The Relation between Patterns of Vacation Sun Exposure and the Development of Acquired Melanocytic Nevi in German Children 6 7 Years of Age

Transcription

1 American Journal of Epidemiology Copyright ª 2007 by the Johns Hopkins Bloomberg School of Public Health All rights reserved; printed in U.S.A. Vol. 165, No. 10 DOI: /aje/kwm007 Advance Access publication March 3, 2007 Original Contribution The Relation between Patterns of Vacation Sun Exposure and the Development of Acquired Melanocytic Nevi in German Children 6 7 Years of Age Olaf Gefeller 1, Jasmin Tarantino 1, Peter Lederer 2, Wolfgang Uter 1, and Annette B. Pfahlberg 1 1 Department of Medical Informatics, Biometry, and Epidemiology, University of Erlangen-Nuremberg, Erlangen, Germany. 2 Regional Health Authority, Erlangen, Germany. Received for publication April 12, 2006; accepted for publication November 2, Sun exposure is the main environmental risk factor for the development of melanocytic nevi. Although the general association is not disputed, the interplay between intense intermittent and the cumulative amount of sun exposure in defining the promoting effect on melanocytic nevus development is an area of debate. Trained staff members ascertained total body counts of melanocytic nevi in a cross-sectional study of 2,189 children 6 7 years of age who were recruited in two German centers in Their parents provided information about a variety of exposure factors. The distribution of melanocytic nevi was skewed markedly to the right; therefore, a negative binomial regression model provided the appropriate framework for a multivariable analysis. A steep gradient with respect to the (adjusted) number of melanocytic nevi was apparent only for the frequency of vacation episodes associated with sun exposure in areas with an intense ultraviolet radiation. In contrast, no such gradients were found for the cumulative duration of vacation sun exposure in such areas or for any variable related to vacation sun exposure in areas with a low ultraviolet radiation. This observation supports the hypothesis that intermittent exposure to high doses of ultraviolet radiation plays an especially important role in nevus development. child; dermatology; environmental exposure; holidays; melanoma; nevus; sunlight; ultraviolet rays Numerous epidemiologic investigations during recent years have focused on the determinants of acquired melanocytic nevi in children and adolescents (1 24). This interest stems from the well-established link between melanocytic nevi and the occurrence of cutaneous melanoma in adults (25). The total number of melanocytic nevi has consistently been identified as the strongest risk factor for melanoma (25). Additionally, some melanocytic nevi are considered to be precursor lesions of melanoma and are hypothesized to constitute one step in the causal pathway leading to the disease (26 28). The main environmental risk factor for the development of both cutaneous melanoma and melanocytic nevi is sun exposure (2 4, 6, 9 11, 13, 15, 19 23, 29). However, studying the effects of sun exposure is a complex issue. It is hampered by problems of ascertaining frequency, duration, latitude, and altitude of the exposure, as well as by behavioral aspects regarding the application of sun-protective measures such as clothing and sunscreen use based on the study subjects recall. Although the general picture of the association between high levels of sun exposure and an elevated number of melanocytic nevi has been pinpointed by all studies on this topic, some details of the relation have remained an area of controversial discussion. In particular, the interplay between intense intermittent and the cumulative amount of sun exposure in defining the promoting effect on melanocytic nevus development has been debated for some time with no consensus yet reached (2, 3, 6, 11, 13, 15, 19, 21, 23). Our study on German children aged 6 7 years attempted to disentangle the effects of intermittent and cumulative sun exposure on melanocytic nevus development. The key idea of the approach was to focus on the children s history of vacation sun exposure and to define different patterns of vacation sun exposure by using detailed information on the Correspondence to Dr. Olaf Gefeller, Department of Medical Informatics, Biometry, and Epidemiology, University of Erlangen-Nuremberg, Waldstr. 6, D Erlangen, Germany ( gefeller@rzmail.uni-erlangen.de). 1162

2 Vacation Sun Exposure and Nevus Development 1163 frequency, duration, latitude, and altitude of that exposure. The comparison of results obtained in the categories of these exposure patterns would then allow us to distinguish the effects of intermittent and cumulative sun exposure. MATERIALS AND METHODS Study design and participants The NAEVAC Study investigated the determinants of the number of acquired melanocytic nevi since birth among children 6 7 years of age in two German centers. The crosssectional study was carried out in Salzgitter, a city in Lower Saxony situated at latitude 52 N, and in Erlangen-Hoechstadt, a county in Bavaria located at latitude 49 N, from November 2001 to April Ethical approval of the NAEVAC Study procedures was obtained by the ethics committee at the Medical Faculty of the University of Goettingen, Lower Saxony. In the two centers, the study was integrated as an add-on into the mandatory medical examination prior to school enrollment conducted by the staff from the respective local health authorities of Salzgitter and Erlangen-Hoechstadt. After the child s standard medical examination, parents received an information sheet and were then asked to consent to their child s participation in the additional NAEVAC Study part. Parents were not informed in detail about the actual objectives of the study, because this information was expected to influence their behavior in filling out the questionnaire. Altogether, 1,162 children from Salzgitter and 1,145 children from Erlangen-Hoechstadt were registered by their parents at the local health authorities for school enrollment in Of these presumably eligible children, the parents of 82 children refused to participate, the parents in 28 cases could not be reached for the interview, and eight children were ill on the day of examination. Parents who agreed to participate represented 2,189 (94.9 percent) children. However, 87 of these 2,189 children were excluded from the study, because the child was not of Caucasian origin (n ¼ 20) or the skin examination could not be completed (e.g., the child was not willing to be examined) (n ¼ 67). Because of missing values in one or more variables, data from 204 children could not be considered, leaving 1,898 children for the final analysis. Nevus counts and assessment of pigmentary traits The main outcome was the total number of acquired melanocytic nevi 2 mm or larger in diameter. In both centers, physicians and other staff members specifically trained by two dermatologists for the recognition of skin pigmented lesions performed full-body examinations on all children (excluding the scalp, genitals, buttocks, and soles of feet). Counting of nevi was standardized by use of transparent plastic stencils pierced with a 2-mm hole to distinguish the target lesions from smaller ones. According to a standard protocol, nevi were defined as brown to black pigmented macules or papules, darker in color than the surrounding skin, excluding lesions with the clinical characteristics of freckles, solar lentigines, or café-au-lait spots (30). Congenital nevi, nevi spili, and blue nevi were not counted, while halo nevi were included. Further physical examinations of the children included 1) measurement of height and weight, 2) visual assessment of hair and iris color by use of high-quality color reference photographs of hair strands and irises of artificial eyes, and 3) a semiquantitative estimation of the degree of freckling, modified after the method of Gallagher et al. (31) supported by reference charts. Skin color of the inner upper arm was determined with reflectance photometry at 660 nm (Reflektometer RM 100; Courage and Khazaka, Cologne, Germany). The reaction of the unprotected skin to midday summer sun for the first time during the year was determined via personal interview according to the Fitzpatrick classification (32). Questionnaire Parents were asked to complete a questionnaire regarding the child s age, nationality, usual outdoor activities during the summer, detailed history of vacations, history and severity of sunburns, and sunscreen use, among other items. To assess habitual sun exposure during the summer months, parents were asked how much time their child usually spent outside in the sun between 10 a.m. and 5 p.m. and how often their child visited an open-air swimming pool or a lake for bathing or played in a paddling pool (excluding vacation periods) each year. Regarding sun exposure during vacations, information on the frequency and average duration of the child s vacations from birth up to the time of examination in different geographic regions was collected. The geographic areas were classified by latitude into two categories, one representing locations in sunny areas south of Salzgitter and Erlangen- Hoechstadt, such as Turkey, Italy, or the Canary Islands, and the second category comprising sunny beach vacations in areas situated north of the respective study centers, for example, at the North Sea or the Baltic Sea. Vacation sun exposure was described with seven different variables. The first three pertained to number of vacation periods, where the overall frequency of vacation periods was derived by summing up the number of vacations in the South and North. For the analysis, all three variables were classified into four categories (table 2). Another three variables summarized the cumulative duration of vacations in weeks for overall vacations, vacations in southern areas, and vacations in northern areas (table 2). Overall cumulative duration of vacations and cumulative duration of vacations in southern areas were classified into five categories, while cumulative duration of vacations in northern areas was divided into only four groups because of the small frequencies for long aggregate durations. Finally, a variable depicting the exposure pattern was constructed by combining the cumulative duration and the frequency of vacations. For determining the categories, these last two variables were cross-tabulated, and the cumulative duration of vacations was first categorized and then divided, where possible, into subgroups with different frequencies of vacations according to the data. Such a subdivision was performed for the last two categories

3 1164 Gefeller et al. TABLE 1. Univariate analyses of variables related to lifetime vacation sun exposure in 1,898 German children aged 6 7 years in 2002 (5 8 and >8 weeks, respectively), yielding a variable with eight categories (table 3). Sunburn history was derived from two questions addressing the frequency of sunburns as redness without peeling and redness for longer than a day and pain with either blistering or peeling, respectively. The frequencies of both types were combined in a severity score. A sun-protection score was obtained by merging the frequency of sunscreen application and the level of sun protection factor used. Statistical analysis The total body count of melanocytic nevi on the children was the primary outcome variable of the study. Its distribution in the sample was graphically inspected and compared with a fitted Poisson and a negative binomial distribution with parameters estimated from the data via maximum likelihood techniques. To take account of the distributional skewness in the descriptive part of the analysis, the distributions of melanocytic nevi in various subgroups were characterized with median values and interquartile s in addition to means and standard deviations. We assessed the homogeneity of distributions between groups using the nonparametric Wilcoxon test for two groups and the Kruskal-Wallis test for more than two groups. We constructed a negative binomial regression model to analyze the association between total body count of melanocytic nevi and different vacation sun exposure variables in a multivariable framework and to control for confounding. All models incorporated the following characteristics, which have been identified as confounders for the association under study: age (as a continuous variable), gender, center, body surface area (as a continuous variable), eye color, hair color, skin color, freckles, skin type according to Fitzpatrick (32), average number of hours spent outdoors between 10 a.m. and 5 p.m., average number of outdoor swimming pool visits during summer, number of sunburns since birth, and Total (n ¼ 1,898) Males (n ¼ 958) Females (n ¼ 940) sunscreen use. Results of the negative binomial regression analyses are presented in terms of estimated (adjusted) percentages of change in melanocytic nevus number relative to the corresponding reference category, accompanied by 95 percent confidence intervals computed by the profile likelihood method. We regarded p values lower than 0.05 as statistically significant. All statistical analyses were performed using SAS, version 9.1.3, software (SAS Institute, Inc., Cary, North Carolina), with the GENMOD procedure used for negative binomial regression modeling. RESULTS p value* Vacations Frequency of periods Cumulative duration (weeks) Vacations in southern areas Frequency of periods Cumulative duration (weeks) Vacations in northern areas Frequency of periods Cumulative duration (weeks) * The Wilcoxon test was used to test heterogeneity in the distribution of vacation variables by gender. The study group for the final analysis comprised 940 girls (49.5 percent) and 958 boys (50.5 percent), with 1,040 children from Erlangen-Hoechstadt (54.8 percent) and 858 children from Salzgitter (45.2 percent). The median age was 6.0 years (interquartile : years), and most of the children (88.8 percent) were German. The distribution of melanocytic nevi was skewed to the right, with a few children having very high counts. The median number of melanocytic nevi was three, with an interquartile of 1 5, and with counts ranging from 0 to 43. No nevus was identified on 334 children (17.6 percent), 15.1 percent of the children had one nevus, and 4.2 percent of the children had 10 or more melanocytic nevi 2 mm or more in diameter. The pattern was best approximated by a negative binomial distribution, while the Poisson distribution did not fit the observed data well because of overdispersion. There was a slight, nonsignificant sex difference regarding the number of melanocytic nevi, with medians of two melanocytic nevi in girls and three melanocytic nevi in boys. Altogether, 79.8 percent of the children had been on vacations associated with sun exposure at least once in their life. The average overall cumulative duration of vacations

4 Vacation Sun Exposure and Nevus Development 1165 TABLE 2. Bivariate associations of variables related to lifetime vacation sun exposure with number of melanocytic nevi at all body sites in 1,898 German children aged 6 7 years in 2002 No. of cases Mean Melanocytic nevi (no.) Standard deviation p value* Frequency of vacation periods > Frequency of vacation periods in southern areas > <0.001 Frequency of vacation periods in northern areas 0 1, > Cumulative duration of vacations (weeks) > Cumulative duration of vacations in southern areas (weeks) > Cumulative duration of vacations in northern areas (weeks) 0 1, > * The Kruskal-Wallis test was used to test heterogeneity in the distribution of the number of melanocytic nevi by all the variables shown in table. since birth up to the day of examination was 5 weeks per child, with a median number of overall vacation periods of three (table 1). No gender differences were found in any of the vacation variables. The association between the total number of melanocytic nevi and different vacation sun exposure variables is shown in table 2. Overall, with the rising number of total vacation periods, the mean number of melanocytic nevi increased

5 1166 Gefeller et al. TABLE 3. Bivariate associations of lifetime vacation sun exposure in southern areas (combination of cumulative duration and frequency of vacations) related to number of melanocytic nevi at all body sites in 1,898 German children aged 6 7 years in 2002 Lifetime vacation sun exposure in southern areas (combination of cumulative duration and frequency) No. of cases Mean significantly from 3.09 to 3.84 (p ¼ 0.007). An even stronger association was found between the number of melanocytic nevi and the frequency of vacation periods in southern areas, where the mean number of melanocytic nevi rose from 3.25 to 4.71 in the identical frequency categories (p < 0.001). In contrast, no such association could be seen for the frequency of vacation periods in northern areas (p ¼ 0.594). With respect to the cumulative duration of vacations in southern areas, the mean number of melanocytic nevi rose with the number of weeks, reaching a peak at 5 8 weeks, whereas values decreased subsequently. The association between the number of melanocytic nevi and the overall cumulative duration of vacations followed a similar but more fluctuating pattern, while no such association was apparent regarding vacations in northern areas. Thus, neither the frequency of vacation periods nor the cumulative duration of vacations in northern areas was significantly associated with the total number of melanocytic nevi. Therefore, further analyses were limited to variables concerning vacations in southern areas, because the effects of overall exposure are presumably driven by the vacations in the South and diluted by vacations in the North. When considering the association between the number of melanocytic nevi in children and the vacation sun exposure pattern in southern areas using the combined variable of duration and frequency of vacation periods, a complex picture could be seen (table 3). First, the number of melanocytic nevi increased with the cumulative duration from zero across 1 2 weeks and 3 4 weeks of vacations. Then, within the categories 5 8 weeks cumulative duration and >8 weeks cumulative duration, the number of melanocytic nevi increased steadily with an ascending number of vacation periods. Finally, three multivariable regression models were analyzed (table 4). Each of the three exposure variables of Melanocytic nevi (no.) Standard deviation interest frequency of vacation periods in southern areas (p ¼ 0.010), cumulative duration of vacations in southern areas (p ¼ 0.025), and the combination of cumulative duration of lifetime vacation sun exposure in southern areas with the frequency of vacation periods (p ¼ 0.002) showed a significant association with the number of melanocytic nevi, yet the gradients of the percentage change in melanocytic nevus number relative to the reference category were rather different for the three variables. In the first model, there was a clear gradient of increasing number of melanocytic nevi with ascending number of vacation periods. The second model showed a mixed picture concerning the relation between an increasing cumulative duration of vacation and number of melanocytic nevi. Only in two categories (3 4 and 5 8 weeks) did we observe a significantly higher number of melanocytic nevi compared with the reference category. Conclusions from the third model confirmed the crude association found between lifetime vacation sun exposure in southern areas (combination of cumulative duration and frequency of vacations) and number of melanocytic nevi in table 3. The same positive gradient of an increasing number of melanocytic nevi with an increasing number of vacation periods was identified in the subgroups of the categories 5 8 weeks cumulative duration and >8 weeks cumulative duration. DISCUSSION p value* No vacations weeks weeks weeks 1 3 times >3 timesy >8 weeks 1 3 times times >5 times <0.001 * The Kruskal-Wallis test was used to test heterogeneity in the distribution of the number of melanocytic nevi by all the variables shown in table. y Further stratification of this category identical to the >8 weeks category was impossible as only four children in this category had been on vacation more than five times. The major finding of our analysis was that a steep gradient with respect to the (adjusted) number of melanocytic nevi was apparent only for the frequency of vacation episodes in areas with relatively intense ultraviolet radiation. In contrast, the relation between the number of nevi and the cumulative duration of vacation sun exposure in such areas

6 Vacation Sun Exposure and Nevus Development 1167 TABLE 4. Estimates of associations between nevus count and variables related to lifetime vacation sun exposure in southern areas in different negative binomial regression models for 1,898 German children aged 6 7 years in 2002 Variables additionally included in model was less regular, and there was no gradient for any variable related to vacation sun exposure in areas with a low ultraviolet radiation. This observation gives credit to the hypothesis that it is intermittent exposure to a high dose of ultraviolet radiation in particular that plays an important role in the development of melanocytic nevi. Our analysis is based on a large study group of German children 6 7 years of age. More than 2,000 children were recruited for the investigation via a medical examination prior to school enrollment that is mandatory in the German system. Therefore, the typical problem of selection bias in a situation when study subjects are voluntarily sampled from some target population and only a limited response is achieved did not occur here. No. of cases Estimate* 95% confidence interval p value Frequency of vacation periods in southern areas y ÿ5, , 37 > , Cumulative duration of vacations in southern areas (weeks) y ÿ3 ÿ15, , , 33 > ÿ4, Lifetime vacation sun exposure in southern areas (combination of cumulative duration and frequency) No vacations 690 0y 1 2 weeks 295 ÿ2 ÿ14, weeks , weeks 1 3 times ÿ7, 25 >3 timesz , 63 >8 weeks 1 3 times 82 ÿ10 ÿ29, times ÿ11, 24 >5 times , * Estimates and 95% confidence intervals are presented in terms of the percentage change in melanocytic nevus numbers relative to the reference category. These were derived from the coefficients of separate negative binomial regression models including the variables age, gender, center, body surface area, eye color, hair color, skin color, freckles, Fitzpatrick skin type, average number of hours spent outdoors between 10 a.m. and 5 p.m., average number of outdoor swimming pool visits during summer, number of sunburns since birth, and sunscreen use. y Reference category. z Further stratification of this category identical to the >8 weeks category was impossible as only four children in this category had been on vacation more than five times. The size of the study group gave us the opportunity to disentangle the interrelated effects of frequency and duration of vacation sun exposure. We accomplished this by considering the joint distribution of the two variables and estimating the interaction structure in a multivariable negative binomial regression model. This type of regression approach provided the best fit to our nevus count data and allowed us to incorporate all relevant confounders simultaneously. When considering frequency and duration of vacation sun exposure separately, we found a statistically significant relation between each variable and the number of melanocytic nevi. However, in a joint model incorporating their interaction structure, we observed that, when holding the

7 1168 Gefeller et al. cumulative duration of sun exposure constant, a strong effect of a variation in the frequency of sun exposure could be identified. On the other hand, when the frequency was held constant, the effect of a variation in the cumulative duration was less pronounced and in the other direction; that is, a longer duration of vacation sun exposure was associated with fewer melanocytic nevi relative to a shorter duration in the same frequency category. This complex pattern has not been revealed in the previous epidemiologic studies on this topic because of limited sample sizes that precluded fine stratifications of the joint distribution of frequency and duration. Autier et al. (13) also analyzed the impact of duration and number of vacation periods on melanocytic nevus development in a study of 628 European children 6 7 years of age. They identified both sun exposure markers as significantly associated with melanocytic nevus development but could not address the joint effect directly. Their analysis also corroborated our finding that the latitude of vacation is an important characteristic modifying the association between sun exposure and melanocytic nevus development. This research issue can probably best be addressed by studies in regions such as Central Europe or North America, where the usual exposure of children to ultraviolet radiation outside vacation periods is not as high as, for example, in Australia, where children are continually exposed to high doses of ultraviolet radiation. Because of the geographic heterogeneity of the study group of Autier et al., they considered the difference between the latitude of vacation places and the latitude of habitual living places. Their principal finding is in line with our observation that vacations in southern areas act as a promoter of melanocytic nevus development, while this could not be verified for vacations in northern areas. Examining vacation sun exposure and not the occurrence of sunburns or sun exposure in general during the first years of life by approaching parents (mainly mothers) has been chosen to minimize potential recall bias and to enhance the precision of exposure ascertainment. Parents were only vaguely informed about our actual study objectives and had no knowledge of the particular aspect of the relation between sun exposure and melanocytic nevus development we were looking for. Experience from studies on risk factors for melanoma indicates that the assessment of sunburn history is prone to problems of recall bias (33, 34), which has influenced our decision to focus on the history of vacation sun exposure rather than on its consequences for the skin. We refrained also from ascertaining general sun exposure in detail, which is fraught with problems of misclassification (2). The frequency and duration of vacation periods over a period of 6 years can, however, be remembered well by the parents and yield exposure data with sufficient precision for meaningful analysis. Limitations of our study include the assessment of the number of melanocytic nevi by different observers. Nine observers were involved in the study for logistic reasons. However, all were specifically trained by two dermatologists using a standard protocol (30) and showed high interrater reliability in evaluations of their nevus counts during the training sessions. The inspection of the children s skin was performed outside the summertime when freckling and skin pigmentation are not yet increased because of sun exposure. Therefore, heavy freckling, which may complicate differential diagnosis between melanocytic nevi and freckles, constituted a problem only in very rare cases. To minimize the risk of misclassification even further, we counted only pigmented lesions 2 mm or larger in diameter on the whole body (excluding the scalp, genitals, buttocks, and soles of feet). Most recent studies on nevus development in children assessed all melanocytic nevi on the body (10, 11, 15, 19 23) and sometimes additionally distinguished between the overall number of melanocytic nevi and melanocytic nevi of different sizes (11, 21, 22). The restriction on larger melanocytic nevi leads to lower absolute melanocytic nevus counts in our study, which has to be kept in mind when comparing results across studies. In conclusion, the exposure to intense intermittent doses of ultraviolet radiation induced by frequent vacations early in life shows a strong association with nevus development. Public health education should focus on the message to minimize that type of exposure to the sun in young children. ACKNOWLEDGMENTS The NAEVAC Study was supported by grant from the Wilhem Sander Foundation, Munich, Germany. The first author was additionally supported by a grant from the German Research Foundation (Ge ). The authors thank Sabine Billmann, Udo Reulbach, Dr. Annette Wicovsky, Christine Wolf, and Dr. Yurdaguel Zopf, University of Erlangen-Nuremberg, and the staff members of the local health authorities in Erlangen and Salzgitter, who assisted with data collection. Members of the NAEVAC Study Group in addition to the authors are Dr. Klaus F. Koelmel, Department of Dermatology, University of Goettingen, and Dr. Stefan Mueller- Dechent, Regional Health Authority, Salzgitter. Conflict of interest: none declared. REFERENCES 1. Daures JP, Sancho-Garnier H, Pourrin-Bourdonneau C, et al. Study of demographic, environmental and congenital factors in the development of nevi in children between three and fifteen years old. Early results. Rev Epidemiol Sante Publique 1995; 43: Dwyer T, Blizzard L, Ashbolt R. Sunburn associated with increased number of nevi in darker as well as lighter skinned adolescents of northern European descent. Cancer Epidemiol Biomarkers Prev 1995;4: Luther H, Altmeyer P, Garbe C, et al. Increase of melanocytic nevus counts in children during 5 years of follow-up and analysis of associated factors. Arch Dermatol 1996;132: Enta T, Kwan TY. Melanocytic nevi in sun-protected Canadian Hutterite children. Arch Dermatol 1998;134: Graham A, Fuller A, Murphy M, et al. Maternal and child constitutional factors and the frequency of melanocytic naevi in children. Paediatr Perinat Epidemiol 1999;13:

8 Vacation Sun Exposure and Nevus Development Harrison SL, Buettner PG, MacLennan R. Body-site distribution of melanocytic nevi in young Australian children. Arch Dermatol 1999;135: Azizi E, Iscovich J, Pavlotsky F, et al. Use of sunscreen is linked with elevated naevi counts in Israeli school children and adolescents. Melanoma Res 2000;10: Gallagher RP, Rivers JK, Lee TK, et al. Broad-spectrum sunscreen use and the development of new nevi in white children: a randomized controlled trial. JAMA 2000;283: Carli P, Naldi L, Lovati S, et al. The density of melanocytic nevi correlates with constitutional variables and history of sunburns: a prevalence study among Italian schoolchildren. Int J Cancer 2002;101: Darlington S, Siskind V, Green L, et al. Longitudinal study of melanocytic nevi in adolescents. J Am Acad Dermatol 2002; 46: Dulon M, Weichenthal M, Blettner M, et al. Sun exposure and number of nevi in 5- to 6-year-old European children. J Clin Epidemiol 2002;55: Siskind V, Darlington S, Green L, et al. Evolution of melanocytic nevi on the faces and necks of adolescents: a 4 y longitudinal study. J Invest Dermatol 2002;118: Autier P, Severi G, Pedeux R, et al. Number and size of nevi are influenced by different sun exposure components: implications for the etiology of cutaneous melanoma (Belgium, Germany, France, Italy). Cancer Causes Control 2003;14: MacLennan R, Kelly JW, Rivers JK, et al. The Eastern Australian Childhood Nevus Study: site differences in density and size of melanocytic nevi in relation to latitude and phenotype. J Am Acad Dermatol 2003;48: Wiecker TS, Luther H, Buettner P, et al. Moderate sun exposure and nevus counts in parents are associated with development of melanocytic nevi in childhood: a risk factor study in 1,812 kindergarten children. Cancer 2003;97: Pfahlberg A, Uter W, Kraus C, et al. Monitoring of nevus density in children as a method to detect shifts in melanoma risk in the population. Prev Med 2004;38: Oliveria SA, Geller AC, Dusza SW, et al. The Framingham school nevus study: a pilot study. Arch Dermatol 2004;140: Synnerstad I, Nilsson L, Fredrikson M, et al. Frequency and distribution pattern of melanocytic naevi in Swedish 8 9- year-old children. Acta Derm Venereol 2004;84: Bauer J, Buttner P, Wiecker TS, et al. Risk factors of incident melanocytic nevi: a longitudinal study in a cohort of 1,232 young German children. Int J Cancer 2005;115: Harrison SL, Buettner PG, Maclennan R. The North Queensland Sun-Safe Clothing Study: design and baseline results of a randomized trial to determine the effectiveness of sunprotective clothing in preventing melanocytic nevi. Am J Epidemiol 2005;161: Valiukeviciene S, Miseviciene I, Gollnick H. The prevalence of common acquired melanocytic nevi and the relationship with skin type characteristics and sun exposure among children in Lithuania. Arch Dermatol 2005;141: Whiteman DC, Brown RM, Purdie DM, et al. Melanocytic nevi in very young children: the role of phenotype, sun exposure, and sun protection. J Am Acad Dermatol 2005;52: English DR, Milne E, Simpson JA. Ultraviolet radiation at places of residence and the development of melanocytic nevi in children (Australia). Cancer Causes Control 2006;17: Kallas M, Rosdahl I, Fredriksson M, et al. Frequency and distribution pattern of melanocytic naevi in Estonian children and the influence of atopic dermatitis. J Eur Acad Dermatol Venereol 2006;20: Gandini S, Sera F, Cattaruzza MS, et al. Meta-analysis of risk factors for cutaneous melanoma. I. Common and atypical naevi. Eur J Cancer 2005;41: Stolz W, Schmoeckel C, Landthaler M, et al. Association of early malignant melanoma with nevocytic nevi. Cancer 1989;63: Skender-Kalnenas TM, English DR, Heenan PJ. Benign melanocytic lesions: risk markers or precursors of cutaneous melanoma? J Am Acad Dermatol 1995;33: Carli P, Massi D, Santucci M, et al. Cutaneous melanoma histologically associated with a nevus and melanoma de novo have a different profile of risk: results from a case-control study. J Am Acad Dermatol 1999;40: Gandini S, Sera F, Cattaruzza MS, et al. Meta-analysis of risk factors for cutaneous melanoma. II. Sun exposure. Eur J Cancer 2005;41: English DR, McLennan R, Rivers JK, et al. Epidemiological studies of melanocytic naevi: protocol for identifying and recording naevi. Lyon, France: International Agency for Research on Cancer, (IARC internal report 90/002). 31. Gallagher RP, McLean DI, Yang CP, et al. Suntan, sunburn, and pigmentation factors and the frequency of acquired melanocytic nevi in children. Similarities to melanoma: the Vancouver Mole Study. Arch Dermatol 1990;126: Fitzpatrick TB. The validity and practicality of sun-reactive skin types I through VI. Arch Dermatol 1988;124: Gefeller O, Brenner H. Re: Reliability of reported sunburn history in a case-control study of cutaneous malignant melanoma. (Letter). Am J Epidemiol 1996;144: Weinstock MA, Colditz GA, Willett WC, et al. Recall (report) bias and reliability in the retrospective assessment of melanoma risk. Am J Epidemiol 1991;133:240 5.