Clinical governance is one of the most frequent terms encountered in health care management, in both the national health services and private healthcare systems. It describes a systematic approach maintaining and regularly improving the quality of patient care while safeguarding high standards of care through creation of an environment in which excellence in clinical care will flourish. The elements of clinical governance advance health care professional education, clinical audit, clinical effectiveness, risk management, good professional practice-based research and development, confidential-based honesty procedures. In this study, we delineate the characteristics of an EQA scheme that should be taken into consideration when establishing a national or local EQA scheme. We consider that if a quality assurance scheme is to be suitable as revalidation method it should have two circulations per year, requirement for membership or a substantial pediatric pathology workload. It should be open to overseas participants, have CPA accreditation, provide CPD credits and certificates, and have regular slides discussion and scheme improvement meetings and an annual report. Internal audits and a 'black box' of tissue slides are important, but external proficiency testing is advocated as an independent means for continuing professional development and to support laboratory accreditation. Indeed, laboratory services are a central core in the diagnostic procedures of a general hospital . The "Pediatric Pathology" module of slide survey programs is targeted for pediatric pathologists, general pathologists who have pediatric cases in their diagnostic routine, pediatric pathology fellows and pathology residents.
The concept of EQA programs relies on the fact that participants should view cases without consulting colleagues before submission to the program office. Uniformity of standards to set up a quality assurance program at national level is an important task. The use of physician specialization to assess the quality of care provided by individual physicians represents a structural approach to measuring quality. In the last 10 years there has been an enormous change in revalidation in the United States. This is because physician specialization as represented by board certification may be an unreliable measure of the quality of a physician's performance over time, unless his or her knowledge and skills in a specialty area are periodically updated or assessed.
In such a way, there is no guarantee that physicians have maintained the same level of skills and knowledge they demonstrated for their initial certification without a revalidation process. Thus, the search for a practicable revalidation method is very valuable. According to our data and considerations, the UK scheme has an educational component largely as a by-product of participation, but it can easily be used as a form for medical revalidation. This is because it possesses a partial confidentiality, contains multiple sources of data and assessors, and has a sampling procedure over several time frames rather than at a single point. The German scheme would appear to be essentially that of an educational program and possibly the North American scheme too. The Australian scheme seems to have both educational and revalidation components, but the number of slides is low and there is only one circulation per year that we consider insufficient as a revalidation method.
In assessing EQA schemes one aspect that has to be emphasized is the 'consensus answer'. The EQA utilizes the consensus answer as the correct answer which may not necessarily be the same as an 'expert' answer. Inter-observer reliability (consistency) may be reassuring, but there is a risk of a group response that can become very unpredictable.  The participants may all agree, but the correct or best answer may not be made as all of the participants are wrong. Similarly, an expert opinion is not necessarily correct. It is well recognized in pathology that there are instances when pathologists cannot agree as to the correct answer even on cases that are not unusual or rarities. Disagreement between experts is to be expected and should not necessarily give rise to concern over their competence . Consequently, the EQA, rather than measuring expertise, is measuring a minimum common level of attainment (i.e. conformity to consensus). Therefore how the consensus should be assessed? There are several forms of consensus used today. The promise of achieving a consensus creates the impression that one group may be able to operate without conflict. Consensus' philosophy fosters an atmosphere that encourages the expression of different opinions and even conflict, and then provides a process that may lead to a resolution in a creative and very supportive environment. It recognizes that everyone has a contribution to make, and that all views are encouraged, but it may take several meetings to work towards a decision and at the end of the process it is not uncommon for some people to continue to disagree with the final decision. The Delphi Technique and Nominal Group Technique are two well-recognized consensus-formation methodologies specifically designed to combine judgments from a group of experts [19, 20]. The Delphi Technique utilizes a series of well-defined questionnaire-based surveys, whereas Nominal Group Technique is a structured face-to-face meeting designed to facilitate consensus. Consensus-formation techniques require that each step builds on the results of the previous steps. In the majority of the EQA survey programs for pathologists the correct diagnosis is not indicated as 'expert judgment' but conversely a 'democratic' consensus is sought by judgment of the majority of participants and discussion in the case of conflicting results. In some situations, a consensus cannot be reached and the controversial tissue slide is excluded from the final EQA performance. The rejection of cases not reaching 80% agreement between participants can be criticized, as it is artificial, but it is a compromise . As a consequence of rejecting such cases, the distribution of the score profile becomes skewed.
Another aspect at the base of some controversial issues in EQA in pathology may be the lack of uniform guidelines that are valid worldwide. Thus, our study first tries to compare four international survey programs, emphasizing the need for harmonization. However, lack of harmonization also relies on the different interpretation of some pathology. An example is the diagnosis of chronic intestinal pseudo-obstruction in children. The diagnosis of intestinal neuronal dysplasia as an isolated entity is indeed very controversial. Intestinal neuronal dysplasia is characterized by hyperplasia of the submucosal and myenteric plexuses and the isolated form has been described in the distal colon and rectum, and its clinical presentation, with constipation and intestinal obstruction, can mimic that of Hirschsprung disease or aganglionosis [22, 23]. During the years since the entity was first described, the criteria have also been modified. However, in many countries, mostly of Anglo-Saxon origin, the current view is that isolated intestinal neuronal dysplasia is seen in a variety of clinical settings and is more a descriptive entity than a specific disease requiring surgical intervention .
It is difficult to identify the minimal number of slides that needs to be assessed for consideration as a revalidation method. We initially considered the number of cases of pediatric and placental pathology in a tertiary academic center. The number of cases of pediatric and placental pathology was found to be approximately 1500 in many institutions (personal communication) over the years. Subsequently, we considered the percentage of 'permitted wrong diagnoses' in the diagnostic routine identified in the English literature. The results of studies concerned with error rates in histopathology vary widely; no serious errors,  0.26%,  up to 1.2% of histopathological reports ) but these were performed in academic or teaching institutions. In reality, the percentage is quite variable when considering sub-specialistic, inter-individual and intra-individual variability studies . It has also been suggested in the USA that false negative rates of 5–10% may be an admirable goal in cytopathology, and rates below 15–20% are a possible standard . We allowed 2% in consideration of typographical errors, and hypothesized that a valuable number of EQA diagnoses could be 30 (2% of 1500 diagnoses), keeping in mind an artificial assumption that the highest number of wrong diagnoses can be equivalent to the number of histological glass slides that run in a annual EQA program. Thus, we propose that 30 should be the minimal number of slides per year assessed in an EQA program by a specialist pediatric pathologist. In this sense, the UK scheme with 30 slides per year might be considered as a standard and may serve better as a revalidation method.
Health services are awash with data, but safety is a category that needs to be continuously improved for health services. IT resources appear to be the greatest barrier to obtaining access to virtual images. This can be overcome if hospitals or government upgrade workstations to have compatible IT systems and allow access for quality assurance purposes. Smooth running quality assurance programs can improve this relationship and strengthen the link with the clinicians. Modernizing the pathology laboratory commences with virtual microscopy and services will provide a remarkable wealth for the child's growth in the 3rd millennium. Image digitalization is a new tool to use for biopsy specimens that cannot be cut in 40–60 slides for all participants. Indubitably, digital imaging with virtual microscopy will be more closely linked to practice in the future. To date, glass slides mimic the routine practice worldwide. Thus, digital images for EQA are a compromise offering the possibility to extend the range of cases to include small specimens, such as endoscopic and fine needle biopsies. Another possibility is to manage the use of glass slides from small biopsies through a postal system, circulating slides sequentially between participants. In the recent UKNEQAS Meeting in Glasgow it was stated that the EQA remains an extremely valuable resource for clinical pathologists and needs a well-organized, rapid and manageable system to run efficiently .
It seems evident that EQA drives standardization and there are many examples supporting this fact, including those where peer-driven changes are influenced by EQA findings . One aspect that was not considered in this study was the difficulty of the histological glass slides utilized in the circulations. However, in our opinion, there was no significant disparity in the difficulty of diagnosis because some recommendations have been proposed and followed in all four EQA schemes examined. In particular, the diagnosis should be made using the hematoxylin and eosin stained slide and no immunohistochemistry should be needed to arrive to the diagnosis. Lymph node cases usually represent a frequent source of difficulty and specific details have to be provided in submitting such cases. Thus, the method for selecting cases has a crucial impact on whether the cases are more difficult (and hence more educational) or more representative of the routine workload (and therefore more relevant to performance surveillance). Cases should be contributed by all participants in rotation, following agreed guidelines. Extremely 'simple' cases should be avoided, to be determined at meetings of the participants, but bizarre cases and case-report material remain inappropriate. The number of cases circulated must be sufficient to give reasonable confidence that serious sub-standard performance will be promptly identified.
Inappropriate tissue slides, limited number of circulations, lack of secretarial support, and low number of participants may jeopardize the EQA as a revalidation method. The importance is to establish a quality management system, allocate privileged time for it and ensure a reasonable cost load for healthcare organizations. It is important to specify well defined implications for schemes and participants. Further, we believe that a periodically reviewed and updated quality policies manual in addition to continuing audit of performance should be standard in every histopathology department. The association with a CPA accredited laboratory (contractual agreement) should be considered. If an EQA scheme is not run in a department, it is recommended that the department is not given accreditation.
Business and healthcare organizations regularly use the process of benchmarking to learn how others address policy issues and solve problems. The improvement of diagnostic skills in pathology is of paramount importance and the interest in programs that provide external proficiency testing, quality assessment and appropriate education programs to public and private laboratories of pathology and laboratory medicine is growing rapidly. There are some associations and companies supplying both quality assurance programs and supporting services for the benefit of pathology laboratories and personnel working within the pathology environment. The substantial advantage of internal audit and slide survey is intrinsically present in these programs intended to continually improve pathology services for the well being of communities. There is no direct evidence to support the validity of histopathology EQA, but there are various strands of indirect evidence that can be drawn together to underpin validity. Indirect evidence to support validity includes the response process used in the EQA. This reflects actual working practice undertaken by pathologists. EQA started originally as a 'hobby' for pathologist participants, but now it has acquired or is acquiring a central role as part of continuing professional development. It may be used as a revalidation method, because it represents clear of medical qualification, its' results may be given to the public, it may contain multiple sources of data and assessors and it represents a sampling procedure over several time frames rather than at a single point. However, the purposes and programs of EQA schemes can be different worldwide and in consideration of its possible use as revalidation method, it is strongly advisable that all EQA programs are unified as soon as possible.