Logo of amjphAmerican Journal of Public Health Web SiteAmerican Public Health Association Web SiteSubmissionsSubscriptionsAbout Us
Am J Public Health. 2007 March; 97(3): 559–565.
PMCID: PMC1805012

Promoting Sun Safety Among US Postal Service Letter Carriers: Impact of a 2-Year Intervention


Objectives. We examined whether US Postal Service letter carriers who received a sun safety intervention would wear wide-brim hats and sunscreen significantly more often than those who did not receive the intervention.

Methods. We used a 2-group randomized design with 2662 evaluation cohort participants from 70 US postal stations. Evaluations were conducted at baseline, 3 months, 1 year, and 2 years. Questionnaire items assessed occupational use of sun-screen and wide-brim hats. The 2-year sun safety intervention included the provision of wide-brim hats, accessible sunscreen, reminders, and 6 educational sessions.

Results. At the 3-month follow-up evaluations, the odds ratio (OR) for regular sun-screen use was 2.8 times higher among the intervention group than among the control group (95% confidence interval [CI]=2.2, 3.5); at the 2-year follow-up evaluations, the rate was still significantly higher (OR=2.0; 95% CI=1.6, 2.6). Intervention group participants also had significantly higher rates of hat use, with the differences remaining consistent across all follow-ups (OR=2.9; 95% CI=2.3, 3.6).

Conclusions. The intervention should be disseminated to postal stations nationwide and possibly to other occupational groups that work outdoors.

In 2006, approximately 62 190 people in the United States were diagnosed with melanoma, and more than 1 million were diagnosed with nonmelanoma skin cancers.1 Exposure to solar ultraviolet radiation (UVR) is a key risk factor for these diseases.25 Outdoor workers may receive 6 to 8 times the dose of UVR that indoor workers receive,6 and they have a higher risk for squamous cell cancer7,8 and possibly other forms of skin cancer.9,10

In a descriptive study, US Postal Service (USPS) letter carriers reported working outdoors an average of 5 hours per day, but only 14% reported that they always wore sun-screen, and only 33% reported that they always wore a hat.11 Using these data and the results from multiple formative evaluation strategies, we developed an intervention for increasing specific sun safety behaviors among this population. The intervention targeted the regular occupational use of wide-brim hats and sunscreen with a SPF of 15 or higher for 2 reasons: (1) previous data have shown the efficacy of these strategies in reducing UVR exposure and its consequences,1226 and (2) these target behaviors likely would be feasible for letter carriers to perform. We predicted that letter carriers at the postal stations assigned to receive the sun safety intervention would wear wide-brim hats and sunscreen while on the job more often than would letter carriers at the control postal stations.


Design and Settings

Project SUNWISE was a randomized controlled trial of a sun safety intervention, with randomization at the USPS station level. The 70 participating postal stations represented 3 geographic regions in Southern California: San Diego County (n=53), Riverside County–nondesert and San Bernardino County (n = 11), and Riverside County–desert (n = 6). The local postmaster and the USPS district manager ensured the participation of these stations. To determine the sample size, we considered the following: (1) a projected minimal-effect size, (2) the potential clustering of responses within postal stations with the intraclass correlation coefficient,27 and (3) a projected 50% minimal-participation rate by employees within these stations. To meet the required number of stations, all but 2 stations in San Diego County were included, and the remaining stations were those located closest to San Diego County. Employees at the remaining 2 San Diego stations provided input on all intervention strategies during their development. Stations were stratified by region and were randomly assigned to either intervention or control (delayed intervention) group. Within each station, we attempted to recruit all letter carriers to participate voluntarily in the evaluation cohort; small incentives valued at less than $10 each were used to promote participation in each evaluation. Data were collected at baseline (late June 2001–late August 2001), and outcome data were collected 3 months, 1 year, and 2 years later. At the baseline evaluation, letter carriers were unaware of their station’s study group. Data collectors were blinded to group assignment at baseline. Until the completion of the 2-year evaluation, control stations received evaluation procedures only. After that, they received all components of the intervention during a 1-year period.

Dependent Measures

The primary outcomes were self-reported occupational use of sunscreen with an SPF of 15 or higher and wide-brim hats (≥ 2.5 inches wide). These behaviors were measured onsite with a paper-and-pencil questionnaire,11 which asked how often respondents had used each item while carrying mail during their past 5 workdays. Response options were never, sometimes, about half the time, often, and always. In all analyses, data for these variables were dichotomized by always versus all other responses, because consistent occupational use of sun-screen and wide-brim hats were the main goals of the intervention.

To validate the hat-wearing item during the baseline, 3-month, and 1-year evaluations, research assistants directly observed a portion of intervention and control station participants and recorded clothing worn during mail delivery. These procedures have been described elsewhere.28 During informed consent procedures, letter carriers were told that they may be observed several times during the study; however, the timings of observations were not specified. Observations were typically conducted within 1 week before survey administration.

The sunscreen-wearing item also was validated with a station-level approach at intervention stations only. The amount of sun-screen removed from the communal pump bottles provided by Project SUNWISE was monitored at each station for each intervention year and was adjusted by the number of letter carriers. We performed a correlational test with that value and with each station’s percentage of cohort who reported “always used sunscreen in past 5 workdays” at the 1-year and 2-year follow-ups.

Dimensions of skin color known to be associated with cumulative UVR exposure29,30 were assessed by trained data collectors during each measurement session with colorimeters (Minolta Chroma Meter 300); between-group differences at each time point in these dimensions served as secondary outcomes. The 2 color dimensions were (1) L*, which quantifies color on a scale from black to white, with higher scores indicating lighter or less tanned, and (2) b*, which quantifies color on a scale from blue to yellow, with higher scores indicating yellower or more tanned.

Three sites on each participant’s face (right cheek, left cheek, and forehead) were measured twice by the primary data collector; approximately every third participant was measured independently by a second data collector. For each time point, a composite face score for L* and a composite face score for b* were computed and were used in the analysis. For example, to compute the composite L* face score for a given time point, we first averaged the 2 values obtained by the data collector for each facial site. The mean L* scores for the left and right cheeks were then averaged to form the composite cheek L* score. Finally, the mean of the composite cheek L* score and the forehead L* score were averaged to form the composite face L* score. At each time point, scores for each of the 3 sites were highly inter-correlated for both L* and b*.

Intervention Procedures

After completion of the baseline measurement, all letter carriers at each intervention station—regardless of their participation in the evaluation cohort—were included in a multicomponent intervention. The intervention was based on an ecological model of behavior that emphasizes the roles of environment and policy.31 The intervention also was based on key constructs from operant models, which emphasize reinforcement and environmental prompts for changing behavior,32 and the Social Learning Theory’s constructs of modeling, social influence, reciprocal determinism, and self-efficacy.33 One objective of Project SUNWISE was to develop social and physical environments that were supportive of sun safety; therefore, the intervention package included (1) increased access to protective hats, (2) increased access to sunscreen, (3) visual cues that prompted use of solar protective strategies, and (4) delivery of brief educational sun safety messages. Whereas the first 3 components were primarily environment-based, the educational sessions created opportunities for modeling and social influence.

Within 2 weeks of the baseline measurement, Project SUNWISE health educators visited intervention stations to give out hats, install and dispense sunscreen, distribute materials that prompted use of solar protective strategies, and deliver the initial educational presentation. When the project was planned, the existing brim hat that was part of the USPS uniform was judged by our research team to be incompatible with intervention goals because 1 of the 2 models of the hat had perforations and because formative research had shown that letter carriers perceived both models of the hat to be uncomfortable. Consequently, we obtained permission for letter carriers at participating intervention stations to wear the hat that was distributed free of charge by Project SUNWISE staff. The color, fabric, weight, and style of the hat were determined using input from both letter carriers and local USPS officials. The machine-washable dark blue hat was made of Cordura nylon, it had a brim that was 4 inches wide in the front and back and 3 inches wide on the sides, and it had an adjustable cord chin strap. In addition to the initial free hat provided by Project SUNWISE, letter carriers at intervention stations were given discounts on replacement hats by the vendor (Watership Trading Companie, Bellingham, Wash).

Locker rooms at intervention stations were stocked with large pump bottles of sunscreen (Coppertone Sport, SPF 30, Schering-Plough HealthCare Products, Inc, Memphis, Tenn) that were refilled regularly by the research staff. Additionally, letter carriers were given free 12-ounce bottles of the sunscreen, which they could refill with sunscreen from the pump bottles. The decision about which sunscreen to use was made on the basis of formative work that identified a product with a high SPF that had an acceptable fragrance and consistency and minimal rub-off from newsprint onto skin.

To encourage sun protective strategies, Project SUNWISE staff distributed water bottles, mouse pads, key chains, and magnetic clips that had sun safety messages. An “ask the skin doctor” poster was placed in the break room and was updated monthly; it had 2 sun safety questions (from letter carriers) and brief answers from one of the authors (L. F. E), who was the Project SUNWISE dermatologist.

Finally, Project SUNWISE health educators delivered 6 brief onsite educational presentations over 2 years. The 5- to 10-minute presentations were modeled after the “stand-up talks” letter carriers regularly participated in; the educators used large flip charts with colorful graphics that were tailored to letter carriers. Key points of the introductory presentation included the amount of UVR carriers are exposed to and UVR as a skin cancer risk factor, a case example of a former carrier who recently had a precancerous growth removed, feasible protection strategies, and specific information about the hats and sunscreen. The themes of subsequent presentations were (1) importance of sun safety, even in winter; (2) sun safety for the eyes; (3) sharing sun safety tips with loved ones; (4) relevance of sun safety to letter carriers of all races/ethnicities; and (5) recap and encouragement to continue practicing sun safety behaviors.

Statistical Analysis

We used descriptive statistics to characterize both the control and the intervention samples. Intraclass correlations were estimated for all outcomes to determine the extent of postal station clustering; the correlations suggested that clustering must be taken into account in the analyses. For the primary outcome measures, we analyzed trends over 2 years with generalized linear mixed models that treated 3 months, 1 year, and 2 years as a set of repeated measures on each participant. Additionally, we adjusted for postal station clustering using a multilevel model. All analyses were based on intent to treat (analyzed according to the original random group assignment) and included all participants who contributed data for at least 1 time point past baseline. Analyses were adjusted for the baseline level of the corresponding outcome variable (either wide-brim hat use or sunscreen use). For each analysis, we tested the time-by-group interaction to determine if the intervention effect remained constant over time and the group main effect. All analyses were repeated with adjustments for age, gender, and race/ethnicity. The results were nearly identical to the unadjusted analyses (data not shown).

The secondary outcomes (colorimeter data) were analyzed with mixed-effects regression models that treated 3 months, 1 year, and 2 years as a set of repeated measures on each participant and that adjusted for participants nested in postal stations within a multilevel model. The approach to modeling the colorimeter data was the same as that described for the primary outcomes. Additionally, we conducted subanalyses for White participants only. We used SAS version 8.2 (SAS Institute Inc, Cary, NC) to analyze the data; an alpha level of 0.05 was used to determine significance.


Response Rates and Participant Characteristics

Figure 1 shows the flow of participants in each group, from recruitment through the 2-year evaluation. Between the 1-year and 2-year follow-up evaluations, 2 intervention stations in the same jurisdiction withdrew their participation because of the impact of a union grievance filed by 1 employee, who initially had reservations about the study.

Flow chart of participants in Project SUNWISE through 2 years of follow-up: Southern California, 2001–2003.

Data about the participants have been published elsewhwere28,34,35 and are briefly summarized here. Participants had a mean age of 43.0 years (SD=8.6), had worked for the USPS for 12.4 years (SD=7.9), and had worked outdoors an average of 3.9 hours daily (SD=1.9). The majority were men (69.9%), were married (69.3%), had completed at least some college (71.7%), and reported a relatively low level of sun sensitivity (Fitzpatrick skin types III or IV; 77.0%).36 The sample was 51.3% non-Latino White, 19.3% Latino, 12.4% Asian, 8.3% African American, 4.3% Pacific Islander, 0.6% American Indian, and 3.7% other race/ethnicity. Approximately 5.1% of the participants reported a history of some type of skin cancer. Rates of all variables were comparable between study groups.

Mean high daily temperatures across the baseline, 1-year, and 2-year evaluation periods during direct observations (late June through August) were 74°F (23°C) for San Diego County, 96°F (35°C) for Riverside County–nondesert and San Bernardino County, and 107°F (41°C) for Riverside County–desert.37


As reported elsewhere,28 at baseline (n=1036) there was strong concordance (κ =0.62; 95% CI=0.57, 0.67)38 between the questionnaire response to and direct observation of wide-brim hat use. Concordance between these measures at the 3-month (n=800) and 1-year (n=843) follow-up evaluations also was good (κ =0.54; 95% CI=0.46, 0.62; and κ =0.63; CI=0.56, 0.69, respectively).

At the station level (n = 35 intervention stations during year 1 and 33 intervention stations during year 2), self-reported sunscreen use also was found to be accurate. The correlations between average ounces of communal sunscreen used per letter carrier per station and the station’s percentage of cohort who reported they always wore sunscreen were R = 0.42, P < .05, for year 1 and R = 0.62, P < .001, for year 2.

Table 1 shows the distribution of responses for the 2 primary outcomes at each time point. As already noted, the primary analysis used the proportions of participants in each group that reported they always wore sun-screen or a wide-brim hat while delivering mail during their past 5 workdays. Comparisons at baseline were made using a mixed-effects model that compared the intervention group with the control group after adjusting for the clustering effects of postal stations; the intraclass correlation was 0.014 for sunscreen and 0.091 for wide-brim hat. At baseline, the groups did not differ significantly on either sunscreen use (P = .15) or wide-brim hat use (P = .09) after we adjusted for clustering.

Unadjusted Distributions of Primary Outcomes at 3-Months, 1-Year, at 2-Years of Follow-Up: Project SUNWISE Letter Carriers Study, Southern California, 2001–2003

Table 2 shows results of the generalized linear mixed models that treated the data from 3 months, 1 year, and 2 years as a set of repeated measures on each participant and that adjusted for participants nested within postal stations and baseline level of the corresponding outcome variable. There was a significant interaction for sunscreen use, which was a result of the difference between intervention and control groups diminishing in size over time. At 3 months, the estimated odds ratio (OR) for sunscreen use was 2.8 times higher among the intervention group compared with the control group; at 2 years, the OR was 2.0 times higher among the intervention group. Nevertheless, the difference at 2 years was still highly significant (95% CI=1.60, 2.58; P<.001). The interaction was not significant for wide-brim hat use, which suggests no differential change over time. However, the group main effect was significant, with the intervention group that used wide-brim hats having a higher rate (OR=2.88; 95% CI=2.31, 3.61). After we adjusted for age, gender, and race/ethnicity, results of the tests were substantially the same. The patterns of change in each outcome for intervention group participants (and no change for control group participants) that were found for the entire sample also were found for each racial/ethnic group separately (data not shown).

Results of Generalized Linear Mixed-Model Analyses (N = 2501): at 3-Months, 1-Year, at 2-Years of Follow-Up: Project SUNWISE Letter Carriers Study, Southern California, 2001–2003

Secondary Outcomes

We computed colorimeter interrater agreements with Pearson’s r coefficients at each of the 4 time points for each of the 3 body sites and for each of the 2 color dimensions. With sample sizes ranging from 650 to 845, all coefficients were statistically significant (P < .01) and ranged from 0.75 to 0.98, with 22 of the 24 values (92%) at r = 0.90 or higher.

Table 3 shows the adjusted means at 3 months, 1 year, and 2 years and results of the mixed-effects regression models. For the L* color dimension, neither the group-by-time interaction nor the group main effect was significant. For b*, there was a significant group-by-time interaction (P = .009). The rate of decrease over time (from more tanned or more yellow to less tanned or more blue) was more pronounced among the intervention group compared with the control group. When the data were adjusted for age, gender, and race/ethnicity, comparable results were found. We observed similar patterns in the colorimeter results when the sample was restricted to non-Latino Whites.

Adjusted Means, Standard Errors, and Tests of Significance for Colorimeter Data: Project SUNWISE Letter Carriers Study, Southern California, 2001–2003

Selected Process Data

The number of educational sessions participants attended at the intervention sites was computed on the basis of responses to questionnaire items by those who completed all 3 follow-up surveys (n = 915). The mean number of sessions that participants attended was 5.18 (SD = 1.12). In exploratory analyses with the intervention station cohort, we evaluated whether number of sessions attended (i.e., dose) was associated with the outcomes (always vs other frequencies of wide-brim hat use or sunscreen use during past 5 workdays). Results of a logistic regression analysis controlled for age, gender, and race/ethnicity showed that the odds for reporting always wore a wide-brim hat was 21% higher for each increase in dose (adjusted OR = 1.21; 95% CI = 1.06, 1.38; P = .005). A parallel analysis with sunscreen use as the outcome resulted in an adjusted OR for attendence level of 1.18 (CI = 1.03, 1.34; P = .017).


Results of our randomized controlled trial indicated that a 2-year, multicomponent sun safety intervention that targeted letter carriers was successful in increasing and sustaining regular occupational use of sunscreen and wide-brim hats. We consider the increases from baseline to the 2-year follow-up of the outcomes within the intervention group to be meaningful: 27% to 39% for sunscreen use and 27% to 40% for wide-brim hat use. The fact that the proportions of compliance post-intervention were still far from ideal was in part the result of our choice to define compliance stringently (i.e., always). Yet, we believe the stringency is justified because of the UVR exposure levels among this population. If these intervention effects can be generalized throughout the USPS nationwide (which employs 344 580 letter carriers39), there would be a relatively large core group of sun safety–compliant outdoor workers (i.e., early adopters) who might then influence their coworkers.40 In other words, the combination of our intervention strategies, social influence, and an eventual change in cultural norms within the USPS may result in considerably higher rates of long-term compliance.

The number of sun-protective behavioral studies in occupational settings is small, and the number of randomized controlled trials is even smaller.41 Additionally, the occupations, interventions, and outcome measures that have been targeted in published studies vary widely. Thus, comparisons of findings among previous studies and between those studies and our study should be made with caution. The intervention strategies used in 4 previous randomized controlled trials that targeted sun safety and preventive behaviors included education and training; provision of sun safety items, such as hats and sunscreen; supportive print media, such as posters and pamphlets; or some combination of these strategies.4245 Results across studies were inconsistent: compared with control groups, interventions were not effective in increasing use of hats, sun-screen, or other protective behaviors among outdoor aquatics staff in Hawaii42 or among ski resort staff throughout the United States,43 even though the rates of sunburns decreased in both studies. In contrast, behavioral changes were found in trials with outdoor recreation staff in Hawaii44 and electric company outdoor workers in New South Wales, Australia.45 Although our results also showed intervention effectiveness, more studies are needed41 with comparable measures30 to determine which intervention strategies are effective with which outdoor occupational groups.

Our study has several strengths. First, participation and retention rates of eligible employees were respectable. Second, we used objective measures to verify verbally reported hat use and sunscreen use; the concordance data indicated that participants were accurately reporting their behaviors. Third, the relatively inexpensive intervention package we tested was model-driven and was carefully targeted to address the specific needs and occupational culture of USPS letter carriers.

Among the limitations, findings from the colorimeter data were not consistent between the 2 color dimensions. Changes in the b* color dimension suggest the possibility of a secular trend in face protection to which our intervention had an additive effect. The L* data showed no main or interaction effects. In previous work with child participants, we found that both b* and L* data showed clear, predicted differences between summer and winter.46 However, it is possible that the measure does not perform as well and is not sensitive to actual changes in UVR protective behaviors among adults with long-term year-round sun exposure.

The study design did not permit inferences about which intervention components were more powerful. In future studies, it would be useful to evaluate the effects of the availability of hats and sunscreen both with and without educational sessions. Results from our analysis on the dose–response effect of educational sessions did suggest that this component was effective. Were the intervention package as a whole to be disseminated throughout the USPS, sessions could be delivered by safety captains, by local volunteers from health organizations, or by video. The program could be sponsored by either the USPS or the National Association of Letter Carriers or a combination of the 2 entities.

Our intervention package was successful in increasing sun safety behaviors among USPS letter carriers. It was highly feasible to institutionalize all components of the intervention. Wide-brim hats that are effective for sun protection could be added to the official uniform and could be eligible for the uniform allowance. High-quality sunscreen could be made available at all USPS stations, and sun-screen manufacturers likely would provide discounts to this large consumer group. We have described opportunities for low-cost sun safety education. The Project SUNWISE intervention has the potential to reduce skin cancer risk among the nearly 345000 US letter carriers39 and other outdoor workers who are at risk because of extensive UVR exposure.


This study was funded by the National Institutes of Health, National Cancer Institute (NIH/NCI) (grant numbers R01 CA085980, R01 CA085980S1, R01 CA085980S2, and K05 CA10051).

We are appreciative of the input from our contacts at the United States Postal Service (USPS) and the National Association of Letter Carriers (NALC) during the design and implementation of this study. We thank Jeffrey Flinders and Julie Frank for their assistance with developing the educational sessions, and we thank the other health educators for their competent delivery of the sessions. We thank Greg Frechette of Watership Trading Companie for designing and producing the hat used in the study. We are grateful to Jennifer Renzi, Jessica van der Pers, Maureen LaRoe, Lisa Yee, and Shannon Crosby for their diligence with data collection, entry, and cleaning and to all of the data collectors for their high-quality performance. Katherine Hoerster provided helpful comments on an earlier draft of the manuscript, and Debra Rubio is acknowledged for assistance with manuscript preparation. We especially thank the letter carriers who participated in the evaluation of this project, the letter carriers at the 2 “vanguard” stations who helped us develop the intervention, and the station managers and safety captains who facilitated the study.

Note. The contents of this article are solely the responsibility of the authors and do not necessarily represent the official views of the USPS, NALC, or the NIH/NCI.

Human Participant Protection …Approval for all study-related procedures and materials was obtained from USPS officials (regional) and from the National Association of Letter Carriers—local, regional, and national. All study procedures were approved by the institutional review board of San Diego State University.


Peer Reviewed

Contributors…J. A. Mayer originated and designed the study and supervised all aspects of its implementation. D. J. Slymen developed the data management and quality control procedures and conducted and interpreted the analyses. E. J. Clapp helped design and supervise the intervention procedures and coordinated all aspects of the study. L. Pichon designed and implemented formative evaluation procedures, developed intervention content, supervised data entry and data management procedures, and helped analyze data. L. Eckhardt helped design and supervise data collection procedures. L. Eichenfield developed content for the intervention. J. P. Elder helped design intervention protocols. J. F. Sallis helped design intervention and evaluation protocols. M. Weinstock helped design intervention protocols. A. Achter helped design and supervise data collection and intervention procedures. C. Balderrama helped to refine and deliver the educational sessions. G. Galindo designed and implemented formative evaluation procedures and developed intervention content. S. S. Oh helped design and supervise data collection and data management procedures. All authors originated ideas, interpreted findings, and reviewed drafts of the manuscript.


1. American Cancer Society. Cancer Facts and Figures 2006. Atlanta, Ga: American Cancer Society; 2006.
2. Friedman RJ, Rigel DS, Bersons DS, Rivers J. Skin cancer: basal cell and squamous cell carcinoma. In: Holleb AJ, Fink DJ, Murphy GP, eds. American Cancer Society Textbook of Clinical Oncology. Atlanta, Ga: American Cancer Society; 1991:290–305.
3. Singletary S, Balch C. Malignant melanoma. In: Holleb AJ, Fink DJ, Murphy GP, eds. American Cancer Society Textbook of Clinical Oncology. Atlanta, Ga: American Cancer Society; 1991:263–270.
4. Marks R. An overview of skin cancers. Incidence and causation. Cancer Suppl. 1995;75:607–612. [PubMed]
5. Elwood JM. Recent developments in melanoma epidemiology, 1993. Melanoma Res. 1993;3:149–156. [PubMed]
6. Holman CD, Gibson IM, Stephenson M, Armstrong BK. Ultraviolet irradiation of human body sites in relation to occupation and outdoor activity: field studies using personal UVR dosimeters. Clin Exp Dermatol. 1983;8:269–277. [PubMed]
7. English DR, Armstrong BK, Kricker A, Fleming C. Sunlight and cancer. Cancer Causes Control. 1997;8: 271–283. [PubMed]
8. Kricker A, Armstrong BK, English DR. Sun exposure and non-melanocytic skin cancer. Cancer Causes Control. 1994;5:367–392. [PubMed]
9. Strickland PT, Vitasa BC, West SK, Rosenthal FS, Emmett EA, Taylor HR. Quantitative carcinogenesis in man: solar ultraviolet B dose dependence of skin cancer in Maryland watermen. J Natl Cancer Inst. 1989; 81:1910–1913. [PubMed]
10. Delzell E, Grufferman S. Mortality among white and nonwhite farmers in North Carolina, 1976–1978. Am J Epidemiol. 1985;121:391–402. [PubMed]
11. Stepanski BM, Mayer JA. Solar protection behaviors among outdoor workers. J Occup Environ Med. 1998;40:43–48. [PubMed]
12. Keeling JH, Kraus EW, Pathak M, Sober AJ. Hats: design and protection from ultraviolet radiation. Mil Med. 1989;154:250–255. [PubMed]
13. Wong JC, Airey DK, Fleming RA. Annual reduction of solar UV exposure to the facial area of outdoor workers in Southeast Queensland by wearing a hat. Photodermatol Photoimmunol Photomed. 1996;12:131–135. [PubMed]
14. Diffey BL, Cheeseman J. Sun protection with hats. Br J Dermatol. 1992;127:10–12. [PubMed]
15. Rosenthal FS, Phoon C, Bakalian AE, Taylor HR. The ocular dose of ultraviolet radiation to outdoor workers. Invest Ophthalmol Vis Sci. 1988;29:649–656. [PubMed]
16. Naylor MF, Boyd A, Smith DW, Cameron GS, Hubbard D, Neldner KH. High sun protection factor sunscreens in the suppression of actinic neoplasia. Arch Dermatol. 1995;131:170–175. [PubMed]
17. Thompson SC, Jolley D, Marks R. Reduction of solar keratoses by regular sunscreen use. N Engl J Med. 1993;329:1147–1151. [PubMed]
18. Damian DL, Halliday GM, Barnetson RS. Broad-spectrum sunscreens provide greater protection against ultraviolet-radiation-induced suppression of contact hypersensitivity to a recall antigen in humans. J Invest Dermatol. 1997;109:146–151. [PubMed]
19. Davenport V, Morris JF, Chu AC. Immunologic protection afforded by sunscreens in vitro. J Invest Dermatol. 1997;108:859–863. [PubMed]
20. van Praag MC, Roza L, Boom BW, et al. Determination of the photoprotective efficacy of a topical sun-screen against UVB-induced DNA damage in human epidermis. J Photochem Photobiol B. 1993;19:129–134. [PubMed]
21. Ananthaswamy HN, Loughlin SM, Cox P, Evans RL, Ullrich SE, Kripke ML. Sunlight and skin cancer: inhibition of p53 mutations in UV-irradiated mouse skin by sunscreens. Nat Med. 1997;3:510–514. [PubMed]
22. Elmets CA, Vargas A, Oresajo C. Photoprotective effects of sunscreens in cosmetics on sunburn and Langerhans cell photodamage. Photodermatol Photoimmunol Photomed. 1992;9:113–120. [PubMed]
23. Neale R, Russell A, Muller HK, Green A. Sun exposure, sunscreen and their effects on epidermal Langerhans cells. Photochem Photobiol. 1997;66:260–264. [PubMed]
24. Weinstock M. The role of sunscreen lotions in melanoma prevention. In: Grob J, Stern R, MacKie R, Weinstock M, eds. Epidemiology, Causes, and Prevention of Skin Diseases. Oxford, UK: Blackwell; 1997:159–163.
25. Gasparro FP, Mitchnick M, Nash JF. A review of sunscreen safety and efficacy. Photochem Photobiol. 1998;68:243–256. [PubMed]
26. Naylor MF, Farmer KC. The case for sunscreens. A review of their use in preventing actinic damage and neoplasia. Arch Dermatol. 1997;133:1146–1154. [PubMed]
27. Donner A. An empirical study of cluster randomization. Int J Epidemiol. 1982;11:283–286. [PubMed]
28. Oh SS, Mayer JA, Lewis EC, et al. Validating outdoor workers’ self-report of sun protection. Prev Med. 2004;39:798–803. [PubMed]
29. Creech LL, Mayer JA. Ultraviolet radiation exposure in children: a review of measurement strategies. Ann Behav Med. 1997;19:399–407. [PubMed]
30. Glanz K, Mayer JA. Reducing ultraviolet radiation exposure to prevent skin cancer: methodology and measurement. Am J Prev Med. 2005;29:131–142. [PubMed]
31. Sallis JF, Owen N. Ecological models of health behavior. In: Glanz K, Rimer BK, Lewis FM, eds. Health Behavior and Health Education: Theory, Research, and Practice. 3rd ed. San Francisco, Calif: Jossey-Bass; 2002:462–484.
32. Skinner BF. Science and Human Behavior. New York, NY: Macmillan; 1953.
33. Bandura A. Social Functions of Thought and Action: A Social Cognitive Approach. Englewood Cliffs, NJ: Prentice Hall; 1986.
34. Pichon LC, Mayer JA, Slymen DJ, Elder JP, Lewis EC, Galindo GR. Ethnoracial differences among outdoor workers in key sun-safety behaviors. Am J Prev Med. 2005;28:374–378. [PubMed]
35. Lewis EC, Mayer JA, Slymen D. Postal workers’ occupational and leisure-time sun safety behaviors. Cancer Causes Control. 2006;17:181–186. [PubMed]
36. Fitzpatrick TB. The validity and practicality of sun-reactive skin types I through VI. Arch Dermatol. 1988;124:869–871. [PubMed]
37. National Weather Service Forecast Office. San Diego, Calif. Available at: http://www.wrh.noaa.gov/sgx. Accessed August 22, 2005.
38. Rosner B. Fundamentals of Biostatistics. 4th ed. Belmont, Calif: Wadsworth Publishing; 1995.
39. US Department of Labor, Bureau of Labor Statistics. Occupational employment and wages, May 2003. Available at: http://www.bls.gov/oes/2003/may/oes435052.htm#nat. Accessed September 24, 2005.
40. Rogers EM. Diffusion of Innovations. 4th ed. New York, NY: Free Press; 1995.
41. Saraiya M, Glanz K, Briss PA, et al. Interventions to prevent skin cancer by reducing exposure to ultraviolet radiation: a systematic review. Am J Prev Med. 2004;27:422–466. [PubMed]
42. Geller AC, Glanz K, Shigaki D, Isnec MR, Sun T, Maddock J. Impact of skin cancer prevention on outdoor aquatics staff: the Pool Cool program in Hawaii and Massachusetts. Prev Med. 2001;33:155–161. [PubMed]
43. Buller DB, Andersen PA, Walkosz BJ, et al. Randomized trial testing a worksite sun protection program in an outdoor recreation industry. Health Educ Behav. 2005;32:514–535. [PubMed]
44. Glanz K, Maddock J, Lew R, Murakami-Akatsuka L. A randomized trial of the Hawaii SunSmart program’s impact on outdoor recreation staff. J Am Acad Dermatol. 2001;44:973–978. [PubMed]
45. Girgis A, Sanson-Fisher RW, Watson A. A work-place intervention for increasing outdoor workers’ use of solar protection. Am J Public Health. 1994;84: 77–81. [PMC free article] [PubMed]
46. Eckhardt L, Mayer JA, Creech L, et al. Assessing children’s ultraviolet radiation exposure: the potential usefulness of a colorimeter. Am J Public Health. 1996; 86:1802–1804. [PMC free article] [PubMed]

Articles from American Journal of Public Health are provided here courtesy of American Public Health Association
PubReader format: click here to try


Save items

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...