Defining Error in Anatomic Pathology

Archives of Pathology & Laboratory Medicine,  May 2006  by Sirota, Ronald L

• E-mail
• Print
• Link

Context.-Although much has been said and written about medical error and about error in pathology since the publication of the Institute of Medicine's report on medical error in 1999, precise definitions of what constitutes error in anatomic pathology do not exist for the specialty. Without better definitions, it is impossible to accurately judge errors in pathology. The lack of standardized definitions has implications for patient care and for the legal judgment of malpractice.

Objectives.-To review the goals of anatomic pathology, to discuss the problems inherent in applying these goals to the judgment of error in pathology, to offer definitions of major and minor errors in pathology, and to discuss error in anatomic pathology in relation to the classic laboratory test cycle.

Data Sources.-Existing literature.

Conclusion.-Definitions for major and minor error in anatomic pathology are proffered, and anatomic pathology error is characterized in the classic test cycle.

(Arch Pathol Lab Med. 2006;130:604-606)

Due in large part to the Institute of Medicine's 1999 report on medical error,1 there is intense interest in reducing medical errors in all disciplines of medicine in the United States. The report broadly mandated a 50% reduction in medical errors across all specialties in the 5 years after its publication. Although the report generally defined error as a failure of a planned action to be completed as intended (an execution error) or the use of the wrong plan to achieve an aim (a planning failure), it did not refine the concept further for each specialty, leaving to each discipline the task of defining and quantifying those types of errors that are important or idiosyncratic to the specialty. The definition of error in each specialty is not only important from the point of view of patient care but also is legally important, because for the legal profession error is usually equated with a deviation from generally recognized standards of care. The objectives of this article are to discuss what is meant by error in pathology, to suggest definitions that may be useful to the specialty of anatomic pathology, and to discuss where errors in anatomic pathology occur in relation to the classic laboratory test cycle.

ERROR IN RELATION TO THE GOALS OF ANATOMIC PATHOLOGY

To understand the concept of error in anatomic pathology, it is important to comprehend the goals of anatomic pathology. Simply stated, the goals of anatomic pathology are to render a correct and complete diagnosis to the correct patient in a timely fashion in a way that is understandable and useful to the physician treating the patient.

When one analyzes the goals carefully, it becomes clear how judgmental the assignment of the term error is. How are the terms correct and complete diagnosis defined? What is meant by timely? Without definition and agreement on what these terms mean, it is unlikely that reproducible estimates of pathology error can be made.

THE CORRECT DIAGNOSIS

Ideally, a correct result would be an accurate result. Accuracy reflects the truth, based on scientifically validated gold standards. Unfortunately, few scientifically validated gold standards exist in anatomic pathology. Morphologic analysis, the cornerstone of anatomic pathology, cannot be considered a gold standard, because it is subjective and based on observer experience. Although some observers have more experience than others or are more competent than others, all observers are human and are subject to the problems of human fallibility that are well known to cognitive psychologists and to those who study error.2-4 Immunohistochemistry has similar problems and is subject to other reproducibility pitfalls stemming from the technique. Cytogenetic or fluorescence in situ hybridization results can serve as gold standards for certain disorders (eg, the classification of acute myeloid leukemia and certain lymphomas); however, cytogenetic criteria do not exist for most disorders encountered in routine surgical pathology. The situation is similar for molecular studies.

In lieu of accuracy, a precise diagnosis usually serves as the correct diagnosis or result in pathology. Precision is a measurement of variation. In the clinical laboratory, precision is measured and expressed statistically. In anatomic pathology, precision usually implies that most competent pathologists will render a similar result (ie, the correct result) when observing or analyzing the same specimen. In anatomic pathology, low interobserver variation usually serves as a surrogate for accuracy.

While low interobserver variation is a goal of anatomic diagnosis, interobserver variation is a fact of life in anatomic pathology, and much of it does not affect patient safety or cause patient harm. For example, although most pathologists try to subclassify most malignant diseases, subclassification of many diseases does not alter therapy. A case in point is that, while most pathologists attempt to classify invasive breast carcinoma as lobular or as ductal, treatment decisions are rarely made on the basis of this subclassification and are based mostly on the stage of the tumor. Therefore, variation in the subtyping of this disease causes little patient harm.