Errors : Detection and minimization in histopathology laboratories

The histopathological diagnosis plays a major role in the treatment of diseases. Errors in these reports affect patient care. Hence, it is of utmost importance for all practitioners of this specialty to be aware of possible errors in histopathology laboratories and the means to minimize them. As with other disciplines of laboratory medicine, errors can occur in the pre-analytical, analytical and post analytical phase. The concept of quality and its control should be applied to all phases to curb errors. Audit can be used as a tool to generate information about the background level of errors in pathology which in turn can be used to reduce and avoid errors in histopathology laboratory. Furthermore, accreditation is a means to ensure patient safety and best quality assurance.

such as lack of objective numerical data, descriptive nature of reports, subjectivity, individual judgment and bias, non uniformity of reporting patterns etc make assessment and implementation of quality control more difficult in histopathology.
To document errors in histopathology laboratory, approach to error investigation and documentation using a validated tool has now been described. 15(table 1)

Pre-analytical phase 1. Defective specimen includes:
Lost specimen, inadequate volume, size, gross description, erroneous measurement or extraneous tissue.
Patient ancillary diagnostic study not initially done.Primary meaning positive/negative or benign/malignant.Secondary meaning grade, stage, margins etc.

Post analytical phase
Erroneous/missing non diagnostic information.

Report delivery.
Computer/format, transmission, upload error.

CATEGORIES OF ERRORS
Literature has described four general types of errors, with three subtypes in the category of defective interpretation. 15 the first major category of defective interpretation, the first subtype of error is false negative, second subtype false positive and third is misclassification e.g there is neither an undercall nor an overcall when a pathologist incorrectly labels an entity in the proper category of disease (e.g fibrosarcoma rather than malignant fibrous histiocytoma).The alternative designation alters neither the diagnostic primary classification (eg.malignancy) nor secondary diagnostic features (e.g high grade, negative margin) among the features mentioned in the report.
The second major category of error is constituted by defective identification of patient, tissue or laterality.These errors can take place at any step in the process of diagnosis.However, they are more common in the pre-analytical phase.This category encompasses misclassification of patients, origin of tissue, sample (e.g stomach vs colon), anatomic location (e.g ascending vs descending colon) or laterality of tissue (e.g right vs left breast).
The third category consists of specimen defects including lost specimens, inadequate volume and size occurring in the pre-analytical phase and inadequate gross description or measurement, extraneous tissue or inadequate sampling occurring in the pre-analytical phase.Pre-analytical phase specimen defects also includes specimens whose representativeness is inadequate or suboptimal at the tissue level, block or slide level because of an action or inaction taken or not taken in the surgical pathology laboratory.Failure to perform pertinent ancillary studies that would initially reveal a correct diagnosis is also a subtype of preanalytical specimen defect.
Fourth major category of error is defective reporting.This includes reports with erroneous or missing non diagnostic information (e.g clinician's name, date of procedure etc), dictation or typing error, report format or upload errors which can arise while using computers.Defective report error usually occurs in the post analytical phase, although missing or incorrect information may have been there in the pre-analytic phase without previously been addressed by anyone until the report's preparation and dispatch.Documentation of the type of change made after an error was discovered should be made.Amendment options include changes in: a) Primary diagnostic characters (e.g negative to positive, benign to malignant, inadequate to adequate).No impact indicates erroneous message not transmitted or message transmitted but ignored.Minimal harm means delay in diagnosis, unnecessary non invasive further diagnostic effort (e.g blood, radiograph, CT), delay in therapy and unnecessary therapy.Minor morbidity is defined as effects and events that can be demonstrated objectively (e.g fever, thrombocytopenia, swelling etc) but do not require hospitalization or surgical intervention.Moderate morbidity includes effects and events that require hospitalization or surgical intervention but not major morbidity defined as loss of organ or function of an organ system (e.g arm/limb, eye/sight, ear/hearing, speech or uterus of a woman of reproductive age). 15l process involved from the submission of specimen to preparation of slide is grouped under pre-analytical phase.Newer models for pre-analytical phase also include aspects like patient satisfaction, the collection process and professional staff satisfaction with arrangement made by the laboratory towards sample collection and transportation etc. 16 Studies have implied that most of the errors in the laboratory are related to pre-analytical phase. 17Same can be said of histopathology as well. 18Documented instructions containing relevant information should available at all points of specimen collection with the laboratory.Correct patients identification by a unique accession number traceable to the specimen and report all throughout the process is of chief importance.Error in this area is common but avoidable.
Similarly, wrong identification of anatomic location or laterality is an error that should be avoided.For this, it is worthwhile for the laboratory to design its own "referral form" for histopathology and immunohistochemistry if available and circulate it to areas of sample collection. 19This form should provide adequate space for relevant clinical data.When clinical data is not provided or not adequate, the laboratory should take imitative to obtain the pertinent data either from the treating physician or hospital files.Further to eliminate the errors in pre-analytical phase standard procedure for sample accession, identification, acceptance/ rejection, gross examination and all other following steps must be documented.The standard operating procedure should be maintained and be made available at workplace.Planned changing of chemical used for processing should be based on tissues passed through.Once predetermined limit is reached, compulsory change should be done.This prevents under processing, unnecessary rework and loss of tissue.Same applies to deparaffinization, staining, dehydrating and cleaning steps of section preparation.

For hematoxylin and eosin, a tissue containing a mixture
Errors : Detection and minimization in histopathology laboratories of hematoxyphilic and eosinophilic tissue (cervix, fibroadenoma ) is to be used as controls.Multiple sections may be cut and stored to be used as controls later.Using same controls avoids variation related to tissue type.Control slide staining should be done before the routine batch staining is done and the staining character should be compared with that of the previous day.A record of staining character should be maintained.
Microtome should be serviced regularly with periodic calibration of micrometer to maintain uniform section thickness.Use of disposable blades is recommended.Lastly, care should be taken not to introduce artifact during any phase of tissue processing and slide preparation.
In the post analytical aspect, report generation without transcription error, report transmission/dispatch to the correct person, storage of report material as well as reported data and safe disposal of specimen is looked into.Billing issues, patient safety issues, turnaround time (TAT) and general customer satisfaction (wait time) have been included in the post analytical phase.16 It is of vital importance to monitor TAT and laboratories should try to achieve the goal of signing out the majority of the specimen within 48hrs of receipt of specimen. 19e diagnostic standard in histopathology laboratory can be maintained and improved by -external quality assessment schemes (EQA).
-clinical audit -laboratory accreditation -continuting medical education (CME) -clinicopathological case review meetings. 20ese processes are related to each other, for example, feedback from EQA provides opportunity for continuous medical education and participation in EQA schemes enables compliance with accreditation standard.Diagnostic external quality assessment (EQA) schemes consist of circulating a "test" material to the participant.The "test" material is a histological section with relevant clinical information.Diagnosis and comments are returned to the organizer of the scheme and feedback of the performance is provided to the participant.EQA is an important educational means in histopathology.It has two components, one: viewing the material will be educational, two: quality assessment requires quantitative feedback to the participant which has educational value as it can provide unambiguous information on areas where continuous medical education (CME) is required and it can confirm the effectiveness of that education. 20e levels of error in diagnostic laboratory can be monitored by audit.Audit involves asking questions and collecting data about selected aspects of one's current practice. 213][24][25] Precise technical standards for diagnostic pathology laboratory are harder to define, though some of the technical processes can be subjected to internal and /or external audit.Audit can assess laboratory speed, overall staining quality and work load for both the laboratory and individual pathologist.In 1991 the American Association of Directors of Anatomic and Surgical Pathology produced a list of recommended types of departmental audit (table 2). 26udies have shown a lack of consensus amongst pathologist for a range of specimen type and have documented that a same pathologist can produce different reports when examining the same specimen on different occasion. 27report on 2046 cases of colonic cancer examined by 22 histopathologist showed considerable observer variation in histological grading, Dukes' staging and the number of lymphnodes involved. 28Similar lack of consensus has been noted in the diagnosis of molar pregnancy 29 and melanocytic lesion. 30[33][34] The baseline and background level of erroneous diagnosis have been examined in several audit studies.In one US study, where an extra pathologist, was specifically appointed to review all cases over a 1-year period, involving 5397 cases, 14 discrepancies (0.26%) of potential clinical significance was detected. 35 a similar audit employing second pathologist, major errors was identified in 1.2% of 2694 cases. 36When a second checker pathologist was involved the discrepancies were divided into oversight errors where the pathologist had missed significant pathology and misinterpretation error where pathological changes had been incorrectly interpreted.57% of the errors were oversight errors and 43% were misinterpretation errors in this study. 35Similarly in a 5 year audit at Southampton of 45 errors, 65% were due to misinterpretation, 31% due to pathological oversight and 4% due to failure to answer a specific clinical query.Almost all the errors in the reports were readily detected and corrected when it was brought to the notice of the reporting pathologist.Discussion with reporting pathologist suggested both oversight error and misinterpretation error occurred while reporting a large batch of surgical specimen. 37In other studies majority of pathologist indicated errors related to excessive workload. 38er review is one of the commonly used methods of audit.The use of an extra pathologist for dual reporting provides continuous peer review but requires significant resources.Hence, random case audits are used to monitor errors.Random case audits a sample size of 2% to 4%. 37,39 The other method of detecting error in pathology that is carried out in almost all pathology departments is the clinicopathological meeting.An audit of cases at clinicopathological meeting is relatively easy to establish and has a low resource requirement. 40creditation is an enabler of quality; it is patient focused, impartial and objective, and ensures an upto date technologies and it procedures that reflect current best practice.Hence, accreditation should also be implemented in histopathology laboratory to improve quality of service and to provide patient satisfaction.

CONCLUSION
Medicine is currently being challenged by society to improve patient safety and to significantly minimize medical errors.As surgical pathology plays an integral part in patient care it is essential for all practitioners of this specialty to detect and modify processes where possible to meet this growing need.
b) Secondary diagnostic characteristics.c) Diagnostic reclassification.d) Patient or specimen re-identification.e) Report of additional specimen sampling that resulted in changed report.f) Other edits that do not change primary or secondary diagnostic information, patient or specimen identification or involve specimen characteristics.Timing of discovery is categorized into those errors detected before sign out and those after sign out.For those changes detected before sign out, mechanisms of discovery are: a) Additional information or material.b) Intradepartmental review sign-out or double read of the current case.c) Preparation or presentation at a conference or at review with a clinician d) External consultation.For revision after sign out the mechanisms are: a) The responsible pathologist's review of a recent case without additional information or material.b) The responsible pathologist's review of a recent case with addition information or material but without clinician prompting.c) At preparation or presentation at conference with clinicians (e.g tumor board).d) Clinician initiated review or reconsideration of a case.e) As a result of external consultation.A part of error classification attempts to standardize assessment of outcomes related to anatomic pathology error. 15Taxonomy of outcome types divide consequence into : a) No impact on care.b) Impact on care with minimal harm.c) Minor harm d) Moderate harm and e) Major harm.