NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Lin JS, Eder M, Weinmann S, et al. Behavioral Counseling to Prevent Skin Cancer: Systematic Evidence Review to Update the 2003 U.S. Preventive Services Task Force Recommendation [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2011 Feb. (Evidence Syntheses, No. 82.)

Cover of Behavioral Counseling to Prevent Skin Cancer

Behavioral Counseling to Prevent Skin Cancer: Systematic Evidence Review to Update the 2003 U.S. Preventive Services Task Force Recommendation [Internet].

Show details



See Appendix E for definitions of terms and abbreviations.

Key Questions and Analytic Framework

We developed an analytic framework with five primary key questions (KQs) based on the previous review and a scan of new primary and secondary research conducted since the previous review (Figure 1).

Figure 1. Analytic Framework and Key Questions.

Figure 1

Analytic Framework and Key Questions.

KQ 1 .

Is there direct evidence that counseling patients in sun-protective behaviors (decreasing sun exposure, avoidance of indoor tanning, and using sunscreen) reduces intermediate outcomes (sunburns, nevi, or actinic keratoses) or skin cancer (melanoma, squamous cell carcinoma, or basal cell carcinoma)?

KQ 2.

Do primary care relevant counseling interventions change sun-protective behaviors (decreasing sun exposure, avoidance of indoor tanning, and using sunscreen)?

KQ 3.

Do primary care relevant counseling interventions have adverse effects?

KQ 4.

Is sun exposure (intentional or unintentional), indoor tanning, or sunscreen use associated with skin cancer outcomes ?

KQ 5.

Are sun-protective behaviors associated with adverse effects (e.g., increased time spent in the sun, reduced physical activity, dysphoric mood, vitamin D deficiency)?

For KQ 4, we did not find studies meeting our inclusion criteria that examined a decrease in sun exposure(e.g., with use of protective clothing, avoidance of midday sun exposure) and skin cancer outcomes. Therefore, we included studies examining the relationship between sun exposure (intentional and unintentional) and skin cancer. We did not examine the association between UV exposure or sun-protective behaviors and intermediate outcomes (e.g., sunburns, nevi, or actinic keratoses). Epidemiologic links between intermediate health outcomes and skin cancer are also not reviewed in this report.

Literature Search Strategy

We searched for relevant systematic reviews published from 2001 to March 2008 in MEDLINE, Cochrane Database of Systematic Reviews, Database of Abstracts of Reviews of Effects, and Clinical Evidence or by the National Institute for Health and Clinical Excellence, Institutes of Medicine, and Agency for Healthcare Research and Quality (AHRQ). Fifteen relevant systematic reviews, in addition to the previous evidence report, were assessed for quality and their potential for answering KQs or identifying primary research for answering KQs. We developed separate literature searches and terms for each KQ (Appendix A Table 1) based on our assessment of the prior evidence report and the subsequent systematic review literature (Table 1).17,22,30–42 We identified 5,387 abstracts through MEDLINE and Cochrane Central Register of Controlled Trials (search dates are provided in Table 1).

Table 1. Search Strategies for Key Questions Based on Existing Systematic Reviews Identified.

Table 1

Search Strategies for Key Questions Based on Existing Systematic Reviews Identified.

For KQs 1 to 3, we searched only for randomized and nonrandomized controlled trials. For KQs 4 and 5, we searched for observational studies and trials. The search for KQs 1 to 3 updated the prior 2002 report. The KQ 4 search began with two more recent, fair-quality systematic reviews with comprehensive literature searches examining the association between indoor tanning devices and skin cancer30 and sunscreen use and melanoma.31 When we did not have existing systematic reviews and the question was not systematically reviewed in the prior report, we used 1966 as our search start date. We evaluated all studies included in the previous report1 against the inclusion and exclusion criteria for the current review. We also obtained articles from outside experts and by reviewing bibliographies of other relevant articles and existing systematic reviews. All searches were limited to articles in the English language.

Article Review and Data Abstraction

We reviewed all abstracts for potential inclusion for any of the KQs using the inclusion/exclusion criteria described in Appendix A Table 2. For KQs 1 to 3, examining the trial literature for the effectiveness and harms of behavioral counseling interventions to prevent skin cancer, we included only those counseling interventions that were conducted in primary care settings, judged to be feasible for delivery in primary care, or widely available for referral from primary care. In general, primary care relevant counseling interventions involved individual-level participant identification, a primary care practitioner or related clinical staff, and individual or small-group format with a limited number of sessions or was viewed as connected to the health care system. Behavioral counseling interventions that included an active component of community outreach, use of community members (e.g., opinion leaders, peer facilitators), use of community programs (e.g., worksite programs, school programs), use of social marketing, or use of public policy changes were not considered primary care relevant. Given the paucity of trial literature in this field, however, we also considered multimodal interventions if the intervention clearly involved primary care. We also required that trials evaluating counseling interventions be conducted in populations representative of primary care patients. Therefore, we excluded studies that exclusively enrolled participants with current or past history of malignant or premalignant skin lesions, or persons with syndromes at risk for skin cancer (e.g., persons with inherited or acquired immunodeficiency, xeroderma pigmentosum, albinism, basal cell nevus syndrome, exposure to arsenic, recessive dystrophic epidermolysis bullosa, medical exposure to psoralen or UVA treatment, familial atypical mole and melanoma syndrome, or more than 100 melanocytic nevi). We did, however, include persons at increased risk based on skin phenotype or family history of skin cancer. For KQs 1 to 3, included trials were conducted in English-speaking countries that are culturally similar to the United States.

For KQ 2, trials had to report behavioral outcomes 3 months or later after the counseling intervention. Behavioral outcomes included self-reported or directly observed measures of sun protection (e.g., limitation/avoidance of midday sun, use of sun-protective clothing, use of sunscreen, or limitation/avoidance of indoor tanning). We did not include behavioral outcomes, including skin self-examinations, because secondary prevention was not in this review’s scope and is addressed in a separate report.3 We also excluded trials that only reported outcomes related to knowledge, attitudes, self-esteem, or ability changes (skills).

For KQs 4 and 5, we had limited a priori exclusion criteria. However, we did exclude studies that focused on populations with syndromes at risk for skin cancer, as described above. We included trials and cohort studies when available. Due to the paucity of trial and cohort studies, we also considered nested case-control studies and population-based case-control studies. We excluded cross-sectional studies that were ecologic analyses and hospital-based case-control studies, as hospital-based controls are not generally representative of the community and hospital-based cases can introduce considerable selection bias.43,44 Outcomes for KQ 5 included potentially significant clinical harms (e.g., paradoxical increase in sun exposure, reduced physical activity, dysphoric mood, vitamin D deficiency, increased incidence of other types of cancer).

Two investigators independently screened 5,387 abstracts for potential inclusion, including all abstracts for KQs 1 to 3, and every fifth abstract (20 percent sample) for KQs 4 and 5. There were a total of four discrepancies between the two reviewers for the 480 dual-reviewed abstracts for KQs 4 and 5 (agreement of 99.2 percent). None of these four abstracts was included in the final review. Therefore, we feel confident that no relevant articles were missed by having a second investigator dual review only a subset of the abstracts.

We reviewed a total of 60 articles for KQs 1 to 3 and 264 articles for KQs 4 and 5. Two investigators independently rated all articles meeting inclusion criteria for quality assessment using the USPSTF’s study-design specific quality criteria, which was supplemented by the Newcastle-Ottawa Scale for assessing cohort and case-control studies (Appendix A Table 3). The USPSTF has defined a three-category quality rating of “good,” “fair,” and “poor” based on these criteria. In general, a good-quality study meet s all criteria well. A fair-quality study does not meet, or it is not clear that it meets, at least one criterion, but it has no known important limitation that could invalidate its results. A poor-quality study has important limitations.2 All poor-quality studies were excluded. Case series and case reports were not included unless they addressed fatal harms. Listings of all excluded articles are included in Appendix B Tables 24 and Appendix C Tables 2 and 4. A flow chart of reviewed abstracts and articles is included in Appendix A Figure 1.

We found no trials for KQ1. This review included 14 articles representing 10 unique trials for KQs 2 and 3; 56 articles representing 32 unique studies for KQ 4; and 18 articles representing 16 unique trials for KQ 5. One primary reviewer abstracted relevant information into standardized evidence tables for each included article (Appendix B Table 1 and Appendix C Tables 1 and 3). A second reviewer checked the abstracted data for accuracy and completeness.

Literature Synthesis

We found no data for KQ 1. For KQs 2 and 3, we were unable to conduct quantitative synthesis, primarily due to the heterogeneity of populations addressed and counseling intervention methods. Instead, we qualitatively synthesized our results stratified by the populations addressed: adults, young adults, adolescents, and parents and children. The Results section and corresponding summary tables reflect these qualitative summaries.

Similarly, for KQs 4 and 5, we were unable to pool estimates of associations due to the heterogeneity in measurement of exposures and outcomes. Instead, we qualitatively synthesized our results stratified by type of exposure addressed (sun exposure, indoor tanning, and sunscreen use) or type of adverse effect.

USPSTF Involvement

The authors worked with three USPSTF liaisons at key points throughout the review process to develop and refine the scope, analytic framework, and KQs; to resolve issues around the review process; and to finalize the evidence synthesis. AHRQ funded this research under a contract to support the work of the USPSTF. AHRQ had no role in study selection, quality assessment, or synthesis, although AHRQ staff provided project oversight, reviewed the draft evidence synthesis, and assisted in external review of the draft evidence synthesis. This systematic evidence review was revised based on comments from five expert reviewers.


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...