From the publication of the first studies revealing GPs' apparent inadequacy at identifying and treating patients with depression, mental health has dominated the primary care research agenda. Much effort has been expended to find a way of transmuting yesterday's erratic GPs into tomorrow's infallible diagnosticians. But for the most part this philosopher's stone has eluded discovery. On page 665 a group in New Zealand document the effect of systematic case finding. The authors conclude that it is unlikely to be a useful approach, illustrating the difficulties of using screening instruments that are always imprecise. They also found that the doctors in the study were better at spotting depression in patients they already knew, placing the discussion, in their phrase, ‘within the context of general practice continuity of care.’ The accompanying leader takes up this aspect, with the statement that ‘the diagnosis of a mental health disorder … evolves in a context of trust’ (page 659). The leader, however, also quotes the NICE guidelines, which recommend screening for depression among certain high risk groups. NICE guidelines state openly the level of evidence on which the recommendations are made, but innocent recipients of the guidelines may be forgiven for occasionally forgetting to check, and others may wonder what on earth this august body is doing including in its recommendations anything that rests on such a flimsy evidential base. Or is NICE playing a more subtle game, deliberately including ludicrous recommendations in order to signal that the others are not to be taken seriously?
Then there is examination of death rates. After the discovery of Shipman's murders it was natural for authorities to call for GPs to monitor their mortality statistics. Once again real life turns out to be more complex and less susceptible to simple solutions. On page 670 a study from Northern Ireland reports that all practices could explain apparently disparate rates in terms of the population they worked with, concluding that ‘case-mix adjustment is not a perfect science and has its own associated risks.’ Among the risks is one described in the leader, where practices finding themselves with high rates might stop registering patients resident in nursing homes, to the obvious disadvantage of such patients (page 660). Guthrie also points out the obvious difficulty that monitoring mortality rates by practice will conceal any individual doctors whose practice is associated with high mortality, and warns against introducing ‘an expensive system that does not work.’ Where death is concerned, we have known for some time that even the death certification system does not work as well as it should, and the paper on page 677 provides some insight. When faced with a dilemma, GPs in New Zealand favoured respecting the feelings for the grieving family over providing a completely accurate certificate for cause of death, once again raising the question about defining the criteria for good quality of care. More familiar questions about euthanasia and assisted suicide have appeared on the agenda recently, provoked by political debates in the House of Lords, and taken up in the mass media. On page 720 Ilora Finlay provides a carefully argued case for opposing any change to legalise assisted suicide, as did many speakers in the RCGP Council debate in June. However, she also acknowledges that the public doesn't appear to share the view of the profession. While doctors and patients value the role that GPs can play in terminal care (page 684), we should be concerned if we find ourselves taking up a position so far from the one held by the general public. We might be right, but it's not comfortable.
Both the mental health debate and the various aspects of the way in which we care for dying patients are different elements of quality of care, and these papers illustrate how difficult it is to define high quality, in two areas that are of enormous importance to patients and doctors. Faced with such difficulties, the notion of judgement by one's peers looks beguilingly attractive. A team in Wales, working on features of revalidation, have tried out such an approach (page 690). They report that the questionnaire could be acceptable. Before we all rush to adopt the system, we should have to consider whether the items where peers consider themselves unable to judge are too numerous to trust the results, especially since the numbers were highest in those areas mostly dealing with clinical competence. Measurement of quality is another area that will keep us all busy for some time to come.
- © British Journal of General Practice, 2005.
