Elsevier

Public Health

Volume 122, Issue 12, December 2008, Pages 1433-1439
Public Health

Original Research
Assessing melanoma risk factors: How closely do patients and doctors agree?

https://doi.org/10.1016/j.puhe.2008.04.012Get rights and content

Summary

Objectives

Self-examination of the skin has been associated with a reduced risk of advanced melanoma, but self-assessment has corresponded imperfectly with clinical examination by a dermatologist, and shown only moderate accuracy in identifying individuals at risk.

Study design

Population-based screening campaign in Styria, Austria.

Methods

One thousand two hundred and twenty-three Caucasians volunteered for a free skin assessment by a dermatologist. First, they answered a questionnaire in which they assessed their own melanoma risk factors, and they were subsequently examined by a dermatologist. Kappa agreements between the two assessments were calculated.

Results

The overall kappa agreements on the estimated number of naevi, the assessment of skin phototype and the perception of increased melanoma risk were 0.34 [95% confidence interval (CI) 0.30–0.37], 0.28 (95%CI 0.24–0.32) and 0.24 (95% CI 0.18–0.30), respectively. Kappa agreements below 0.40 are considered poor. Stratification by age and gender revealed slightly higher scores for subgroups on single items.

Conclusions

Screening for melanoma should be population based and should not be limited to self-referrals since self-assessment for melanoma risk factors is inaccurate. Educational programmes must be developed to improve self-assessment and to target populations at risk.

Introduction

Melanoma is a common cutaneous malignancy with considerable mortality, which is potentially preventable or at least identifiable at an early stage with excellent prognosis.1, 2, 3, 4, 5, 6 Primary prevention programmes, such as teaching the population about melanoma risk factors including skin phototype, sunburn in childhood and intense intermittent sun exposure, are controversial.1, 2, 7, 8 Programmes to reduce the incidence of melanoma require decades to achieve their full potential, must be repeated frequently to be effective and are expensive.1, 7, 8 Widespread public health programmes, such as those used in Australia for the past 20 years and Austria for more than 15 years, have increased public knowledge about melanoma, altered attitudes about sun exposure and tanning, and led to a change in general behaviour.2, 4, 5

Extended educational programmes with instructions on individual risk factors, such as family history, number of naevi, presence of dysplastic (atypical) naevi and changing naevi, which encourage self-assessment and then self-referral carry benefits as well as risks. On one hand, education has been associated with thinner and less invasive melanomas,3, 9, 10 and self-examination has been associated with a reduced risk of advanced disease.7 On the other hand, self-assessment was only moderately accurate in identifying individuals at high risk for melanoma, and corresponded imperfectly with clinical assessment, especially in people at highest risk for melanoma.11, 12, 13 Furthermore, only 15% of patients performed skin self-examinations, and the frequency of skin examinations decreased with age.7

Secondary prevention programmes with screening also have mixed effects. Free skin examinations have led to early detection of melanoma in people who would not have obtained a skin examination without the programme. While specialized examination of pigmented skin lesions was associated with a decreased risk of missing melanomas,7, 14, 15, 16 no study has shown definitively that unfocused screening for skin cancer is cost-effective and scientifically worthy.7, 8, 17, 18 Additionally, only a few melanomas are detected during such screening programmes,19, 20 total skin examinations do not increase the melanoma detection rate,21 and certain groups of individuals, especially men aged >50 years and elderly people, are under-represented in those programmes.9, 22, 23

The aim of this study was to shed light on these controversies. How accurate is self-assessment? Is it a valid instrument to reach populations at higher risk for melanoma? During a melanoma screening campaign in Styria, a province in south-eastern Austria, agreement between self-assessment and a dermatologist's evaluation of the estimated number of naevi, skin phototype and perceived melanoma risk was evaluated.

Section snippets

Screening procedure

The screening campaign was performed on five consecutive Saturdays in summer. Each day, a team of trained dermatologists visited a randomly selected open-air recreation facility in Styria. Swimmers and sunbathers were invited by leaflets and loudspeaker announcements to participate in a free medical assessment for melanoma.

Questionnaire

After specially trained assistants explained the study, 1223 volunteers, all Caucasians, gave informed oral consent and then answered an anonymous comprehensive questionnaire

Characteristics of the study population

In total, 1223 people participated in the study: 529 men (mean age 34.7 years, range 1–87) and 694 women (mean age 36.5 years, range 1–85). The number of subjects <9 (n = 147) and >69 years of age (n = 72) was too small to be included into the stratified analyses.

Self-assessments of risk factors

The majority of subjects (35.7%) estimated that they had between one and 10 naevi, followed by 25.9% who estimated 11–20 naevi and 21.6% who estimated 21–50 naevi. Regarding skin phototype, 50.4% of the subjects assessed themselves as type

Discussion

The only option to reduce melanoma mortality in the short term is early diagnosis, although it is hoped that primary prevention campaigns may reduce the incidence of melanoma in the long term.8 Responsibility for early detection of melanoma rests with the individual, with early diagnosis and prompt treatment depending on early self-referral by people aware of their risk.6

This study evaluated self-assessment of skin phototype and number of naevi; two of the most important risk factors for

Acknowledgements

The authors wish to thank the following dermatologists of the Department of Dermatology, Medical University of Graz, Austria for performing the clinical examinations during the screening campaign: A. Gerger, MD; M. Gruber, MD; P. Kahofer, MD; S. Koller, MD; A. Okcu, MD; H. Reiter, MD; A. Wackernagel, MD; I.H. Wolf, MD; and I. Zalaudek, MD. The authors also wish to thank Walter H.C. Burgdorf for his critical review and editorial assistance.

References (30)

  • J. Aitken et al.

    Population screening for melanoma: current evidence and a community based randomized trial

  • C. Garbe et al.

    Time trends of cutaneous melanoma in Queensland, Australia and Central Europe

    Cancer

    (2000)
  • T.L. Diepgen et al.

    The epidemiology of skin cancer

    Br J Dermatol

    (2002)
  • A. Harbauer et al.

    Validity of an unsupervised self-administered questionnaire for self-assessment of melanoma risk

    Melanoma Res

    (2003)
  • P. Carli et al.

    Melanoma detection rate and concordance between self-skin examination and clinical evaluation in patients attending a pigmented lesion clinic in Italy

    Br J Dermatol

    (2002)

    • Cited by (10)

    • Factors associated with suspected nonmelanoma skin cancers, dysplastic nevus, and cutaneous melanoma among first-time SpotMe screening program participants during 2009-2010

      2023, Journal of the American Academy of Dermatology
      Citation Excerpt :

      Although the validity of self-assessed versus physician-assessed risk factors for melanoma has previously been demonstrated,21 to our knowledge, no such comparable studies have been conducted for NMSCs and DNs. Investigations have demonstrated that personal history of skin cancer22 and prior skin cancer examination history23 self-reports are accurate, and other investigations suggest there can be significant disagreement between physician and patient reports of factors such as mole counts, skin phenotype, family history of skin cancer, and perceived melanoma risk.24-29 One major limitation is the lack of histologic confirmation for the screening diagnoses, which might be particularly relevant for suspected CM and DN analyses, as the ability to distinguish between a DN and CM can be particularly challenging even with known histology.30-32

    • Can People Correctly Assess their Future Risk of Melanoma?

      2021, Journal of Investigative Dermatology

    • The relationship between skin checking and sun protection behaviours: implications for skin cancer prevention campaigns

      2018, Public Health

    • Screening, early detection, education, and trends for melanoma: Current status (2007-2013) and future directions Part I. Epidemiology, high-risk groups, clinical strategies, and diagnostic technology

      2014, Journal of the American Academy of Dermatology
      Citation Excerpt :

      In addition, images of atypical moles and melanomas may play a role in training the public to recognize clinical warning signs of melanoma.42,43 Self-assessment for melanoma risk factors has been shown to be inaccurate,44 and evidence-based assessment tools are necessary to identify individuals at high risk for melanoma. Several risk assessment tools have been used, although no available models target those at highest risk of developing lethal melanoma.

    • Risk stratification for melanoma: Models derived and validated in a purpose-designed prospective cohort

      2018, Journal of the National Cancer Institute

    • Accuracy of self-reported nevus and pigmentation phenotype compared with clinical assessment in a population-based study of young Australian Adults

      2015, Cancer Epidemiology Biomarkers and Prevention
    View all citing articles on Scopus
    View full text
    Copyright © 2008 The Royal Institute of Public Health. Published by Elsevier Ltd. All rights reserved.