Excising moles that worry people seems simplest.

When specialists demand to provide a service, they often complain shortly afterward about the volume of work they acquire, and try to dump it and the risk back into general practice.

Either accept GPs actually managing the presenting condition and patient, or accept the referrals.

Competing interests: None declared

The UK melanoma mortality rate continues to rise (www.ncin.org.uk). Stage at treatment is the main determinant of outcome, related to public awareness, early diagnosis and prompt treatment. A tension exists between early referral of possible melanomas, and over referral of benign pigmented lesions. Our editorial was about a new technology tested as a means of improving diagnostic accuracy in primary care. (1) Based on the findings of the study, we intended to suggest that targeted education of primary health care teams and specialisation within teams might be key to maximising diagnostic accuracy and referral in primary care, rather than reliance on technologies which appear to be difficult to optimise. The respondents have suggested that we have been critical of primary health care physicians, which was not in any way our intention. We understand the difficulties of diagnosis (in primary or secondary care). “Missing” melanomas is not confined to primary care: melanomas may confound the most expert dermatologists.

The variation in survival between countries is considerable: Eurocare 4 reports 5-year survival for patients diagnosed 1995-99 from 63% (Poland) to 92% (Northern Ireland) with Wales at 74% and England at 85%. (www.eurocare.it). NCIN data show a range for those diagnosed 2001 to 2005 from 79% in one region of the UK to 92% in another. The Patient Experience Survey for 2010 suggests that the majority of melanoma patients are referred in a timely fashion. (2). That there is a significant range in survival even within the UK however suggests that there remains scope for improvement, a proportion of which might be within the realm of diagnosis within primary care. This variation most likely largely reflects stage at excision in a relatively small proportion of patients, probably related to late presentation or failure to detect and treat quickly (although the root causes of that variation will be much better understood as data collection in the UK improves).

Intuitively it feels right that increased physician training and specialisation result in better outcomes, particularly in a condition relying on visual recognition. The study from Walter et al (1) suggests that this is indeed the case, with a significant improvement in ‘appropriate’ referral without increasing the false negative rate in the trained GPs. Cave comments that GPs need better scoring systems, and need to be empowered to utilise clinical photos, clinical review and possibly dermoscopy. Education is a key part of this. Most UK medical students however receive little dermatology training. Outpatient dermatology rarely forms part of the training of GPs yet consultation about moles is common. The UK Melanoma Taskforce took the view that increasing education of GPs about skin cancer was desirable. (www.skcin.org). Cancer Research UK and the British Association of Dermatology launched an “on line” tool at doctorsnet.co.uk this summer, which was funded by the DOH to aid in decision making for referral under the 2 week wait rule.

That there is marked regional variation in melanoma outcomes suggests that early diagnosis is something that can be addressed. The screening of pigmented lesions is difficult and our view is that education of health care professionals is to be recommended as an integral component of addressing this.

1 Walter F, Morris H, Humphrys E. et al. Effect of adding a diagnostic aid to best practice to manage suspicious pigmented lesions in primary care: randomised controlled trial. BMJ. 2012 Jul 4;345:e4110. doi: 10.1136/bmj.e4110
2 Lyratzopoulos G, Neal R, Barbiere J, et al. Variation in number of general practitioner consultations before hospital referral for cancer: findings from the 2010 National Cancer Patient Experience Survey in England. Lancet Oncology 2012, 13; 353-365

Competing interests: PLorigan - no competing interests. JNewton-Bishop - previously took part in a study designed to assess another diagnostic aid.

My GP colleagues make good points in their rapid responses to perceived criticism. But to be fair to Newton-Bishop and Lorigan, they point out in following the 'expert' guidance these are the results. Diagnosing melanoma clinically is clearly a very difficult business. They point out that known clinical methods have high false-positive rates. Unsurprisingly, teaching them to GPs does not appear to improve things. Even MoleMate did not improve the "rate of appropriate referrals". But would it reduce the sum total of morbidity and mortality from malignant melanoma? Faced with a subset of moles pre-selected by GP referral, a Specialist in doubt will perform an excision biopsy. Studies show that such biopsies are frequently benign histologically. Newton-Bishop and Lorigan state: "The increased rate of referral in the MoleMate group is reminiscent of the observation that dermatologists show increased concern about borderline lesions when they first start to use dermoscopy to examine naevi. This increased referral rate might therefore have settled over time. Nonetheless, the technology was not of benefit. "

How can we do better? Perhaps our aim (GPs and Specialists) should be to ensure that all malignant melanoms are promptly diagnosed and excised, rather than pursue 'a rate of appropriate referral'. Given the imprecision of clinical tools, this would carry costs - an increase in the numbers of benign lesions referred, or an increase in the number of benign lesions biopsied. Both events seem to be happening.

Dr Ashworth makes the salient point that we GPs really need a good clinical way of confidently reassuring the worried patient that she does NOT have a malignant melanoma. Presently the only sure way to do this is excision-biopsy.

Seeking such reassurance his own wife was found on biopsy to have malignant melanoma. I well remember two such cases in my own practice, back in the days when a GP excising a mole was a commonplace.

Competing interests: None declared

Whilst I won't labour the very pertinent points already made by my colleagues , I would also draw attention to the fact that according to the authors >90% of the referrals followed the referral guidelines (which were presumably drawn up according to national guidance with local, secondary care, interpretation).

If this has 'only' yielded a 'successful' detection rate of 11.8%, would the authors suggest that GPs ignore the guidelines, and only refer when they were 'more convinced'?

Competing interests: I am a GP