General Practice

Commentary: Screening for melanoma risk is misguided

BMJ 1995; 310 doi: (Published 08 April 1995) Cite this as: BMJ 1995;310:916
  1. Duncan Keeley

    With an incidence of malignant melanoma of 1 in 10000 a year1 2 a general practitioner with a registered list of 2000 patients will see about one new case every five years. Risk factors for melanoma have been identified, and one strategy for prevention is to target those at higher risk for more intensive advice on reducing their exposure to the sun and on what to do if moles change. This study investigates in an “affluent practice” some aspects of the feasibility of such a strategy. But is the strategy worth investigating at all?

    In this study 18% of adults were found to have high mole counts, while 5.5% had both freckling and high mole count. Five per cent of all adults reported “major” change in a mole in the preceding three months. Johnson et al estimate that at least 4% of a general practitioner's list have a fivefold or greater relative risk of developing a melanoma.3 Implementing a strategy of intensive counselling of high risk individuals therefore involves getting the entire population to count their moles and provide other information pertaining to melanoma risk, and then labelling around 5% of the adult population as having an increased risk of melanoma. Screening for disease may generate anxiety.4 This strategy would have significant adverse consequences in this respect.

    No evidence yet exists that such a strategy would reduce mortality. Cochrane and Holland stated that as an ethical prerequisite for a screening programme there should be conclusive evidence that screening can alter the natural course of disease in a significant proportion of those screened.5 Johnson et al3 estimate that a randomised trial of this strategy aiming at detecting a 20% reduction in mortality would require 100000 high risk adults in each arm and would need to involve all adults living in at least two health regions in Britain.

    The implementation of this strategy would substantially increase the workload of practices. The screening process itself is complex, and the consultation rate in connection with pigmented lesions would increase. Given a 5% rate of reported major change in moles in the adult population every three months, the rate of their excision and submission for histology would increase substantially. The concept of prioritising practices for screening for melanoma risk according to the age or social class distribution of their patients is unacceptable.

    The authors of this study have not made any assessment of the psychological effects of their screening method or assessed its effect on practice workload or health service costs. Feasibility studies of possible screening strategies that fail to address these issues should not be undertaken. Further research on the prevention of melanoma and death from melanoma should concentrate on finding better methods of communicating the message that sunburn is bad for you to the general population, and on improving doctors' skills in early diagnosis.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    View Abstract