Journal Watch Dermatology<br>February 3, 2009

Screening for Skin Cancer: The Evidence Is Still Insufficient

June K. Robinson, MD


Journal Watch. 2009;8(2) 

Randomized, controlled trials of sufficient size and scope are still needed

The United States Preventive Services Task Force (USPSTF) has updated their recommendations on skin cancer screening by primary care physicians and patient self-examination, reviewing any new evidence published since their last review in 2001. Little has changed over the years in their recommendations: Task force members again found insufficient evidence to make a recommendation on screening in the general adult population without history of premalignant or malignant lesions. The task force did not examine outcomes related to surveillance of patients at extremely high risk (e.g., those with familial syndromes). They noted the possibility of different risk-benefit ratios in those with history of skin cancer or other high-risk factors.

To determine the accuracy of whole-body skin screening by primary care physicians, the task force analyzed 11 studies, finding sensitivity ranges from 42% to 100% and specificity of 98% in the diagnosis of melanoma. In an Australian community-based RCT of screening by primary care physicians, the overall positive predictive value for all types of skin cancer was 29% (J Am Acad Dermatol 2006; 54:105). The USPSTF found insufficient evidence to determine whether dermatologists and primary care physicians differ in accuracy, citing one analysis of trials published from 1996 to 1999 (Arch Dermatol 2001; 137:1627). That the USPSTF cited this study, regardless of the age of the primary sources, demonstrates the task force’s focus on systematic reviews. Dermatologists’ use of dermoscopy to enhance diagnostic accuracy was not considered.

The task force identified critical gaps in the evidence, including a lack of studies on the benefits of early skin cancer detection for reducing mortality and morbidity and of studies on potential risks of screening (e.g., misdiagnosis, overdiagnosis, and harm from biopsies and overtreatment).


Consensus conferences and voluntary skin cancer screenings by dermatologists cannot produce the level of evidence needed for USPSTF recommendations. Needed are well-designed, well-conducted prospective, controlled studies, of sufficient duration, in representative primary care populations of sufficient size, to determine the efficacy of screening for reducing mortality and morbidity. In addition to efficacy, researchers need to consider potential harms of self-examination (e.g., unnecessary physician visits and needless skin biopsies) and the cost-effectiveness of physician surveillance.

Dermatologists believe that our accuracy exceeds primary care physicians’. Limited studies support this contention, although they lack a sufficient level of evidence to be considered by the USPSTF (Arch Dermatol 2008; 144:97). Dermatologists usually screen higher-risk patients than primary care physicians do. It is important to compare accuracy in the same populations (i.e., those at average or high risk). Dermatologists’ accuracy in detecting melanomas in patients with dysplastic nevus syndrome cannot be compared with the accuracy of primary care physicians in average-risk populations. Without prospective RCTs to evaluate accuracy in real-world clinical situations, future USPSTF findings will resemble the findings of 1996 and 2001. The research community has not yet had the resources to perform the necessary population-based studies.

If you would like to weigh in on this issue, please e-mail us at

CLICK HERE for Journal Watch subscription information.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.