Common Errors in Pathology Reporting

The pathologist is in the sometimes uncomfortable position of being expected to make a ‘definitive’ diagnosis. While clinicians are often seen as undertaking an ‘art’ of interpretation and assumption, the pathologist is perceived to employ a range of tests and stains to arrive at a ‘real’ diagnosis. However, pathological reporting is rarely that simple, with much relying on the interpretation of appearances down the microscope. Therefore, just like any other doctor, pathologists are at risk of making mistakes.

Published reports tell us that error is uncommon in pathology reporting, occurring in only around 1% of cases [1]. However, when errors do happen, not only can they impact on patient care, they can also result in expensive legal claims. Clearly, reducing errors is in everyone’s best interests.

How to Define Error

microscope-385364_1280 (1)Error in pathology reporting can be difficult to define. Everything from differences in interpretation or semantics, omissions, to problems leading to a change in diagnosis and /or treatment might in different circumstances be considered errors. We can also think about whether a mistake is noted prospectively or retrospectively, and indeed whether it results in any changes to patient management or is merely an academic difference. In the current climate of litigation against medical teams, we can of course also look at medicolegal claims and use these as a surrogate measure of significant errors in pathology reporting.

Causes of error

Pathologists are highly trained professionals, often with years of experience, surrounded by skilled technicians working to reach the correct diagnosis [2]. So how do mistakes happen? Industry derived error-identification systems allow scrutiny of all stages of the specimen management process and help to identify where errors are occurring [3]. Such scrutiny has identified errors in pre-analytic (poor specimen labelling, delayed delivery to the laboratory, poor clinical information), analytic (incorrectly applied tests, errors in interpretation) and post-analytic (incorrect patient identification on the report, errors in interpretation of the report findings) stages [4].

For the individual pathologist, the key element is ensuring that training and continuing professional development (CPD) enable accurate interpretation of the sample and associated tests. Workplace error analysis identifies that problems can occur due to lack of knowledge, distraction during work and bias during ‘rapid thinking’ analysis [5]. During learning people typically employ ‘slow thinking’ – the methodical stepwise processing of information to reach a diagnosis; with increasing experience they switch to ‘fast thinking’ – the recognition of patterns which allows some steps to be skipped in order to rapidly reach a conclusion. Although more efficient, occasionally assumptions result in errors.

There have been relatively few published reports into the type and spread of errors in pathological reporting. One paper from the United States examined 335 litigation claims and found that over 57% of cases arose from only five types of specimen. Within these, 68% were failure to diagnose cancer, resulting in delayed or incorrect treatment [6]. It is difficult to say whether these diagnoses really did generate more errors, or whether the emotive and potentially life-changing nature of the disease resulted in a greater number of legal cases.

Ways to reduce medical errors

Medical Blood TestThinking about the causes of error gives us some insight into how to reduce their number. The first and most obvious point is to have robust systems in place within the department or unit to ensure that procedural mistakes are minimised, use of safety implementation systems such as the Lean or 5S models can be effectively employed to reduce error rate [7].

Secondly training, CPD and support within the unit need to be maximised [8]. In fact, units where reports are ‘reviewed’ prior to release exhibit a lower rate of errors. For example, one group uncovered errors in 10% of reports at review [9]. Although reviewing reports inevitably leads to some delay in the report being available to clinical teams, greater accuracy outweighs this in most analyses. Report review can take a variety of forms – including review by the author, a second pathologist, or in a multi-disciplinary meeting. In some cases, review consists simply of an administrator assessing for any missed elements. This last example highlights the importance of simply ensuring all required elements are present in the report. The American College of Surgeons mandated the inclusion of all pertinent elements in the vast majority of reports in 2004, an action supported by the College of American Pathologists through production of checklists for a wide range of diagnoses. Using such a method of synoptic reporting has been demonstrated to dramatically reduce the incidence of errors of omission, with one group finding a report completeness of 88% when synoptic reporting was used versus 34% where it was not [10]. Furthermore, use of synoptic reporting tools which can generate parts of the diagnosis (for example staging information for cancer) in light of inputs generated by the pathologist are found to improve reporting even further by reducing ‘fast thinking’ bias and assumptions [11].

Summary

Errors in pathology reporting are rare, but can lead to significant consequences for the patient, and potentially expensive lawsuits. Aside from system errors, a large proportion of errors arise from a small number of diagnoses and for a small number of reasons [6]. Many problems arise from omissions, perhaps due to elements being forgotten or a lack of familiarity with the diagnosis and all the information required. With increasing complexity of diagnosis, and the rising use of supplementary tests, this becomes ever more challenging. Synoptic reporting dramatically improves report completeness and accuracy. Differences in interpretation are seen, and may warrant peer review, or review at a second institution, to ensure the best possible diagnosis is achieved [12].

References:

  1. Prospective peer review in surgical pathology. Lind AC, Bewtra C, Healy JC, Sims KL. Am J Clin Pathol. 1995 Nov;104(5):560-6. https://www.ncbi.nlm.nih.gov/pubmed/7572817
  2. Patient safety and error reduction in surgical pathology. Nakhleh RE. Arch Pathol Lab Med. 2008 Feb;132(2):181-5. https://dx.doi.org/10.1043/1543-2165(2008)132[181:PSAERI]2.0.CO;2
  3. A Six Sigma Trial For Reduction of Error Rates in Pathology Laboratory. Tosuner Z, Gücin Z, Kiran T, et al.Turk Patoloji Derg. 2016;32(3):171-7. https://dx.doi.org/10.5146/tjpath.2015.01356
  4. Surgical Specimen Management: A Descriptive Study of 648 Adverse Events and Near Misses. Steelman VM, Williams TL, Szekendi MK, Halverson AL, Dintzis SM, Pavkovic S. Arch Pathol Lab Med. 2016 Dec;140(12):1390-1396. https://dx.doi.org/10.5858/arpa.2016-0021-OA
  5. Quality and patient safety in the diagnosis of breast cancer. Raab SS, Swain J, Smith N, Grzybicki DM. Clin Biochem. 2013 Sep;46(13-14):1180-6. https://dx.doi.org/10.1016/j.clinbiochem.2013.04.024
  6. Medicolegal aspects of error in pathology. Troxel DB. Arch Pathol Lab Med. 2006 May;130(5):617-9. https://dx.doi.org/10.1043/1543-2165(2006)130[617:MAOEIP]2.0.CO;2
  7. Pitfalls in lymphoma pathology: avoiding errors in diagnosis of lymphoid tissues. Wilkins BS. J Clin Pathol. 2011 Jun;64(6):466-76. https://dx.doi.org/10.1136/jcp.2010.080846
  8. Study of parameters to ensure quality control in histopathology reporting: A meta-analysis at a tertiary care center. Qureshi A, Loya A, Azam M, et al. Indian J Pathol Microbiol2012 ;55:180-2.http://www.ijpmonline.org/text.asp?2012/55/2/180/97857
  9. Adequacy of surgical pathology reporting of cancer: a College of American Pathologists Q-Probes study of 86 institutions. Idowu MO, Bekeris LG, Raab S, Ruby SG, Nakhleh RE. Arch Pathol Lab Med. 2010 Jul;134(7):969-74. https://dx.doi.org/10.1043/2009-0412-CP.1
  10. Does standardised structured reporting contribute to quality in diagnostic pathology? The importance of evidence-based datasets. Ellis DW, Srigley J. Virchows Arch. 2016 Jan;468(1):51-9.https://dx.doi.org/10.1007/s00428-015-1834-4
  11. Breast pathology second review identifies clinically significant discrepancies in over 10% of patients. Khazai L, Middleton LP, Goktepe N, Liu BT, Sahin AA. J Surg Oncol. 2015 Feb;111(2):192-7. https://dx.doi.org/10.1002/jso.23788
  12. Second-opinion pathologic review is a patient safety mechanism that helps reduce error and decrease waste. Middleton LP, Feeley TW, Albright HW, et al. J Oncol Pract. 2014 Jul;10(4):275-80. https://dx.doi.org/10.1200/JOP.2013.001204