All changed, changed utterlyBMJ 1998; 316 doi: https://doi.org/10.1136/bmj.316.7149.1917 (Published 27 June 1998) Cite this as: BMJ 1998;316:1917
All rapid responses
QUALITY ASSESSMENT BY REGISTRARS - FIRST HAND, IN-DEPTH, ACCURATE AND
We read with interest the article by Smith, highlighting serious
issues facing the medical profession and the need to implement mechanisms
to ensure high quality of care.
Accurate, on-going assessment is primary to quality control in health
care. The complex nature of modern medicine necessitates that such
quality assessment should come from within each speciality.
The use of statistical analysis to assess healthcare personnel is
fraught with inaccuracy, distortion and unacceptable risk to patients,
since a large number of failures are needed to constitute statistical
significance. Moreover, statistics can be tampered with. It is easy to
understand the temptation to do so when career and honour are at stake.
Undue importance to statistical analysis may also generate reluctance to
treat high-risk patients, for fear of adversely affecting morbidity and
mortality rates. This applies especially to highly specialised senior
clinicians, to whom are referred the cases with the most appalling
Perhaps the most effective instrument for improving clinical
performance is structured, constructive and critical feedback from
professional colleagues. However, limited resources and temporal
constraints preclude consultants assessing one another.
An alternative strategy involves the institution of ongoing formal
assessment of consultants by their registrars, possessed of considerable
clinical experience and acumen. Registrars rotate from firm to firm and
are thus in a position to effectively disseminate best knowledge and
practice amongst the consultants they work with. Moreover, a registrar has
the advantage of assessing a consultant over a six-month period. Such a
long-term review, with the advantages of depth and comprehensiveness may
well be superior to a short assessment by an external consultant.
It is recognised that assessment by multiple observers is essential
to obtain acceptable reliability . Political implications and personal
bias can also be minimised if every registrar passing through a firm is
required to assess the consultant and the responses averaged. Such a
system would also be time and resource effective. Most importantly, this
would ensure the retention of quality control within the medical
profession, preferable to its enforcement by external, non-medical
For such a system to work, it is vital that an environment be created
in which senior clinicians do not feel insulted by the idea of being
assessed by their trainees, while junior doctors are free from the threat
of retribution from their consultants. An unbiased and open-minded
approach to this concept may facilitate the implementation of a highly
effective system of quality control in health care.
Dr. Sharmila J. Menon
Honorary Research Worker
Dept. of Human Resources
Yorkhill NHS Trust
Competing interests: No competing interests
Richard Smith refers to 'key protagonists over-reacting' in his editorial on 'the Bristol case'(1). He joins 'even the strongest supporters of the Labour Government' in bemoaning 'its excessive concern with media opinion'. Yet he tells us (and, from his privileged platform, the world) that this case is a 'once in a life-time drama', 'Shakespearean in its scale and structure', which will have the result that 'the trust that patients place in their doctors' 'will never be the same again'.
This is a strange stance for the editor of a scientific journal committed to the encouragement of rational ideas in medicine. It is of course true that 'dramas' like this (even dramas which occur more often than 'once-in-a-lifetime') are 'powerful levers for change'. But that is not a reason for senior medical journals to seize on them and use them to promote pre-existing agendas of change which have only a tangential bearing.
The lesson of the Bristol tragedy, when the dust and the shouting outside the GMC has subsided, is that there was a gross and inexcusable failure of EXISTING mechanisms of control. It provides no rational support whatsoever for the wholesale imposition of systems of monitoring and control on doctors in general. Such systems may or may not be necessary, and the BMJ should be arguing that if they are introduced it should only be done with proper testing, so that ill effects, predicted by myself(2) amongst others, can be determined and weighed against the benefits, predicted by Richard Smith amongst others. But that has very little to do with the Bristol case. Here the problem, as we now understand it, was not a failure of detection, it was a failure by those in authority to take any action on the warnings that they repeatedly received. What purpose will all the monitoring in the world serve if society cannot respond to the shouts of warning it is already receiving?
Last week I didn't find the level of trust in my practice had, "all changed, changed utterly". This is partly due to patients well-known capacity for exempting their own doctors from the strictures they apply to the profession in general. Of course on the broader field we must look for careful progress, but this is not the time for responsible journals to indulge in emotive polemic.
James A R Willis, General Practitioner
1 Smith R, All changed, changed utterly, BMJ 1998;316:1917-8
2 Willis JAR, The Paradox of Progress, Radcliffe Medical Press, 1995
Competing interests: No competing interests