Necessary features
Demonstration
Ease of use
Ease of use in all settings, not just tertiary centers; the process should be simple
Reproducibility
Sufficient studies demonstrating good diagnostic agreement with moderate or substantial kappa statistical agreement
Clinically relevant
Diagnostic categories should have relevance to patient’s therapy and prognosis
1.
Ease of use: a system should be easy to use, teach, and learn . A system should be based on relatively few objective findings. Grading of breast cancer is as complex as a system may be and still be successful. It is dependent on three factors, tubular content, nuclear grade, and mitotic rate. These three elements are graded and combined to achieve the overall grade. This works because each of these elements is a simple three-level choice. Any more elements would probably make this system too complex with loss of ease. Features have to be clear, e.g., “the presence of marked variation in size, prominent nucleoli, and chromatin clumping is nuclear grade 3,” “> 75 % of the tumor is grade 1 tubular formation” etc.
2.
Reproducibility: the use of any system should be tested for diagnostic agreement between pathologists. This is ultimately the goal in reducing errors and achieving greater diagnostic agreements. Typically, a set of cases that are vetted to cover the spectrum of findings is circulated among a number of expert pathologists, and diagnostic agreement is calculated using a Kappa statistic. A Kappa value of 0.4–0.6 is considered moderate agreement and is generally acceptable in most systems. The substantial agreement level (> 0.6–0.8) is considered excellent .
3.
Clinical relevance: the levels devised by the grading or staging system must be relevant to patient treatment or prognosis if they are to be of benefit to patients. For example, it may be easy to classify a process into ten levels, but this is not helpful, if there are only two treatments that can be used to manage the patient. The only clinically relevant level will be the point at which therapy should be changed.
The failure of any scheme may lie in any of these factors. Pathologists are likely to avoid using a scheme if it is too complex or too difficult to learn. And if they are forced to use it, they are likely to apply the rules poorly. If a system is too vague or dependent on subjective criteria, pathologists will interpret finding in different ways and this will result in poor diagnostic agreement. Lastly, if a scheme is not clinically relevant, clinicians will not ask for it and pathologists will have no incentive to use it .
The Benefits of Using Standardized Terminology
1.
The ultimate benefit is improved diagnoses and improved patient care . In a perfect world using the most relevant terms to describe a person’s disease lets the clinician know exactly what to do. And it would not matter where geographically the patient was being treated, if the pathologist is using the same terminology that the clinician understands. Problems occur when pathologists use a system of disease classification that is obsolete and is no longer applicable to current disease treatment recommendations. Problems also occur when there are two or more systems of classification, and it is not apparent which system is being used.
2.
Standardized diagnostic information lends itself to ease of use. If the diagnostic categories are predetermined, it is easy to incorporate these categories into menus to choose from. It has been shown repeatedly that synoptic computerized reports are more complete and are preferred by clinicians. Pathologists more easily complete reports with more confidence in having included all the pertinent information needed to manage the patient.
3.
Standardized diagnoses enhanced data collection and comparison of treatments. Progress in medicine is dependent on ongoing clinical trials to improve upon current treatment. This process is heavily dependent on having accurate standardized diagnoses. When comparing different treatment modalities, researcher needs confidence that patients enrolled in the study have the same disease and even the same stage of disease. This is the only way to make direct and equitable comparisons regarding outcome of different therapies.
4.
Standardized diagnostic terminology facilitates adherence to standards of care. If there is an agreed upon language of terms to use in the diagnosis of a particular disease and a pathologist uses the standardized terms, the clinician is then able to take that information and apply existing guidelines for the treatment of that disease. For example, the pathologist’s diagnosis stage II squamous cell carcinoma of the lung, the clinician is able to access current recommendation for that tumor and stage and offer the patient the most current available treatment.
5.
Standardized diagnostic terminology facilitates the assessment of care where treatment guidelines exist. If guidelines exist for the use of standardized terminology such as the cancer protocols, then a director of a service may directly assess the appropriate use of those protocols. Lack of compliance with the use of cancer protocols in accordance to departmental or national policies would indicate clear suboptimal practice. The same could be said for clinicians that have practice guidelines for specific disease processes. With the availability of standards or standardized language, these assessments can be made objectively.
Additional Indirect Benefits of Standardization
When diagnostic as well as grading and staging criteria are agreed upon and published and pathologists start using the schemes, certain aspects of practice related to classification systems are then recognized and addressed. A series of articles published by the Cancer Committee of the College of American Pathologists demonstrate the ongoing evolution of these schemes [11–15]. Subtle deficiencies are identified in the schemes that are addressed with investigations to determine solutions or answers to questions. Examples of this type of work include practical problems in staging of breast cancer, highlighted in an article by Dr. James Connelly [12]. This includes the classification of isolated tumor cells in lymph nodes and the determination of tumor size when multiple tumors are present. Other problems in the lung cancer checklist are addressed by Dr. Alberto Marchevsky [13]. In particular, defining pleural involvement in the staging of lung cancer and defining tumor location with the presence of multiple tumors. He also addressed determination of lymphatic involvement by tumors. Dr. John Srigley [14] emphasized preanalytic aspects of specimen handling in order to maximize the information gained during microscopic examination and synoptic reporting . Dr. Carolyn Compton [15] addresses problems with colorectal standardized reports focusing on the evolving issue of radial margin in colorectal surgery. She also addresses issues concerning microscopic tumor satellites and determination of lymph node metastasis when lymph node tissue is not identified. In summary, establishment of standardized diagnostic as well as staging and grading criteria is not a static process but is continuously evolving with the study and identification of selected problems, leading to further investigation and clarification of criteria.
Agreement on a single set of diagnostic, staging, and grading criteria makes it easier for pathologists to learn a system and use it for patient care. When more than one system is used, confusion exists as to which system should be used. Studies have demonstrated that pathologists’ lack of familiarity with an organ system leads to less than adequate pathology reports [16].
Conversely, the use of synoptic reports with standardized grading and staging leads to improvement among nonspecialists as demonstrated by Messenger [17]. He demonstrated that nongastrointestinal pathologists attained a level of report completeness that was comparable with gastrointestinal pathologists.
Meier et al. [18] showed that with standardization of diagnostic criteria enforced through systematic case reviews of breast and prostate cases, there was a reduction of amended reports for diagnostic misinterpretations over a 4-year period.
Examples of Successfully Used Current Systems
It is difficult to understand the power of standardization without using actual examples. Table 8.2 lists examples of existing classification schemes. Table 8.3 lists areas that generally benefit from classification schemes.
Table 8.2
List of existing standardized diagnostic criteria and staging
Disease and tissue type | |
|---|---|
Cancer protocols | Over 50 cancer protocols have been developed and are used. Available at www.cap.org |
Barrett esophagus | Illustrates ease of use, reproducibility, and clinical relevance |
Bethesda cervical cytology and thyroid cytology | Illustrate important aspects of reporting in cytology specimens including report adequacy, risk assessment of borderline lesions, and associate management |
Banff transplant rejection | Defines forms of rejection and the necessary features/ancillary studies needed for different levels of rejection. This is correlated with treatment protocols |
Hepatitis grading and staging | Defines active disease as well as chronic sequelae (fibrosis) of the disease. Extent of disease is decoupled from etiology |
Table 8.3
Lesions and conditions that may benefit from standardizations
Lesion or situation | Benefit of standardization |
|---|---|
Limited sample specimens (e.g., cytology, small biopsies) | Define adequacy, limit variability in diagnostic terms |
Cancer | Determine diagnostic criteria, Standardize prognostic features including grade and stage Suggest useful ancillary studies for diagnosis and prognosis Include and emphasize feature relevant to therapy |
Borderline lesions (e.g., dysplasia, carcinoma in situ)
 Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
Get Clinical Tree app for offline access
|