Intended for healthcare professionals

Education And Debate GMC and the future of revalidation

A way forward

BMJ 2005; 330 doi: (Published 02 June 2005) Cite this as: BMJ 2005;330:1326
  1. Mayur Lakhani, chairman of council (mlakhani{at}
  1. Royal College of General Practitioners, London SW7 1PU

    Revalidation is a summative assessment and must be seen as separate fromappraisal and clinical governance


    The General Medical Council's proposals for recertifying doctors in the United Kingdom every five years have been thrown into confusion. A recent judicial inquiry into the case of Harold Shipman, who murdered at least 215 patients, found that the method for revalidating doctors based on regular appraisal would not detect poorly performing doctors.1 The government suspended plans for revalidation and ordered an urgent review led by the chief medical officer.2 This article gives a proposal for a way forward. Although the article is from general practice, the framework described is applicable to other specialties.

    Problems with current model

    The GMC's proposal for revalidation requires doctors to keep a portfolioshowing participation in appraisals and completion of an agreed personal development plan. They also have to provide a statement of “no concerns” fromtheir primary care organisation and evidence of probity and health. To understand the problems with the current model, we have to be clear about the definitions of appraisal, clinical governance, and revalidation (box 1).

    The three methods of assessing performance overlap.5 This overlap has advantages and disadvantages. On the one hand, data collected for appraisal can be used for revalidation, thus reducing the burden of assessment. On the other hand, a summative process such as revalidation could hijack a formative process such as appraisal. A fundamental observation of the Shipman report was that although appraisal and clinical governance inform revalidation, they are not sufficient to evaluate a doctor's practice. The current proposals therefore give an “illusion of protection,” particularly as patient safety and reassurance of the publicare important functions of revalidation. What can be done to overcome this fundamental shortcoming?

    Making revalidation work

    My proposal for successful revalidation is based on 10 guiding principles (box 2). Below, I describe the rationale behind the principles.

    Revalidation is a summative process

    Revalidation is an evaluation of a doctor's fitness to practise—that is, an assessment that requires a summative judgment (pass or fail). This concept is important strategically and has not received sufficient attention. It is consistent with the reform of postgraduate medical education and the introduction of training and assessment based on competencies. Postgraduate assessment validates doctors for specialist practice in the first place by allowing them to be entered on to the specialist register. Revalidation is affirmation of continuing fitness to practise and therefore must relate to compliance with defined competencies. The academic criteria for an assessment process include6:

    Box 1: Definitions

    Appraisal is a formative and developmental process.3 It is about identifying development needs not performance management

    Clinical governance is a framework through which NHS organisations are accountable for continuously improving the quality of their services and safeguarding high standards of care4

    Revalidation is an episodic process to show fitness to practise to the professional regulator (the GMC)

    • The need for criteria, standards, and evidence

    • Involving the public in judging assessments

    • Using multiple sources of data and assessors

    • Sampling over several time frames rather than at a single point.

    Box 2: Ten guiding principles for revalidation

    1. To recognise that revalidation is a summative assessment process which should conform to the established academic criteria for assessment

    2. Clear criteria, standards, and evidence are required for revalidating doctors.

    3. There should be meaningful lay involvement in the design and implementation of the revalidation system at both national and local level

    4. Neither appraisal nor clinical governance alone, or in combination, as currently constituted can be a suitable basis for revalidation. They can, however, help a doctor gather evidence for revalidation.

    5. Appraisal is primarily a formative assessment, and should notbe used as a regulatory instrument. Both appraisal and clinical governance need to become more robust if they are to inform revalidation

    6. Local certification should be an integral part of revalidation. Clinical governance processes need to be designed to provide informationon the individual doctor

    7. To ensure maximum reliability, the information presented by adoctor in his or her portfolio should be examined by more than one method or from more than one point of view

    8. Additional approved routes and intervals for revalidation should be permitted

    9. There should be a tighter definition of a managed clinical environment; for independent contractors this should entail practice accreditation schemes

    10. The principles and framework for revalidation should so far as possible be consistent and comparable across the whole medical profession

    Need for clear criteria standards and evidence

    Neither appraisal nor clinical governance defines the detailed criteria for revalidation. Although the GMC's Good Medical Practicesets out the general requirements for good practice,7 detailed criteria, standards, and predefined evidence need to be developed to facilitate explicit assessment of a doctor's portfolio for revalidation. This would also allow doctors to know what they have todo to be revalidated. The Royal College of General Practitioners (RCGP) hasdeveloped criteria and standards for general practice.8 9 Supporting policies are also available that define poor practice in general practice.10

    All disciplines should define criteria, standards, and required evidence for three dimensions of care (adapted from R Baker, personal communication):

    • Quality of technical care (mainly clinical audit, medical records)

    • Quality of doctor-patient interaction (such as communication skills)

    • Compliance with codes of practice, such as Good Medical Practice and royal college standards such as membership in good standing.

    Meaningful lay involvement in revalidation

    The public must be involved in revalidation to ensure accountability andconfidence in the system. This is one of the principles stated in the criteria for excellence in assessment. Public involvement will also prevent revalidation being seen as a closed process, confined to the medical profession.

    Need for separate revalidation procedures

    Neither appraisal nor clinical governance assures a doctor's competence or performance, and even together they are unlikely to pick up poor performance. Both can inform the process, but an additional element of evaluation is needed (see principle 7).

    Appraisal should not be used as a regulatory instrument

    Appraisal should remain formative and educational. It has the potential to become a powerful tool for personal development. To be effective, however, appraisal has to be a confidential process. If it becomes a regulatory tool, its purpose is undermined. Appraisal is not designed to provide an objective judgment about a doctor's practice. Moreover, appraisers are not trained to be regulatory assessors, and many would not wish to be so.9

    Appraisal should become more robust and be quality assured. Important elements of this include defining the data requirements of appraisal, qualityassurance of personal development plans, and training and accreditation of appraisers.

    Local certification should be part of revalidation

    Certification from clinical governance systems gives additional information that may not be available in the doctor's folder. To provide local certification, clinical governance will have to be redesigned to provide a minimum dataset for individual doctors such as performance indicators. Proactive examination of this information and the doctor's portfolio by a local health community panel can help us move towards making a positive statement about any concerns relating to a doctor (I know that there are none) rather than a negative statement (I know of none).

    Portfolios should be examined by several methods

    Appraisal and clinical governance are processes that inform revalidation but are insufficient by themselves. The information gathered by a doctor in his or her portfolio also needs to be assessed according to explicit criteria. This therefore requires a process, and a local panel could do this. Trained assessors are necessary to do this. The exact composition of a localpanel needs further discussion. The local assessment provides triangulation(figure). Triangulation allows judgments to be made about attributes that are important when single assessment methods lack reliability.


    Additional routes and intervals for revalidation should be permitted

    Many doctors report a heavy workload and excessive inspection. To make revalidation more feasible and proportionate, we should consider allowing royal college assessments for postgraduate qualifications to be accepted as a method of revalidation. An example of a scheme is the MRCGP by assessment of performance.11 The fifth Shipman report suggested that the interval for revalidation for some doctors could rise to seven years. What is important is that a doctor complies with specified criteria and standards and not necessarily how compliance was achieved. This issue is of strategic importance. It allows flexibility without losing rigour. It is a radical departure from the GMC's one size fits all proposals. Local certification will still be necessary. Medical royal colleges potentially have an important contribution to make to revalidation and their role has been underestimated.

    Tighter definition of managed clinical environment

    The GMC model is predicated on doctors working in a managed environment.In general practice doctors have a variety of contractual options, including independent contractor status. Independent contractor status is a valued and important contractual option. However, Dame Janet Smith argued that independent contractor status may be an obstacle to implementing clinical governance in primary care.1 She therefore recommended a system of practice accreditation. Many practices already undertake organisational development schemes.12 Such schemes allow doctors to show that they are working in clinical environments that have been peer reviewed and externally accredited.

    Consistent principles and framework for whole profession.

    The principles outlined here apply to all disciplines. It should not be harder or easier to revalidate in one craft then another; besides, many doctors have portfolio careers—for example, having a post in general practice and in a hospital. Consistency requires clarity about the roles of organisations. The NHS, medical royal colleges, and the GMC need to work more closely together to clarify roles and define the linkage between NHS systems andlicence to practise:

    • The GMC should set the overall standards and framework for revalidation including quality assurance

    • The royal colleges should establish detailed criteria, standards, and evidence for their specialties

    • The NHS should work with professional organisations to develop clinical governance and appraisal into more effective processes.

    Knowledge tests

    Dame Janet Smith recommended a mandatory knowledge test for general practitioners as part of revalidation. Factor analysis of the RCGP membership examination shows that tests of knowledge are by far the most accurate predictor of poor performance in the overall examination (R Neighbour, personal communication). Although knowledge tests can confirm what a doctor knows, they cannot confirm what a doctor can do (competence) or actually does (performance). Nevertheless, a doctor cannot have competence without knowledge. Knowledge tests are a reliable assessment instrument and could be useful for formative self assessment. A knowledge test by itself is not sufficient to revalidate a doctor. Tests should be considered as one of several sourcesof information in a doctor's portfolio. If revalidation were to include a mandatory or voluntary test of relevant clinical knowledge, the RCGP could use its in-house assessment procedures to provide suitable materials.


    Revalidation should protect the public from poorly performing doctors and provide reassurance. Current systems do not provide sufficient protectionfor patients. The GMC must assert itself and promote a reliable system of revalidation for all doctors. Harold Shipman would probably have done quite well in a general practice appraisal. His patients trusted him, but tragically their trust was misplaced. Revalidation can be made to work, and many tools and resources already exist to support it, particularly in the royal colleges. All major general practice organisations have supported revalidation for a long time. The principles described above will enable a reliable and feasible system of revalidation to be put into place. Additional resources are essential to support doctors undertaking revalidation.

    Summary points

    Revalidation is an essential tool to protect patients from the minority of poorly performing doctors

    Current systems of appraisal and clinical governance do not evaluate a doctor's fitness to practise

    Revalidation should be regarded as a form of summative assessment and conform to established academic criteria for assessment

    Additional approved routes and intervals for revalidation should be permitted

    A model for revalidation has been defined for general practitioners and tools are available to support it

    This is the fourth article in a series examining regulation of doctors

    This article represents the opinions of the author and does not constitute official RCGP policy.


    • Contributor ML has been a clinical governance lead and a GP appraiser and was involved in deeloping Good Medical Practice for GPs and the RCGP policy on revalidation as a council member and an officer. He has dealt with underperformance issues in primary care.

    • Competing interests If this plan is accepted the RCGP might benefit financially through the uptake of its tools and resources for quality


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.