesday        2         LT1 (PM)
  A2       November 9, 2006, Thursday       5         LT1 (PM)
  A3       November 13, 2006, Monday        2         LT1 (PM)
  A4       November 21, 2006, Tuesday       8         LT2 (AM)
  B        December 1, 2006                 0

Abbreviation: LT, laboratory technician.

(a) A1 through A4, Bar-coded printed labels were overlaid on the
handwritten portion of the glass slides. B, Bar-coded printed
label was placed opposite the handwritten portion of the glass
slide.
Gale Copyright: Copyright 2009 Gale, Cengage Learning. All rights reserved.