Early detection of melanoma: an ounce of prevention may be a ton of work

Risk factors for presumptive melanoma in skin cancer screening: American Academy of Dermatology National Melanoma/Skin Cancer Screening Program experience 2001-2005

BACKGROUND

Since its inception in 1985, the American Academy of Dermatology (AAD) National Melanoma/Skin Cancer Screening Program has strived to enhance early detection of cutaneous malignant melanoma (MM) by providing nationwide skin cancer education campaigns in combination with free skin cancer screenings.

OBJECTIVE

To analyze the AAD screening data from 2001 to 2005 in order to identify factors associated with MM detection, and thereby derive a model of increased likelihood for MM detection through visual skin examinations at screenings.

MATERIALS AND METHODS

Patients completed a standardized AAD pre-screening form with historical and phenotypic information. Clinicians then recorded suspected clinical findings noted at visual skin examination. Statistical analyses were conducted using SPSS 14 (SPSS Inc., Chicago, Ill).

RESULTS

Five factors, which can be remembered with the acronym HARMM, independently increased the likelihood of suspected MM being found in the 362,804 persons screened: History of previous melanoma (odds ratio [OR] = 3.3; 95% confidence interval [CI], 2.9-3.8); Age over 50 (OR = 1.2; 95% CI, 1.1-1.3); Regular dermatologist absent (OR = 1.4; 95% CI, 1.3-1.5); Mole changing (OR = 2.0; 95% CI, 1.9-2.2); and Male gender (OR = 1.4; 95% CI, 1.3-1.5). Individuals at highest risk (4 or 5 factors) comprised only 5.8% of the total population, yet accounted for 13.6% of presumptive MM findings, and were 4.4 times (95% CI, 3.8-5.1) more likely to be diagnosed with suspected MM than individuals at lowest risk (0 or 1 factor). Receipt of a total skin examination at screening independently increased the likelihood for identifying suspected MM (OR = 1.4; 95% CI, 1.3-1.6). However, significantly fewer screenees in the highest risk group versus those in the lowest risk group underwent total skin examinations (53.7% vs 62.5%).

LIMITATIONS

Risk factors studied limited to variables collected in screenee enrollment form.

CONCLUSIONS

A higher-risk subgroup of the skin cancer screening population can be identified through assessment of MM risk factors using the HARMM criteria. Refocusing efforts to provide a total skin examination to those individuals with multiple risk factors has the potential to both reduce costs and increase yields for suspected MM in future mass screening initiatives.

Día del Euromelanoma. Resultados en España de las campañas de 2000, 2001 y 2002

BACKGROUND

In the year 2000, the European Academy of Dermatology and Venereology initiated a campaign, called "Euromelanoma Day," for the early detection of melanoma.

OBJECTIVE

To discuss the results of "Euromelanoma Day" in Spain for the campaigns carried out in 2000, 2001 and 2002.

MATERIAL AND METHODS

Participating members of the Spanish Academy of Dermatology examined everyone who requested an appointment, free of charge, in order to detect suspicious lesions. Four months after the visit, telephone contact was made with all patients who had been diagnosed with suspected melanoma to ask them whether the lesions had been excised, and if so, the result of the histological study of the specimen.

RESULTS

33,750 calls to set up an appointment were received, and 12,487 patients were examined. The average participation by dermatologists was 399 academy members per year. A total of 164 lesions suspected of being melanoma were detected, above all in Andalusia and Catalonia, and the diagnosis was histologically confirmed in 31 cases. The clinicopathological correlation was 23.3 %, a higher figure than the one for the USA and similar to the one for Mediterranean countries. The average thickness of the tumors was 0.93 mm.

CONCLUSIONS

Although these campaigns have the drawback of possible bias in the findings and may cause some degree of public alarm, the benefit to the population was unquestionable in terms of heightening awareness of a health problem.

Screening for malignant melanoma: A cost-effectiveness analysis

BACKGROUND

Skin cancer is the most common cancer in the United States. Increasing evidence suggests that screening for malignant melanoma is effective, but its cost-effectiveness has not been determined.

OBJECTIVE

We attempted to determine the effectiveness and costs of a visual screen to diagnose malignant melanoma in high-risk persons.

METHODS

We developed a decision analysis comparing no skin cancer screen with a single screen by a dermatologist. Clinical outcomes included malignant melanoma, nonmelanoma skin cancer, or no skin cancer. Life expectancy and costs of care were projected on the basis of clinical findings.

RESULTS

Skin cancer screening increased average discounted life expectancy from 15.0963 years to 15.0975 years. Based on the prevalence of malignant melanoma, however, this translates into an increased discounted life expectancy of 0.9231 years for each person with diagnosed melanoma. Using a cost of $30 per screen, total skin cancer-related costs for a cohort of 1 million people increased from $826 million with no screen to $861 million with screening, with an increase of 1200 years of life. This results in an incremental cost-effectiveness ratio of $29,170 per year of life saved (YLS) with screening. Sensitivity analysis showed that the cost-effectiveness ratio for screening remained below $50,000/YLS if the prevalence of melanoma in the screened population was at least 0. 0009, the probability that a melanoma detected in screening was localized was at least 94.8%, or the cost of each screen was below $57.

CONCLUSION

Skin cancer screening in high-risk patients is likely to be associated with a small increase in discounted life expectancy and is reasonably cost-effective compared with other cancer screening strategies.

Follow-up and evaluation of skin cancer screening in British Columbia

BACKGROUND

Skin cancer screening is thought to be a useful public health tool for the early detection of skin cancers. However, few studies have reported on follow-up and outcome of subjects who have a positive screen.

OBJECTIVE

The aims of this study were to evaluate attendance at skin cancer screening clinics in British Columbia for the period 1994 and 1995 and to assess follow-up outcome among participants who were identified to have a potentially serious skin lesion that warranted further medical review.

METHODS

A self-administered questionnaire was sent to participants screening positive for skin cancer and to their attending physicians.

RESULTS

Five hundred twenty people were screened. Of these, 105 were referred for evaluation of a potential malignancy or precursor lesion. One melanoma, 3 basal cell carcinomas, 4 atypical nevi, and 1 actinic keratosis were histologically confirmed in 76 referred participants for whom follow-up information was available. The positive predictive values ranged from 17% to 89% depending on the screening diagnosis. Several false-positive results and one false-negative result were observed. Reasons for not seeking recommended follow-up were addressed.

CONCLUSIONS

Our yield and positive predictive values for different screening diagnoses were virtually identical to those previously reported in larger US studies. Improved communication between screening physicians and screening participants may improve follow-up rates in those people who would benefit from further medical care.

The role of screening in the management of skin cancer

The incidence of skin cancer has been rising at an alarming rate for the past several years. This poses a significant public health problem in the United States. Detection and treatment of melanoma early in its course is critical for improved outcome. Of the approaches to cancer control that can reduce mortality from melanoma and nonmelanoma skin cancer, screening holds the greatest promise for a rapid and major impact. Prevention and early detection are crucial in reducing morbidity and mortality from skin cancer. For a number of reasons, however, the full effect of screening for both melanoma and nonmelanoma skin cancers has not been achieved. Controversy exists regarding who should perform screening, who should be screened, and whether screening should be performed at all. It is clear that melanoma and nonmelanoma skin cancer control programs combining primary prevention, education, and screening are in developmental stages. This review will discuss the advantages and disadvantages of screening for skin cancer.

Cutaneous melanoma: patient surveillance and tumor progression

BACKGROUND

The goal of surveillance examinations (SVE) in patients at high risk for cutaneous melanoma (CM) is to diagnose premetastatic tumors.

OBJECTIVE

We attempted to determine whether SVE before CM diagnosis influences the histopathologic features of tumor progression.

METHODS

A retrospective review was conducted of consecutive patients who presented with an intact primary CM during a 36 month period (1991 to 1994) in a university dermatology ambulatory setting.

RESULTS

Thirty-nine intact CMs were diagnosed in 34 patients. Compared with 23 CMs in as many patients presenting at the first encounter, 16 CMs diagnosed in 11 patients during SVE had a smaller mean tumor diameter (P = .007) and tumor thickness (P = .002) and were more likely anatomic level I or II (P = .003) with microscopic thickness less than 0.76 mm (P = .01) and less than 0.50 mm (P = .002).

CONCLUSION

Primary CMs are more likely to be smaller and thinner when diagnosed during SVE than at first encounter, features that are likely to have a positive impact on CM mortality.

Public health interventions for melanoma. Prevention, early detection, and education

Worldwide melanoma control programs that include some combination of primary prevention, education, and screening activities have only recently begun to undergo an evaluation process. More studies with rigorous design and evaluation are needed. Until then, the proper public health policy guidelines for melanoma control, especially screening, are open to debate. Future studies must determine how screening, early detection, case finding, and education can best be used to reduce mortality and achieve optimal melanoma control.

Are young adults checking their skin for melanoma?

This study examined the prevalence and predictors of self screening for melanoma in a large sample of young New Zealanders. A self-report questionnaire was administered to a sample of 909, 21-year-olds to investigate if young adults check their skin for changes in lesions which could be melanoma, and to identify the factors which influence this behaviour and any subsequent help seeking. Fifty-three per cent reported checking their skin in the past year, with 20% noticing a change in a mole or freckle. Forty-five per cent of those who noticed a change sought medical advice. The most common reason for not seeking advice was cost. Women were more likely that men to have checked their skin, to have noticed a change and to have sought medical advice. In addition to gender, tendency to self check was also associated with knowledge of melanoma and perceived risk of melanoma. These results are discussed in light of the current debate regarding skin cancer screening. This study fills a gap in the literature regarding self screening for melanoma in young adults and identifies ways in which future prevention campaigns might be modified to address the concerns of this age group.

Basal cell carcinoma

Basal cell carcinoma is the commonest malignancy in Caucasians with incidence rates of 300 per 100,000 reported in the USA. Rates are increasing at over 10% per year leading to a lifetime risk of 30%. Although mortality is low, the disease is responsible for considerable morbidity and places a substantial burden on health service provision in the UK. Furthermore, lesions may recur and patients often develop multiple tumours giving major implications for treatment and follow-up. Four main types of basal cell carcinoma are seen: nodulo-ulcerative; pigmented; morpheaform and superficial. Diagnosis is by histological evaluation although many tumours have a characteristic clinical appearance. The differential diagnosis is large. Identified risk factors include male gender, skin type 1, red/blonde hair and increasing age. Patients with basal cell carcinoma are more likely to develop malignant melanoma and squamous cell carcinoma but it is still unclear whether there is a link with internal malignancy. The main treatment modalities are surgery and radiotherapy. Each has advantages and disadvantages. The choice of treatment depends on many factors. Principles of treatment include identification of high-risk patients to enable early detection, complete removal of the lesion, and careful follow-up to detect recurrence or new lesions. Approximately 10% of tumours recur, depending on site, size and treatment modality. Metastatic basal cell carcinoma and the association of ultraviolet radiation to basal cell carcinoma risk are reviewed.

Full proceedings from the National Conference to Develop a National Skin Cancer Agenda. American Academy of Dermatology and Centers for Disease Control and Prevention, Washington, D.C., April 8-10, 1995

National efforts to reduce skin cancer incidence and mortality require scientifically coordinated efforts. This report summarizes the first American Academy of Dermatology/Centers for Disease Control and Prevention national conference to develop a skin cancer agenda. Leading experts in dermatology, public health, medicine, health education, nursing, behavioral sciences, environmental health and epidemiology identified and prioritized skin cancer control issues in five key areas. Discussion centered around strategies for reducing UV exposure and increasing public and professional awareness of skin cancer. Panelists in five sessions developed consensus on several public and professional recommendations and a series of research strategies.

Evaluation of the American Academy of Dermatology's National Skin Cancer Early Detection and Screening Program

BACKGROUND

Increasing incidence and mortality rates from cutaneous melanoma are a major public health concern. As part of a national effort to enhance early detection of melanoma/skin cancer, the American Academy of Dermatology (AAD) has sponsored an annual education and early detection program that couples provision of skin cancer information to the general public with almost 750,000 free skin cancer examinations (1985-1994).

OBJECTIVE

To begin to evaluate the impact of this effort, we determined the final pathology diagnosis of persons attending the 1992-1994 programs who had a suspected melanoma at the time of examination.

METHODS

We directly contacted all such persons by telephone or mail and received pathology reports from those who had a subsequent biopsy.

RESULTS

We contacted 96% of the 4458 persons with such lesions among the 282,555 screenings in the 1992-1994 programs. We obtained a final diagnosis for 72%, and the positive predictive value for melanoma was 17%. Three hundred seventy-one melanomas were found in 364 persons. More than 98% had localized disease. More than 90% of the confirmed melanomas with known histology were in situ or "thin" lesions (< or = 1.50 mm thick). The median thickness of all melanomas was 0.30 mm. The 8.3% of AAD cases with advanced melanoma (metastatic disease, regional disease, or lesions > or = 1.51 mm) is a lower proportion than that reported by the 1990 Surveillance, Epidemiology and End Result Registry. The rate of thickest lesions (> or = 4 mm) and late-stage melanomas among all participants was 2.83 per 100,000 population. Of persons with a confirmed melanoma, 39% indicated (before their examination) that without the free program, they would not have considered having a physician examine their skin.

CONCLUSION

The 1992-1994 free AAD programs disseminated broad skin cancer educational messages, enabled thousands to obtain a free expert skin cancer examination, and found mostly thin, localized stage 1 melanomas (usually associated with a high projected 5-year survival rate). Because biases impose possible limitations, future studies with long-term follow-up and formal control groups should determine the impact of early detection programs on melanoma mortality.

The incidence of malignant melanoma in the United States: issues as we approach the 21st century

The risk of malignant melanoma developing in an American in the United States has now reached 1 in 87 (up more than 1800% since the 1930s). This rising incidence of malignant melanoma is, in fact, real because (1) it is not due to increased surveillance; (2) it is not due to better cancer-counting methods in general; (3) it is not due to changes in histologic diagnostic criteria; (4) it is being noted worldwide; and (5) most importantly, despite rising survival percentages, the mortality rate from malignant melanoma also continues to rise. On the basis of these trends, incidence rates for malignant melanoma will continue to rise for at least the next 10 to 20 years, although the demographics of those affected may change. Effective programs to improve public and professional education must be developed to enhance early clinical detection and behavioral changes. An establishment of a National Melanoma Registry is needed to more effectively assess the magnitude and impact of future incidence and the success of prevention program efforts into the next century.

Screening for melanoma by primary health care physicians: a cost-effectiveness analysis

BACKGROUND AND DESIGN - Australia has the highest rates of skin cancer in the world, and the incidence is estimated to be doubling every 10 years. Despite advances in the early detection and treatment of melanoma about 800 people still die nationally of the disease each year. A possible strategy for further reducing the mortality from melanoma is an organised programme of population screening for unsuspected lesions in asymptomatic people. Arguments against introducing melanoma screening have been based on cost and the lack of reliable data on the efficacy of any screening tests. To date, however, there has been no systematic economic assessment of the cost effectiveness of melanoma screening. The purpose of this research was to determine whether screening may be potentially cost effective and, therefore, warrants further investigation. A computer was used to simulate the effects of a hypothetical melanoma screening programme that was in operation for 20 years, using cohorts of Australians aged 50 at the start of the programme. Based on this simulation, cost-effectiveness estimates of melanoma screening were calculated. RESULTS - Under the standard assumptions used in the model, and setting the sensitivity of the screening test (visual inspection of the skin) at 60%, cost effectiveness ranged from Aust$6853 per life year saved for men if screening was undertaken five yearly to $12 137 if screening was two yearly. For women, it ranged from $11 102 for five yearly screening to $20 877 for two yearly screening. CONCLUSION - The analysis suggests that a melanoma screening programme could be cost effective, particularly if five yearly screening is implemented by family practitioners for men over the age of 50.

Skin cancer screening

The key to reducing morbidity and mortality from skin cancer is early detection. Prevention through education is, of course, ultimately the best outcome of collective human efforts. Many questions are still unresolved, but skin cancer screening by general internists, family practitioners, dermatologists, and other health professionals could prove cost-effective and reliable. As more knowledge becomes available, clinicians will have the resources to improve skills in prevention and early detection.

Screening for cutaneous malignant melanoma: a feasibility study

OBJECTIVES

To study two different procedures for inviting subjects to screening for malignant melanoma, and to study the role of specially trained nurses as primary examiners.

METHODS

The study was carried out in Stockholm county and comprised 5000 randomly selected men and women aged 40 to 60, of whom 2,500 were invited with a personal invitation letter and with a set time of appointment and 2,500 were sent a questionnaire and instructions for self examination of the skin. To test the nurses' competence, a survey was made, comparing the judgments of skin examination between nurses and doctors.

RESULTS

1654 (66%) of those invited for a skin examination took part. 1507 (60%) replied in the questionnaire group, of whom 1024 (41%) requested and attended a subsequent skin examination. Of the subjects attending from both groups, the nurses judged 9% v 11% as having skin lesions suggesting malignant melanoma, whereas the two doctors referred 4% v 3% of the subjects for surgical biopsy. Two malignant melanomas were detected, both of them in the questionnaire group. In the study of nurses' screening competence, both nurses and doctors equally judged one case out of 19 as having a lesion suspicious for malignant melanoma.

CONCLUSIONS

When the questionnaire method was used the results were achieved with one third fewer skin examinations, but it is difficult to conclude whether one of the procedures is better than the other. The results also indicate that nurses can be trained to be the primary examiners in a melanoma screening programme.

Trends of skin cancer in the Canton of Vaud, 1976-92

Trends in incidence and mortality for basal cell carcinomas (BCC), squamous cell carcinomas (SCC) and cutaneous malignant melanoma (CMM) for the period 1976-92 were analysed using data from the Cancer Registry of the Swiss Canton of Vaud. Among the 12,473 cases registered, 63% were basal cell carcinomas, 25% squamous cell cancers, 9% cutaneous malignant melanomas and 3% other miscellaneous histological types. Age-standardised incidence rates increased substantially for all histological types considered, from 44% increase for BCC in males to a more than 3-fold increase for SCC in females, with only signs of a levelling off in 1991-92, following a peak of incidence rates in 1986-90. From 1976-80 to 1991-92 CMM incidence increased by approximately 80% in both sexes. In 1991-92, age-standardised (world) incidence rates per 100,000 were 69.3 for basal cell, 29.1 for squamous cell cancers and 11.5 for melanomas in males, and, respectively, 62.2, 18.0 and 12.3 in females. With respect to mortality, in males rates increased for both non-melanocytic cancer (> 40%) and CMM (> 53%) whereas in females CMM, BCC and SCC rates remained approximately stable over the calendar periods examined. In 1991-92, age-standardised mortality rates per 100,000 were 2.6 for melanoma and 0.7 for other skin cancers in males, and, respectively, 1.6 and 0.2 in females. Upward trends in incidence were also present, and relatively homogeneous across, various age groups examined. However, SCC and CMM levelled off over the last period, and some decline was apparent in males below age 45. Separate analysis by anatomical site showed substantial increases in the head and neck for SCC and BCC, and in the trunk for CMM. In 1991-92, middle-aged women had almost equalled male incidence rates of BCC and SCC. A female excess of CMM incidence seemed to have disappeared since 1981-86. The increase in skin cancer incidence thus continued in this population up to the late 1980s, with a plateau only after 1990.

A pigmented lesion clinic: analysis of the first year's 1055 patients

The Pigmented Lesion Clinic (PLC) at Frenchay Hospital was started in January 1993 to deal rapidly and effectively with a growing number of referrals of suspicious pigmented lesions. Its objectives were to offer expert assessment and either reassurance or excision at the first PLC following referral by the patient's general practitioner. During 1993, 1055 patients were seen in 37 PLCs and 357 excision biopsies were performed. We have compared the Breslow thicknesses of malignant melanomas diagnosed from the PLCs with those of all other (non-PLC) malignant melanomas referred to our unit over the same period. We discuss the development and results of the first year of our PLC and the advantages it confers.

The early detection of and screening for melanoma. International status

Cutaneous melanoma is an increasingly common public health problem in industrialized nations. In theory, education and early detection should decrease melanoma morbidity and mortality, because the lesion is external and visible, risk factors are known, and thin tumors are associated with a high 5-year survival rate. The efficacy of early detection and screening programs, however, remains untested by randomized trials. Some early data are available from melanoma prevention and control efforts that have begun in many nations over the past decade. Textbook definitions of cancer screening are particularly challenging when applied to skin cancer control. Although some propose that there are rigid differences among screening, education, and early detection of skin cancer, the visible nature of this cancer makes early detection through education and screening inextricably intertwined. This article discusses the intermediate measures of education and screening within the context of the American Academy of Dermatology programs. Over the next few decades, screening and education for melanoma and other skin cancer will receive increasing worldwide attention. Multiple strategies will be required to combat increasing incidence and mortality rates. Screening and educational programs have been implemented worldwide, but data on their effects are only just being collected. Future collaborative work in melanoma/skin cancer prevention, education, and early detection holds promise as a way to decrease mortality and save lives.

Basal cell and squamous cell carcinomas are important risk factors for cutaneous malignant melanoma. Screening implications

BACKGROUND

This study was designed to determine the risk of developing malignant melanoma (MM) in patients with a history of basal cell and/or squamous cell skin cancer (BCC/SCC) and to determine whether surveillance efforts can be directed toward these patients for the detection of early MMs.

METHODS

The study cohort was followed by annual total cutaneous examination (TCE). Controls consisted of individuals from the United States population matched for age, sex, and length of follow-up. The anatomic locations of the study cohorts' newly diagnosed MMs were plotted on an anatomic chart. The setting was a private dermatology practice. Two hundred, ninety consecutive white patients with a history of BCC/SCC but with no personal or family history of MM were followed by annual TCEs. The main outcome measures were the relative risk of developing MM and their prognosis.

RESULTS

Ten of the 290 patients developed an MM within an average of 109 months of follow-up (range, 3-17 years). All MMs were less than 0.70 mm in Breslow thickness and 80% occurred on usually clothed cutaneous sites. The expected number of MMs in the control population was 0.59 (P = 0.006), resulting in a relative risk of 17.

CONCLUSION

Patients with BCC/SCC skin cancer are at substantial increased risk for developing MM. Regular and life-long surveillance TCE is an inexpensive and effective method for detecting curable MMs in such patients.

Melanoma control in the United States: current status

Screening and education for melanoma/skin cancer are receiving increasing worldwide attention. In an attempt to improve melanoma control in the United States, where incidence and mortality rates continue to increase, the American Academy of Dermatology (AAD) has provided free skin cancer screenings to more than 600,000 Americans and prevention messages to millions more. This chapter reviews and summarizes ongoing melanoma control activities in the United States, including prevention, education, and early detection/screening. We also address outcome measures in melanoma control and suggest that the unique external and visible nature of skin cancer blurs the distinction between screening and education. Measuring and evaluating the benefits of international melanoma control activities presents many challenging tasks. Future collaborative work must include complementary strategies in melanoma/skin cancer prevention, education, and screening to combat the worldwide increase in melanoma incidence and mortality.

Behavioral science in translational research and cancer control

Behavioral science plays an integral, essential role in cancer control. This review examines that role in two broad areas: (1) in the development and implementation of cancer control technologies, and (2) in the emerging focus on the process of translating cancer research into clinical practice. In developing and implementing cancer control technologies, what we know of human processes as a result of behavioral science is identified as the common link in all stages of the cancer process (prevention, detection, treatment, rehabilitation, terminal care). The value of behavioral science is shown in conservative projections (incorporating likely rates of patient noncooperation) of the health benefits obtainable by the year 2000 if effective behavioral strategies in three key areas were broadly implemented. Smoking prevention and cessation programs could lead to 7.2 million fewer smokers; application of cancer screening programs might result in 45,000 additional 5-year cancer survivors each year; and behavioral programs could lead to improved quality of life in 150,000 patients per year who experience treatment side effects despite the use of antiemetics. Behavioral science also is shown to contribute to translational science by providing a pragmatic conceptual model for clinical practice and facilitating collaborative research between basic scientists and clinical researchers. As an example of behavioral science's contribution to translational research, behavioral issues in genetic screening for cancer risk are outlined based on the experience of several families with the BRCA1 genetic marker and patients with Huntington's disease.

Behavioral and psychosocial cancer research in the underserved. An agenda for the future

BACKGROUND

This research was undertaken to review the status of behavioral and psychosocial cancer research in the underserved.

METHODS

The author extensively reviewed the literature related to behavioral and psychosocial cancer research in the underserved for this article. The author's experience as the Principal Investigator for the only National Institutes of Health smoking cessation study targeting Asian-Americans also was used in analyzing the literature and developing an agenda for the future.

RESULTS

Two questions and accompanying agenda items were addressed. The underserved were defined, and the effect of being underserved was illustrated through their disproportionate and earlier (younger) presence in morbidity and mortality statistics. Initiating behavioral and psychosocial cancer research in the underserved begins with a proper attitude and making accommodations to culture. The agenda for research should be practical and assessed on the basis of prevalence and preventability of cancer related behaviors. Examples from underserved Asian-American populations were cited.

CONCLUSIONS

Behavioral and psychosocial cancer research needs to be done.

Screening for melanoma and options for its evaluation [see comment]

A review of the published evidence presented here argues that screening for melanoma is recommended and practised at present, but with wide diversity of opinions about its value; there is evidence that screening has considerable potential for benefit, but the evidence of actual benefit is limited; and there are substantial costs and potential hazards from screening. On this basis the evaluation of screening procedures for melanoma is important, and options for this are discussed. The ideal study design to assess the efficacy of melanoma screening in reducing mortality is a large scale randomised trial. This may need a well coordinated proposal involving several centres in one or more countries, and the cost would be substantial. Without such a trial, however, it is most likely that increasing resources will be put into poorly designed screening programmes of unknown value. The simplest and strongest designs use individual randomisation, but group randomisation designs may have practical advantages, though they require a greater sample size. Designs based on general population screening, and on screening only high risk groups, are both considered. They answer different questions. In countries with high incidence the value of general population screening is probably the more critical. Not enough is known to specify the type and frequency of screening precisely; both screening by doctors and self screening require evaluation, and annual screening should probably be tested. The age range at risk will depend on the local incidence, but is likely to be quite wide-for example, 45-69, and both sexes need inclusion. Thus a suggested design for a moderate to high incidence area would be a trial, randomised by individual or group, assessing at least two annual rounds of both screening by doctor and self screening (ideally by a factorial design), for adults aged 45-69, with mortality over several years' follow up as the critical outcome. In an area with good data systems such a study could compare screening offered to some 260,000 subjects with 10 times that number of controls passively followed up, with 90% power to detect a one third reduction in mortality. A general assessment of costs over five years gave estimates of $8.3 million for the screening programme and $2.4 million for the evaluation. The much weaker designs, area based cohort studies using individual data or a simpler ecological comparison, and case-control studies, are also considered. If well designed with attention to their methodological limitations they may be valuable but are unlikely to be as definitive as a randomised trial.