[Skip to Content]
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
Purchase Options:
[Skip to Content Landing]
August 16, 2000

Early Detection of Melanoma

Author Affiliations

Author Affiliation: Dermatoepidemiology Unit, VA Medical Center Providence, and Department of Dermatology, Rhode Island Hospital and Brown University, Providence.


Controversies Section Editor: Phil B. Fontanarosa, MD, Executive Deputy Editor.

JAMA. 2000;284(7):886-889. doi:10.1001/jama.284.7.886

Melanoma has been increasing in incidence and mortality in recent decades and represents a substantial public health problem. Because melanoma trends tend to follow a cohort pattern, deaths from melanoma occur at a younger age than most other cancers, and melanoma is among the most common sites of cancer in young adults.1 Early detection is feasible because melanoma is usually visible on the skin surface when in a curable stage. Invasive melanomas excised when less than 0.76 mm in Breslow thickness are about 96% curable.2 However, if melanoma is not removed before it reaches 3.6 mm in thickness, case fatality is greater than 70%.2 Death due to melanoma is a tragedy that should not be occurring so frequently.

Admittedly, there is no reliable and precise measure of the magnitude of survival benefit that can be achieved by early detection; no randomized trials have reported on its effectiveness. Nevertheless, clinical experience, expert opinion, and knowledge of the stages of melanoma progression indicate that the potential to substantially reduce melanoma mortality is huge. Furthermore, reports from populations subject to intensive early detection campaigns have associated those campaigns with reduced mortality3 and suggest that they are cost-effective.4 These indications are supported by the trend of improved survival among melanoma patients in recent decades, presumably due in large part to early detection efforts.

Procedure for Early Detection: Skin Inspection

The vast majority of melanomas that have not already metastasized are completely asymptomatic, although visible on the skin surface to the naked eye. Early detection requires inspection of the skin. The skin is already being examined by dermatologists, by primary care physicians, by other professionals who provide primary care, and by patients and their families. The question is not whether it is worthwhile for anyone to look at the skin. Rather, the question is whether and how these skin examinations can and should be improved for the purpose of reducing mortality from melanoma.

The Role of Dermatologists

Dermatologists have the greatest expertise in diagnosis and treatment of skin lesions and have developed their skills during 3 years of residency training and subsequent experience in practice. For high-risk patients, dermatologists are providing cutaneous examinations and are making increasing use of tools, such as photography and in vivo epiluminescence microscopy, which aid in the early detection of melanoma by those with appropriate training.5,6 High-risk patients, such as those with multiple dysplastic nevi, a history of melanoma, or a significant family history of melanoma, frequently obtain periodic dermatologic surveillance. These surveillance programs generally involve periodic dermatologic examination as well as patient and family education about warning signs and skin self-examination. Studies of patients in these programs have noted that melanomas occurring in enrolled patients are substantially thinner and hence of better prognosis than the melanomas that initially brought the patient into the program.7-9 Despite the absence of randomized trials, participation in these programs has been recommended for high-risk individuals.10,11

Most melanomas, however, do not occur in high-risk individuals, so early detection efforts restricted to them will have a limited impact on the overall public health problem. Published evidence suggests the cost-effectiveness of dermatologic screening of broad demographic subpopulations at increased risk,12 but it remains unclear that advocacy of visits conducted purely for screening purposes among the general population (not at high risk) is the most cost-effective approach. Since no one is identifiably immune from melanoma, early detection efforts need to be broadly inclusive for maximum impact.

The role of the dermatologist is limited by several factors. Dermatology is a small specialty, so its ability to affect the entire population by direct skin inspection is modest. A substantial proportion of dermatologists do not perform full body skin examinations on many of their patients, which limits their impact on the melanoma problem in low- and moderate-risk populations.13 Furthermore, even many high-risk patients are not enrolled in surveillance programs. And, despite their training and experience, the predictive value of a positive screen for melanoma by dermatologists is modest, even in relatively high prevalence settings,14-17 although accuracy in diagnosis of melanoma improves with increasing years of experience after dermatologic training.18-20

Dermatologists have a critical role to play, but substantial progress on the overall public health problem requires that others assume responsibility for early detection of skin cancer by examining the skin.

The Role of Patients and Their Families

Patients and their families should and do bear substantial responsibility for their own health, including early detection of skin cancer. In general, no one has more intimate knowledge of a patient's skin or a more frequent opportunity to inspect the skin than the patient and his or her spouse, significant other, or close family member. Patients and their families are the ones who bear the ultimate consequences of failure to detect melanoma promptly. Assuming some responsibility for early detection also may confer a sense of empowerment that may mitigate the anxiety associated with melanoma risk. The good news is that patients and their families are already the ones who usually first recognize melanoma.21 Nevertheless, a great part—if not most—of the delay between awareness of the initial signs of the melanoma and its ultimate excision occurs prior to the first physician visit for the lesion.22-26

In addition to access to appropriate health care and knowledge of the warning signs of melanoma, performance of thorough skin self-examination (TSSE) is key to early detection of melanoma by patients and their families. Regular (typically monthly) TSSE has been advocated by a variety of organizations concerned with skin cancer, such as the American Cancer Society, the American Academy of Dermatology, and the Skin Cancer Foundation. However, TSSE has been subject to only limited formal study. A case-control study estimated that monthly skin self-examination was associated with a 63% reduction in melanoma mortality.27 In the context of existing knowledge of melanoma, this suggests the importance of an active role for patients and their families in skin inspection for the early detection of melanoma. Regrettably, the skin inspection that people perform is generally not systematic and thorough. A recent random-digit dial survey suggested that while more than half (59%) of those contacted reported deliberately and systematically examining their skin, only 9% did so thoroughly on closer questioning.28 Furthermore, public knowledge about melanoma remains limited despite increased publicity in recent years.29

One constraint on both research efforts and advocacy in this area is the limited evidence of the efficacy, effectiveness, and cost-effectiveness of TSSE for reducing melanoma mortality. Organizations such as the US Preventive Services Task Force have viewed the evidence for TSSE efficacy as inadequate support for its advocacy.10 Indeed, TSSE has not been subject to a randomized trial with melanoma mortality as an outcome and is unlikely to be so evaluated in the future. Yet, principles of evidence-based medicine must be interpreted with judgment here. The practice of TSSE that an individual performs on oneself, by oneself, without cost or substantial risk, should not be viewed as a screening test requiring the level of evidence demanded of procedures such as mammography or sigmoidoscopy; rather, it is simply a healthy personal habit for which advocacy requires a lower evidentiary standard. As a person who has knowledge of the warning signs of melanoma, I wonder: when I wake up in the morning, trudge to the bathroom to stand in front of a mirror, crack open that first eyelid, and inspect the skin of my face, am I screening myself for skin cancer? What if, after I shower and now (more fully awake) make a thorough and careful inspection of all of the spots on my skin? If the latter is screening, and the former is not, then should we advocate casual viewing of one's skin with minimal evidentiary support, but require a randomized trial to support advocacy of being careful and systematic? Surely not.

Patients and their families have important roles to play through awareness of warning signs and performance of TSSE. Early detection of melanoma requires looking at the skin; for most of the population, no one is currently doing that in a thorough and systematic manner. Lack of randomized trials should not paralyze research on early detection or advocacy of TSSE by the general public.

The Role of Primary Care Clinicians

Primary care clinicians already assume responsibility for many cancer early detection activities and consider early detection an important aspect of their care of patients. For those individuals for whom no one is examining their skin, the primary care clinician is the health care professional who is left to assume this responsibility, along with the many other health maintenance tasks he or she must oversee.

Many organizations explicitly recommend skin examinations as part of a cancer-related check-up, including the American Cancer Society, the American Academy of Dermatology, and the National Institutes of Health Consensus Panel.30 Complete skin examination adds little to the time needed for an office visit if the patient is appropriately prepared by office staff. The US Preventive Services Task Force has not recommended that primary care clinicians conduct specific skin cancer screening examinations for their entire patient population because of the paucity of rigorous published studies. However, the Task Force does recommend that "clinicians remain alert for skin lesions with malignant features when examining patients for other reasons."10 Since Americans make an average of 1.7 visits to primary care physicians each year,31 detection of melanoma by primary care practitioners has tremendous potential for reducing melanoma mortality without substantially interfering with other primary care activities.

In Australia, where melanoma risk is several-fold higher than in the United States, published analyses have suggested that screening by primary care clinicians would be cost-effective if applied to the population older than 50 years.32,33 We do not know the cost-effectiveness of this approach in the United States, and indeed we cannot evaluate the ultimate potential of this approach when the primary care clinicians being evaluated generally do not have training appropriate for optimal skin inspection and skin cancer triage.

Early detection by primary care clinicians does face significant barriers.34 First, these practitioners often have limited skills in the recognition of melanoma.35,36 This may be due in part to inadequate training about this task in medical school and residency, and in part to the fact that melanoma is uncommon in primary care practice. There are approximately 48,000 invasive melanomas diagnosed each year in the United States,37 and about 267,000 primary care physicians.38 Hence, even if no one other than a primary care physician were ever to diagnose a melanoma, the typical primary care physician would be unlikely to diagnose more than 1 or 2 in a decade of full-time practice. Furthermore, other physicians do diagnose melanoma; most office visits for melanoma in the United States are to dermatologists.39 The less melanoma is seen in routine primary care practice, the more difficult it is for the physician to maintain the visual pattern recognition skills needed for early diagnosis.

Moreover, primary care clinicians lack confidence in their ability to diagnose melanoma.40,41 This lack of confidence is recognized as a major barrier to the conduct of early detection activities, and 1 consequence is that primary care clinicians typically do not carefully examine the skin.42 In a recent survey, most patients reported that their primary care clinician rarely or never examined the skin of their back (the most common site of melanoma among men) or of their legs (the most common site among women).28

While time constraints in primary care practice are not about to disappear, the barriers of skills and confidence are not an unalterable feature of the medical landscape. Although melanoma may be infrequent in primary care practice, concern about skin cancer is common. Although skills in melanoma diagnosis are limited, they can be significantly improved.43-45 For example, a brief (2-hour) curriculum that focuses on the use of an 8-step algorithm to appropriately triage skin lesions and patients into categories of "Act," "Reassure," or "Track" has been developed.46 Preliminary evaluation with primary care physicians suggests that this curriculum improves both confidence and skills in early detection and may result in increased performance of skin cancer counseling and skin examination in routine practice47-49 (M.A.W., unpublished data, 2000).

Primary care provides an important opportunity for improvement in skin inspection for the early detection of melanoma, although its full potential has not yet been realized.


As noted at the outset, the value of early detection, ie, skin inspection by health professionals or the general public themselves, has not been quantified with randomized trials. In Queensland, Australia, which has the highest recorded incidence of melanoma of any large population, an effort is under way to conduct just such an evaluation of multicomponent community-based early detection efforts that include TSSE, whole body examination by primary care clinicians, and specialized screening clinics. Assuming this evaluation proceeds according to plan, results for the mortality end point from this unique and critically important investigation are expected to be complete in the year 2015.

It has been asserted that "no one should die of malignant melanoma."50 That may be an unrealistic goal, but the limits of the benefit to be derived from early detection, or the cost-effectiveness of the various approaches have yet to be determined. Dermatologists, patients and their families, and primary care clinicians and staff all have the potential and the responsibility to contribute to the reduction of melanoma mortality. Let's do it.

 SEER Cancer Incidence Public Use Database, 1973-1996 [CD-ROM]. Bethesda, Md: National Cancer Institute; 1999.
Schuchter L, Schultz DJ, Synnestvedt M.  et al.  A prognostic model for predicting 10-year survival in patients with primary melanoma.  Ann Intern Med.1996;125:369-375.Google Scholar
MacKie RM, Hole D. Audit of public education campaign to encourage earlier detection of malignant melanoma.  BMJ.1992;304:1012-1015.Google Scholar
Cristofolini M, Bianchi R, Boi S.  et al.  Analysis of the cost-effectiveness ratio of the health campaign for the early diagnosis of cutaneous melanoma in Trentino, Italy.  Cancer.1993;71:370-374.Google Scholar
Binder M, Schwarz M, Winkler A.  et al.  Epiluminescence microscopy: a useful tool for the diagnosis of pigmented skin lesions for formally trained dermatologists.  Arch Dermatol.1995;131:286-291.Google Scholar
Kelly JW, Yeatman JM, Regalia C, Mason G, Henham AP. A high incidence of melanoma found in patients with multiple dysplastic naevi using photographic surveillance.  Med J Aust.1997;167:191-194.Google Scholar
Vasen HFA, Bergman W, van Haeringen A, Scheffer E, van Slooten EA. The familial dysplastic nevus syndrome: natural history and the impact of screening on prognosis: a study of nine families in the Netherlands.  Eur J Cancer Clin Oncol.1989;25:337-341.Google Scholar
Masri GD, Clark WH, Guerry D, Halpern A, Thompson CJ, Elder DE. Screening and surveillance of patients at high risk for malignant melanoma result in detection of earlier disease.  J Am Acad Dermatol.1990;22:1042-1048.Google Scholar
Tucker MA, Fraser MC, Goldstein AM, Elder DE, Guerry D, Organic SM. Risk of melanoma and other cancers in melanoma-prone families.  J Invest Dermatol.1993;100:350S-355S.Google Scholar
US Preventive Services Task Force.  Guide to Clinical Preventive Services. 2nd ed. Baltimore, Md: Williams & Wilkins; 1996:141-152.
Hill L, Ferrini RL. Skin cancer prevention and screening: summary of the American College of Preventive Medicine's practice policy statements.  CA Cancer J Clin.1998;48:232-235.Google Scholar
Freedberg KA, Geller AC, Miller DR, Lew RA, Koh HK. Screening for malignant melanoma: a cost-effectiveness analysis.  J Am Acad Dermatol.1999;41:738-745.Google Scholar
Polster AM, Lasek RJ, Quinn LM, Chren M-M. Reports by patients and dermatologists of skin cancer preventive services provided in dermatology offices.  Arch Dermatol.1998;134:1095-1098.Google Scholar
Koh HK, Caruso A, Gage I.  et al.  Evaluation of melanoma/skin cancer screening in Massachusetts: preliminary results.  Cancer.1990;65:375-379.Google Scholar
Koh HK, Norton LA, Geller AC.  et al.  Evaluation of the American Academy of Dermatology's National Skin Cancer Early Detection and Screening Program.  J Am Acad Dermatol.1996;34:971-978.Google Scholar
Jonna BP, Delfino RJ, Newman WG, Tope WD. Positive predictive value for presumptive diagnoses of skin cancer and compliance with follow-up among patients attending a community screening program.  Prev Med.1998;27:611-616.Google Scholar
Engelberg D, Gallagher RP, Rivers JK. Follow-up and evaluation of skin cancer screening in British Columbia.  J Am Acad Dermatol.1999;41:37-42.Google Scholar
Lindelof B, Hedblad MA. Accuracy in the clinical diagnosis and pattern of malignant melanoma at a dermatological clinic.  J Dermatol.1994;21:461-464.Google Scholar
MacKenzie-Wood AR, Milton GW, deLauney JW. Melanoma: accuracy of clinical diagnosis.  Australas J Dermatol.1998;39:31-33.Google Scholar
Morton CA, Mackie RM. Clinical accuracy of the diagnosis of cutaneous malignant melanoma.  Br J Dermatol.1998;138:283-287.Google Scholar
Koh HK, Miller DR, Geller AC, Clapp RW, Mercer MB, Lew RA. Who discovers melanoma? patterns from a population based survey.  J Am Acad Dermatol.1992;26:914-919.Google Scholar
Elwood JM, Gallagher RP. The first signs and symptoms of melanoma: a population-based study.  Pigment Cell.1988;9:118-130.Google Scholar
Cassileth BR, Temoshok L, Frederick BE.  et al.  Patient and physician delay in melanoma diagnosis.  J Am Acad Dermatol.1988;18:591-598.Google Scholar
Krige JEJ, Isaacs S, Hudson DA, King HS, Strover RM, Johnson CA. Delay in the diagnosis of cutaneous malignant melanoma: a prospective study of 250 cases.  Cancer.1991;68:2064-2068.Google Scholar
Blum A, Brand CU, Ellwanger U.  et al.  Awareness and early detection of cutaneous melanoma: an analysis of factors related to delay in treatment.  Br J Dermatol.1999;141:783-787.Google Scholar
Richard MA, Grob JJ, Avril MF.  et al.  Melanoma and tumor thickness: challenges of early diagnosis.  Arch Dermatol.1999;135:269-274.Google Scholar
Berwick M, Begg CB, Fine JA, Roush GC, Barnhill RL. Screening for cutaneous melanoma by skin self-examination.  J Natl Cancer Inst.1996;88:17-23.Google Scholar
Weinstock MA, Martin RA, Risica PM.  et al.  Thorough skin examination for the early detection of melanoma.  Am J Prev Med.1999;17:169-175.Google Scholar
Miller DR, Geller AC, Wyatt SW.  et al.  Melanoma awareness and self-examination practices: results of a United States survey.  J Am Acad Dermatol.1996;34:962-970.Google Scholar
Consensus Development Panel.  Diagnosis and treatment of early melanoma.  NIH Consens Statement.1992;10:1-26.Google Scholar
Nelson C, Woodwell D. National Ambulatory Medical Care Survey: 1993 summary.  Vital Health Stat 13.1998;136:iii-iv, 1-99.Google Scholar
Girgis A, Clarke P, Burton RC, Sanson-Fisher RW. Screening for melanoma by primary health care physicians: a cost-effectiveness analysis.  J Med Screen.1996;3:47-53.Google Scholar
Burton RC, Howe C, Adamson L.  et al.  General practitioner screening for melanoma: sensitivity, specificity, and effect of training.  J Med Screen.1998;5:156-161.Google Scholar
Wender RC. Barriers to effective skin cancer education.  Cancer.1995;75:691-698.Google Scholar
Cassileth BR, Clark WH, Lusk EJ, Frederick BE, Thompson CJ, Walsh WP. How well do physicians recognize melanoma and other problem lesions?  J Am Acad Dermatol.1986;14:555-560.Google Scholar
Gerbert B, Maurer T, Berger T.  et al.  Primary care physicians as gatekeepers in managed care: primary care physicians' and dermatologists' skills at secondary prevention of skin cancer.  Arch Dermatol.1996;132:1030-1038.Google Scholar
Landis SH, Murray T, Bolden S, Wingo PA. Cancer statistics, 2000.  CA Cancer J Clin.2000;50:1-12.Google Scholar
 Physician Characteristics and Distribution in the US, 2001-2002 Edition . Chicago, Ill: American Medical Association. In press.
Smith ES, Feldman SR, Fleischer AB, Leshin B, McMichael A. Characteristics of office-based visits for skin cancer: dermatologists have more experience than other physicians in managing malignant and pre-malignant skin conditions.  Dermatol Surg.1998;24:981-985.Google Scholar
Girgis A, Sanson-Fisher RW. Skin cancer prevention, early detection, and management: current beliefs and practices of Australian family physicians.  Cancer Detect Prev.1996;20:316-324.Google Scholar
Stephenson A, From L, Cohen A, Tipping J. Family physicians' knowledge of malignant melanoma.  J Am Acad Dermatol.1997;37:953-957.Google Scholar
Federman DG, Concato J, Caralis PV, Hunkele GE, Kirsner RS. Screening for skin cancer in primary care settings.  Arch Dermatol.1997;133:1423-1425.Google Scholar
Girgis A, Sanson-Fisher RW, Howe C, Raffan B. A skin cancer training programme: evaluation of a post-graduate training for family doctors.  Med Educ.1995;29:364-371.Google Scholar
Laidlaw JM, Harden RM, Morris AM. Continuing medical education case study series: an innovative programme for general practitioners on malignant melanoma.  Med Educ.1996;30:226-231.Google Scholar
Gerbert B, Bronstone A, Wolff M.  et al.  Improving primary care residents' proficiency in the diagnosis of skin cancer.  J Gen Intern Med.1998;13:91-97.Google Scholar
Weinstock MA, Goldstein MG, Dubé CE, Rhodes AR, Sober AJ. Basic Skin Cancer Triage for teaching melanoma detection.  J Am Acad Dermatol.1996;34:1063-1066.Google Scholar
Weinstock MA, Goldstein MG, Dubé CE, Rossi JS. Impact of training in Basic Skin Cancer Triage on skills and confidence in skin cancer detection [abstract].  J Invest Dermatol.1999;112:664.Google Scholar
Mikkilineni R, Weinstock MA, Goldstein MG, Dubé CE, Rossi JS. Impact of Basic Skin Cancer Triage (BSCT) training on providers' attitudes towards skin cancer control [abstract].  J Invest Dermatol.2000;114:775.Google Scholar
Mikkilineni R, Weinstock MA, Goldstein MG, Dubé CE, Rossi JS. Impact of Basic Skin Cancer Triage (BSCT) training on providers' skin cancer control practices [abstract].  J Invest Dermatol.2000;114:888.Google Scholar
Ackerman AB. No one should die of malignant melanoma.  J Am Acad Dermatol.1985;12:115-116.Google Scholar