Abstract
Solar ultraviolet radiation (UV) exposure is the primary risk factor for skin cancer and children receive about one-quarter of lifetime UV exposure before age 18. Thus, skin cancer prevention is essential for children. The objective of this study was to test an intervention to facilitate implementation of district sun safety policies. Elementary schools (N=118) from 40 California public school districts with a school board-approved policy for sun safety were recruited along with one principal and teacher from each school. Elementary schools were randomly assigned to receive the 20-month Sun Safe Schools intervention (N=58) or to a minimal informational control condition (N=60). Principals were 47.8 and teachers 44.3 years of age and were predominately female (principals 72.7%; teachers 86.7%) and White (principals 72.0%; teachers 68.2%). The 20-month Sun Safe Schools intervention was delivered to principals and included a coaching session, telephone and email contacts, provision of resources for practice implementation, and a small grant program. The principal (N=118) and a teacher (N=113) at each school reported on school sun protection practices 20-months post-baseline. The schools were diverse in student characteristics (M=64.1% free/reduced price meals; M=54.5% Hispanic). Intervention principals reported implementing more sun safety practices overall (control M=2.7, intervention M=4.2, p<0.005) and more practices not present in the district’s policy (control M=0.4, intervention M=0.9, p=0.005). Principals and teachers combined replicated these findings and also reported implementing more practices present in the district policy (control M=0.9, intervention M=1.3, p=0.005). In sum, the intervention increased sun safety practices in public elementary schools.
Trial Registration:
This study is registered in ClinicalTrials.gov, www.clinicaltrials.gov, with the identification number of NCT03243929.
Keywords: children, policy, primary prevention, schools, school control policies, school-based intervention, skin cancer
Introduction
In 2020, over 100,350 cases of melanoma and five million cases of keratinocyte skin cancers will be diagnosed, and melanoma will kill 6,850 Americans.1 Exposure to solar ultraviolet radiation (UV) is the primary risk factor for skin cancer.2,3 It is estimated that children receive approximately one-quarter to one-half of lifetime UV exposure,4,5 spend 3 hours/day outdoors,6 including at school,7 are not routinely protected,8,9 and are frequently sunburned,9,10 increasing their risk for skin cancer as adults. Thus, skin cancer prevention is essential for children and the delivery of effective interventions that can instill lifelong sun safety habits is imperative.
The Centers for Disease Control and Prevention (CDC) have recommended the inclusion of sun protection in educational settings and school policies11 and have issued national guidelines on skin cancer prevention12 that recommend schools a) modify schedules and the physical environment to avoid outdoor activities at midday and increase shade availability; b) encourage students to wear protective clothing, hats, sunglasses and sunscreen; c) enact education on skin cancer prevention; d) advise children and parents about sun protection; and e) provide resources, accountability and evaluation of school sun safety. Despite these recommendations, many schools have not implemented skin cancer prevention programs, policies or practices.13 District administrators may feel UV poses a low risk and that the feasibility of skin cancer prevention is reduced by competing educational priorities and limited financial resources (e.g., to build shade structures).
Results of a randomized trial showed that school district administrators can be convinced to adopt district policies on sun safety.14 Despite adoption of district-level policies, many principals and teachers remain uninformed about the policies15 and few schools are taking action on sun safety.16 Researchers have explored factors associated with implementation of sun safety practices in primary schools including geographic region,17 population density,18 grade level,17,18 enrollment size,17,19 student socioeconomic status17, and resources for implementation.20 These studies provide little information useful for interventions to facilitate policy intervention. A prospective randomized trial testing an intervention that provides resources for implementation would offer better evidence on the drivers of policy implementation. Few such studies have been conducted. Evidence from a randomized trial conducted in Australia indicates that a school policy support intervention may increase the use of sunscreen in the schools21,22 and in a separate study a quasi-experimental comparison suggested that the SunSmart schools program in Australia may promote the development of more comprehensive school-level policies.23
Despite the potential benefits, we are unaware of any evaluated programs designed to promote school-level implementation of district sun safety policy. In this paper, we present results of a randomized trial evaluating an intervention intended to close this policy-practice gap, Sun Safe Schools (SSS). The analyses tested the hypothesis that schools randomly assigned to receive the SSS intervention will implement more sun safety practices than schools in a minimal information control group.
Methods
Design of the Randomized Trial
All procedures were approved by the Claremont Graduate University (CGU) Institutional Review Board (IRB) and the Western IRB. Principals of public elementary schools were contacted through email and telephone and invited to participate. Principals (N=118) consented for themselves and for the school, completed a baseline survey, and nominated one teacher to complete the baseline survey (N=113). Following pretesting of principals and teachers, schools were randomly assigned to either the intervention (N=58 schools) or control (N=60 schools) group by the project statistician using the Ranuni generator in SAS.24,25 Schools enrolled in the past month were randomized at the beginning of each month to experimental conditions. Block size ranged from 1 to 17 schools per month (Mblock size = 6 schools). The assignment status of schools was known only to the project statistician and the project coordinator in Colorado. Research staff in California who were responsible for school recruitment and program delivery were blind to treatment assignment, as were school personnel, until immediately prior to the initiation of intervention. At that time, treatment assignments were communicated to the Project Manager in California and letters were sent to the principals informing them that they would receive either program A or program B along with a brief description of the elements of the respective interventions. Intervention schools received the SSS intervention over 20 months, while control schools received mailings with sun safety information (e.g., CDC’s Guidelines for School Programs to Prevent Skin Cancer). Principals and teachers were sent a computer-generated invitation to complete a posttest survey 20-months after consenting. Surveys were completed online. Respondents received email and telephone reminders and $10 for completing each survey. The use of an online survey methodology eliminated direct contact between research staff and principals and teachers, reducing the risk of biasing effects from research staff. Recruitment and baseline assessment occurred between January 2014 and May 2016 and posttest assessment between November 2015 and March 2018. The trial ended as designed with completion of assessments within the time period available for posttesting among the schools.
Sun Safe Schools Intervention
In the SSS intervention, sun safety coaches met by telephone with principals in an initial intervention session (IIS) to: a) review the district’s sun safety policy, b) complete a sun safety practice checklist identifying up to three practices to implement, c) plan for implementation, and d) provide sun safety resources. The intervention was built on organizational innovation principles in Diffusion of Innovation Theory (DIT).26,27 All school districts in the trial had previously adopted a policy on sun safety (i.e., Board Policy [BP] 5141.7). The SSS intervention provided guidance to a) show how sun protection practices fit the school's needs and environment; b) plan the implementation of practices; c) adapt the practices to the school; and d) communicate with school staff, students, and parents.
To provide principals with resources, coaches were trained to assess readiness for implementing practices using a staging matrix based on DIT. This matrix matched the innovation stage to sun safe practice resources within each policy category. For example, in the Refining/Restructuring stage, principals received information on modifying sun safety practices for their school and/or modifying the school's processes to support implementation. In addition, the model provided guidance to the coach based on stage and suggesting that for Agenda Setting, coaches should promote the importance of sun safety overall as well as the importance of specific policy components; for Matching they would help principals and other school personnel see how sun safety policy can be implemented into processes the school already has in place; for Refining/Restructuring coaches should assist with formulating an action plan and provide resources as the implementation is rolled out; and for Clarifying address barriers the implementers might face and plan for ways to adjust the action plan.
Coaches sent a monthly email containing information pieces aimed at skin cancer prevention topics, tailored to principals' readiness, needs and interest. In follow-up communications, coaches encouraged implementation, received updates, helped solve barriers to implementation, provided resources, and encouraged principals to implement additional practices. Of the 58 intervention principals, 91% met with coaches and 100% selected sun safety practices to implement (38% selected 1-4 actions; 43% 5-8 actions; 13% 9-12 actions; 6% 13-16 actions).
Sample of Public School Districts and Elementary Schools
Elementary schools were eligible for the trial if they a) were in California; b) included students in grades K-6; and c) were in public school districts that posted their sun safety policy, BP5141.7, online. The California School Boards Association (CSBA) provided lists of the 124 member districts that had adopted BP5141.7. Project staff contacted principals at all 475 elementary schools in 56 school districts by email and telephone. Order of contact was stratified by distance from the project’s main office and number of elementary schools within the district. All pretested principals and teachers were invited to complete the posttest; if they had left the school, their replacements were invited to complete the survey (see CONSORT diagram, Figure 1). It was estimated that a sample of 114 schools at posttest would provide 80% power to detect a difference of 15% in the proportion of schools that implement at least one new sun safety practice between control (20%) and intervention (35%) at the p<.05 level, two-sided.
1.
CONSORT Diagram for Trial
Policy Collection and Verification
To verify the eligibility of schools, BP 5141.7 policy documents for the participating districts were coded assessing the presence of policy content (i.e., 0=not present; 1=present) in 11 categories covering components in the CDC’s guidelines12 (i.e., sunscreen use; UV protective clothing; hats; education of students; education of teachers; outdoor shade; scheduling; parent outreach; resource allocation; accountability; and staff modeling). Fifteen percent (n=29) of policies were coded by a second coder. Agreement was calculated using the Kappa statistic which ranged from 0.74 to 1.00 with a mean of 0.9428.
Measures
In the pretest, principals and teachers reported whether the school district had a policy on sun protection for students. Those who answered affirmatively were presented with a list of policy components (e.g., modifying the schedule to avoid high UV) and asked to indicate which components were included in the policy. Participants also responded to items on their attitudes toward sun protection of students, personal sun protection practices, sunburn history, job characteristics (e.g., years in public education), and demographics. At posttest, principals and teachers were also asked about costs incurred when implementing sun safety practices, resources received and used from the program, and sun safety material obtained outside of the SSS program.
The implementation of sun safety practices at the school, was assessed by asking principals and teachers which of a list of sun safety practice categories had been implemented since the last assessment (Yes, No, Don’t Know). If they answered affirmatively for one of these categories, they were presented with specific actions within the category and responded as to whether they had done them (Yes, No, Don’t Know). At baseline, only principals and teachers who were aware of the district board sun safety policy completed this measure. However, during the intervention, coaches learned that some schools had implemented practices prior to the study even though the principal was unaware of the policy. Thus, at posttest, all principals and teachers were asked to report on sun safety practices. Four indicators of implementation were created (calculated for all practices; for practices reflecting district board policy, and for practices not reflected in district board policy), including a) a dichotomous indicator with a Yes value if at least one sun safety practice was implemented in any category; b) number of sun safety practices implemented; c) percentage of categories implemented within the BP 5141.7 policy adopted by the district; and d) number of sun safety practices implemented that were not within the district’s BP5141.7 policy.
School Characteristics
The number of students and administrators (size) and proportion of minority students, English learners, and students in free or reduced-price meal programs, and distance from Claremont, CA (location) were obtained from the California Department of Education records. We did not collect data on the use of school uniforms at each school. Public schools were recruited which typically do not require the wearing of a school uniform.
Statistical Analysis
The effects of intervention on implementation of sun safety practices were modeled for principals and teachers separately and then by combining the two informants in each school, using multi-level analysis with individuals (principal or teacher) nested within schools. Whether the principal/teacher was originally in the sample at baseline or new to the sample at posttest, and school and respondent characteristics, were included as covariates. Initially, models were run including pretest values of sun safety practices but did not converge due to the number of principals and teachers unaware of the district policy. Post-only analyses of the outcome measures were then conducted including baseline covariates but without adjustment for baseline sun safety practices. Post-only analyses can yield valid estimates of intervention effects in randomized trials, when randomization equalizes groups on potential covariates as it did here and adjustments are provided for attrition including through the use of imputation strategies.29 Thus, the analysis was completed twice, first on completers and then using multiple imputation (Ml) for the ten principals and seventeen teachers who did not complete the posttest30 PROC GLIMMIX in SAS 9.3 was used to fit mixed models for logistic regression on binary outcomes or Poisson regression on count outcomes. PROC MIXED was used to fit mixed models for linear regression on percentage outcomes. Alpha criterion levels were set at 0.05 (two-tailed).
Three school characteristics were used as auxiliary variables in the Ml process: total number of students enrolled in school, percent of students in free or reduced-price meals program, and percent of White non-Hispanic students. Due to the monotone missing data pattern, a monotone logistic regression and a monotone regression method were chosen for the binary and continuous variables respectively. Five versions of principal data and five versions of teacher data, with the missing values replaced by imputed values, were created separately using PROC Ml in SAS and each was then analyzed using multi-level analysis, the same modeling techniques that were used for the completers’ analysis. The parameter estimates and the associated covariance matrices obtained were then used to derive the statistical inferences about the parameters, employing PROC MIANALYZE in SAS.
Results
Profile of the Samples
The schools were located in 17 counties (M=133.5 miles [sd=180.1] from CGU), and were diverse in size (M=564.6 students [sd=215.6]) and student characteristics (M=27.8% English-learners [sd=17.0%]; M=64.1% in free/reduced price meal program [sd=28.8%]; M=54.5% Hispanic [sd=26.4%]; M=24.0% non-Hispanic White [sd=22.3%]). Sample characteristics were compared across the treatment conditions at baseline with no statistically significant differences detected (see Table 1).
Table 1:
Profile of the Samples of Principals and Teachers
| Characteristic | Principals | Teachers | ||||||
|---|---|---|---|---|---|---|---|---|
| All (n=118) Mean ± SD or No. (%) |
Control (n=60) Mean ± SD or No. (%) |
Intervention (n=58) Mean ± SD or No. (%) |
p | All (n=113) Mean ± SD or No. (%) |
Control (n=58) Mean ± SD or No. (%) |
Intervention (n=55) Mean ± SD or No. (%) |
p | |
| Years Working in Public Education | 21.1 ± 7.1 | 21.5 ± 6.6 | 20.7 ± 7.6 | .57 | 14.3 ± 7.3 | 14.1 ± 7.5 | 14.5 ± 7.2 | .76 |
| Years Working as a Principal/Teacher in Current School District | 4.6 ± 4.0 | 4.1 ± 3.6 | 5.2 ± 4.4 | .14 | 11.6 ± 7.5 | 11.7 ± 7.6 | 11.6 ± 7.4 | .96 |
| Years Working in Any Position in Current School District | 11.4 ± 9.2 | 10.6 ± 8.9 | 12.4 ± 9.4 | .29 | NA | NA | NA | NA |
| Grades Currently Taught* | ||||||||
| K | NA | NA | NA | NA | 19 (17.0) | 10 (17.2) | 9 (16.7) | .94 |
| 1 | 20 (17.9) | 10 (17.2) | 10 (18.5) | .86 | ||||
| 2 | 21 (18.8) | 9 (15.5) | 12 (22.2) | .36 | ||||
| 3 | 24 (21.4) | 13 (22.4) | 11 (20.4) | .79 | ||||
| 4 | 24 (21.4) | 15 (25.9) | 9 (16.7) | .24 | ||||
| 5 | 21 (18.8) | 12 (20.7) | 9 (16.7) | .59 | ||||
| 6 | 11 (9.8) | 5 (8.6) | 6 (11.1) | .66 | ||||
| 7 or higher | 3 (2.7) | 2 (3.5) | 1 (1.8) | .60 | ||||
| No specific grade | 51 (45.1) | 24 (41.4) | 27 (49.1) | .41 | ||||
| Currently Teach Health or Science Curriculum | ||||||||
| Yes | NA | NA | NA | NA | 47 (42.0) | 28 (49.1) | 19 (34.5) | .12 |
| No | 65 (58.0) | 29 (50.9) | 36 (65.5) | |||||
| Skin Type | ||||||||
| 1 (darkest) | 33 (28.7) | 20 (35.1) | 13 (22.4) | .47 | 29 (26.4) | 15 (26.8) | 14 (25.9) | .93 |
| 2 | 33 (28.7) | 14 (24.6) | 19 (32.8) | 30 (27.3) | 14 (25.0) | 16 (29.7) | ||
| 3 | 29 (25.2) | 13 (22.8) | 16 (27.6) | 25 (22.7) | 13 (23.2) | 12 (22.2) | ||
| 4 | 20 (17.4) | 10 (17.5) | 10 (17.2) | 23 (20.9) | 13 (23.2) | 10 (18.5) | ||
| 5 (lightest) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 3 (2.7) | 1 (1.8) | 2 (3.7) | ||
| Family History of Skin Cancer | ||||||||
| Yes | 46 (39.3) | 24 (40.0) | 22 (38.6) | .88 | 48 (43.2) | 26 (46.4) | 22 (40.0) | .49 |
| No | 71 (60.7) | 36 (60.0) | 35 (61.4) | 63 (56.8) | 30 (53.6) | 33 (60.0) | ||
| Age, Years | 47.8 ± 7.7 | 48.6 ± 7.9 | 47.0 ± 7.4 | .28 | 43.4 ± 10.0 | 42.0 ± 9.4 | 44.8 ± 10.3 | .14 |
| Hispanicity | ||||||||
| Hispanic | 25 (22.3) | 13 (22.8) | 12 (21.8) | .90 | 25 (23.4) | 11 (20.7) | 14 (25.9) | .53 |
| Not Hispanic | 87 (77.7) | 44 (77.2) | 43 (78.2) | 82 (76.6) | 42 (79.3) | 40 (74.1) | ||
| Race | ||||||||
| American Indian/Alaska | ||||||||
| Native | 4 (3.5) | 2 (3.5) | 2 (3.6) | .30 | 5 (4.5) | 0 (0.0) | 5 (9.3) | .18 |
| Asian | 1 (0.8) | 0 (0.0) | 1 (1.8) | 8 (7.3) | 5 (8.9) | 3 (5.5) | ||
| Black/African American | 10 (8.8) | 6 (10.3) | 4 (7.1) | 5 (4.5) | 3 (5.4) | 2 (3.7) | ||
| Native Hawaiian/Other | ||||||||
| Pacific Islander | 1 (0.8) | 0 (0.0) | 1 (1.8) | 3 (2.7) | 1 (1.8) | 2 (3.7) | ||
| White | 82 (72.0) | 38 (65.5) | 44 (78.6) | 75 (68.2) | 39 (69.6) | 36 (66.7) | ||
| More than one race | 6 (5.3) | 5 (8.6) | 1 (1.8) | 4 (3.7) | 1 (1.8) | 3 (5.6) | ||
| None of these | 10 (8.8) | 7 (12.1) | 3 (5.3) | 10 (9.1) | 7 (12.5) | 3 (5.5) | ||
| Gender | ||||||||
| Female | 85 (72.7) | 45 (75.0) | 40 (70.2) | .56 | 98 (86.7) | 49 (84.5) | 49 (89.1) | .47 |
| Male | 32 (27.3) | 15 (25.0) | 17 (29.8) | 15 (13.3) | 9 (15.5) | 6 (10.9) | ||
Teachers could report teaching more than one grade level.
NA=Not asked.
Principal and teacher completion rates on the posttest survey were 91% and 82% respectively. Schools, principals, and teachers who completed follow-up, versus those who did not, were compared using baseline data. Schools with no follow-up data completed by principals or teachers were closer to CGU (60.8 miles; N = 4 schools) than schools with follow-up data (136.1 miles; N =114 schools). Principals not completing follow-up were more likely to indicate that the school communicated the district’s sun safety measures with parents and requested parent provision of sun safety resources (Yes 30.0% not completing follow-up; Yes 5.7% completing follow-up).
Effect of Sun Safe Schools Intervention on Policy Awareness and Knowledge
Principals, and principals and teachers combined, demonstrated greater knowledge of the district sun safety policy in intervention than control schools when measured as the number (Principals (P): Mt=1.0, Mc=0.6; Principals & Teachers combined (P+T): Mt=1.0, Mc=0.6) and proportion of policy elements (P: Mt=30.4%, Mc=15.8%; P+T: Mt=20.4%, Mc=9.5%) known (Table 2). In the completer analysis, principals and principals and teachers combined reported more awareness of school district policy in the intervention group relative to control (P: Mt=40.9%, Mc=17.5%; P+T: Mt=23.1%, Mc=9.7%).
Table 2:
Comparison of implementation measures (in mean percentages ± sd) at posttest by experimental condition (completers analysis)
| Implementation Measure |
Control | Intervention | Test Statistic (t) |
p | Cohen’s d |
Control | Intervention | Test Statistic (F) |
p | Cohen’s d |
|---|---|---|---|---|---|---|---|---|---|---|
| Multiple Imputation School Principals (n=118 in each imputation (n=60 Control and n=58 Intervention), 5 imputations)1 |
Completer Analysis School Principals (n=108 (n=55 Control and n=53 Intervention))1 |
|||||||||
| Any element of district’s sun safety policy known | 22.8% ± 8.0% | 42.5% ± 9.8% | −1.94 | .05 | .38 | 17.5% ± 7.3% | 40.9% ± 10.5% | 5.33 | .02 | .46 |
| Number of district’s sun safety policy elements accurately known | 0.6 ± 0.2 | 1.0 ± 0.3 | −2.46 | .02 | .33 | 0.4 ± 0.2 | 0.8 ± 0.3 | 7.97 | <.01 | .35 |
| Percent of district’s sun safety policy elements accurately known | 15.8% ± 5.7% | 30.4% ± 5.5% | −2.13 | .03 | .41 | 14.2% ± 5.7% | 30.8% ± 5.7% | 5.32 | .03 | .46 |
| Implemented any sun safety practices | 89.0% ± 5.9% | 87.6% ± 5.9% | 0.22 | .83 | −.02 | 90.6% ± 5.4% | 88.2% ± 5.8% | 0.13 | .72 | −.01 |
| Number of sun safety practices implemented | 2.7 ± 0.4 | 4.2 ± 0.6 | −3.09 | <.01 | .52 | 2.7 ± 0.4 | 4.2 ± 0.5 | 14.25 | <.001 | .57 |
| Implemented any of the district’s sun safety policy elements | 74.2% ± 9.0% | 78.1% ± 7.8% | −0.40 | .69 | .08 | 74.4% ± 8.9% | 79.5% ± 7.7% | 0.26 | .61 | .10 |
| Number of district’s sun safety policy elements implemented | 2.1 ± 0.4 | 2.9 ± 0.5 | −1.99 | .05 | .32 | 2.1 ± 0.3 | 2.8 ± 0.4 | 5.46 | .02 | .35 |
| Percent of district’s sun safety policy elements implemented | 33.6% ± 6.1% | 44.2% ± 5.6% | −1.64 | .11 | .32 | 32.9% ± 5.3% | 44.1% ± 5.3% | 3.80 | .06 | .35 |
| Implemented any sun safety practices not specified in district’s policy | 30.7% ± 8.9% | 55.9% ± 10.0% | −2.28 | .02 | .42 | 31.5% ± 9.0% | 56.0% ± 10.2% | 4.69 | .03 | .41 |
| Number of sun safety practices implemented that are not specified in district’s policy | 0.4 ± 0.1 | 0.9 ± 0.2 | −2.98 | <.01 | .59 | 0.4 ± 0.1 | 0.9 ± 0.2 | 10.27 | <.01 | .62 |
| Percent of sun safety practices implemented that are not specified in district’s policy | 9.6% ± 3.3% | 23.1% ± 3.1% | −3.40 | <.001 | .68 | 9.2% ± 3.1% | 23.5% ± 3.1% | 14.33 | <.001 | .74 |
| Multiple Imputation Teachers (n=114 in each imputation (n=58 Control and n=56 Intervention), 5 imputations)1 |
Completer Analysis Teachers (n=97 (n=48 Control and n=49 Intervention))1 |
|||||||||
| Any element of district’s sun safety policy known | 17.3% ± 13.4% | 22.3% ± 15.4% | −0.59 | .56 | .13 | 5.8% ± 5.2% | 12.7% ± 9.8% | 2.28 | .14 | .30 |
| Number of district’s sun safety policy elements accurately known | 0.4 ± 0.2 | 0.6 ± 0.3 | −1.69 | .11 | .24 | 0.1 ± 0.1 | 0.3 ± 0.2 | 11.03 | <.01 | .34 |
| Percent of district’s sun safety policy elements accurately known | 9.2% ± 7.9% | 16.6% ± 7.6% | −1.17 | .24 | .24 | 4.5% ± 8.8% | 15.3% ± 8.3% | 2.75 | .10 | .34 |
| Implemented any sun safety practices | 59.8% ± 25.3% | 76.0% ± 17.8% | −1.67 | .10 | .33 | 70.1% ± 42.1% | 85.8% ± 24.5% | 3.39 | .07 | .38 |
| Number of sun safety practices implemented | 1.2 ± 0.5 | 1.8 ± 0.5 | −2.15 | <.05 | .37 | 1.1 ± 0.3 | 1.9 ± 0.5 | 12.20 | <.001 | .52 |
| Implemented any of the district’s sun safety policy elements | 26.0% ± 21.9% | 47.7% ± 27.9% | −2.19 | .03 | .45 | 23.1% ± 22.4% | 48.2% ± 31.2% | 5.96 | .02 | .52 |
| Number of district’s sun safety policy elements implemented | 0.6 ± 0.3 | 0.9 ± 0.4 | −2.12 | .04 | .32 | 0.5 ± 0.3 | 0.9 ± 0.5 | 10.12 | <.01 | .43 |
| Percent of district’s sun safety policy elements implemented | 15.3% ± 7.9% | 24.8% ± 6.9% | −1.62 | .11 | .34 | 11.1% ± 7.3% | 23.3% ± 6.8% | 4.74 | .03 | .46 |
| Implemented any sun safety practices not specified in district’s policy | 23.5% ± 17.2% | 43.8% ± 20.5% | −1.92 | .06 | .39 | 21.1% ± 15.7% | 48.2% ± 22.0% | 5.81 | .02 | .49 |
| Number of sun safety practices implemented that are not specified in district’s policy | 0.3 ± 0.1 | 0.5 ± 0.2 | −1.34 | .19 | .26 | 0.2 ± 0.1 | 0.4 ± 0.2 | 3.91 | .05 | .36 |
| Percent of sun safety practices implemented that are not specified in district’s policy | 5.0% ± 4.5% | 11.8% ± 4.1% | −2.15 | .03 | .40 | 2.5% ± 4.3% | 11.2% ± 4.1% | 7.83 | <.01 | .55 |
| Multiple Imputation Principals & Teachers Combined (n=232 in each imputation (n=118 Control and n=114 Intervention), 5 imputations)1 |
Completer Analysis Principals & Teachers Combined (n=205 (n=103 Control and n=102 Intervention))1 |
|||||||||
| Any element of district’s sun safety policy known | 19.1% ± 11.3% | 29.8% ± 15.1% | −1.67 | .10 | .28 | 9.7% ± 7.2% | 23.1% ± 14.0% | 7.27 | <.01 | .41 |
| Number of district’s sun safety policy elements accurately known | 0.6 ± 0.3 | 1.0 ± 0.5 | −2.86 | .01 | .34 | 0.2 ± 0.1 | 0.7 ± 0.4 | 11.77 | <.001 | .41 |
| Percent of district’s sun safety policy elements accurately known | 9.5% ± 7.5% | 20.4% ± 7.4% | −2.23 | .03 | .33 | 6.6% ± 8.0% | 20.8% ± 7.8% | 7.46 | <.01 | .42 |
| Implemented any sun safety practices | 73.1% ± 14.4% | 80.9% ± 10.8% | −1.33 | .18 | .19 | 78.4% ± 13.5% | 85.8% ± 9.6% | 1.89 | .17 | .21 |
| Number of sun safety practices implemented | 1.9 ± 0.4 | 2.8 ± 0.6 | −3.88 | <.01 | .45 | 1.7 ± 0.4 | 2.6 ± 0.6 | 11.41 | <.01 | .55 |
| Implemented any of the district’s sun safety policy elements | 38.2% ± 16.9% | 52.0% ± 17.7% | −1.80 | .07 | .26 | 37.8% ± 18.8% | 53.8% ± 19.7% | 3.27 | .07 | .29 |
| Number of district’s sun safety policy elements implemented | 1.0 ± 0.4 | 1.4 ± 0.6 | −3.25 | <.01 | .35 | 0.9 ± 0.4 | 1.3 ± 0.6 | 7.75 | <.01 | .41 |
| Percent of district’s sun safety policy elements implemented | 19.8% ± 7.0% | 30.2% ± 6.7% | −2.16 | .03 | .33 | 17.7% ± 7.2% | 30.0% ± 7.0% | 7.50 | <.01 | .40 |
| Implemented any sun safety practices not specified in district’s policy | 30.4% ± 16.8% | 55.1% ± 18.2% | −3.11 | <.01 | .41 | 39.5% ± 19.9% | 64.6% ± 18.6% | 6.45 | .01 | .43 |
| Number of sun safety practices implemented that are not specified in district’s policy | 0.4 ± 0.1 | 0.7 ± 0.2 | −2.96 | <.01 | .42 | 0.4 ± 0.1 | 0.8 ± 0.2 | 9.66 | <.01 | .47 |
| Percent of sun safety practices implemented that are not specified in district’s policy | 6.1% ± 4.2% | 16.2% ± 4.0% | −3.68 | <.001 | .54 | 5.4% ± 4.2% | 16.8% ± 4.1% | 15.64 | <.001 | .63 |
Comparison was adjusted for whether respondent was assessed at pretest and posttest or was new in position at posttest, respondent phenotypic index, total number of students enrolled in school, percent of students in Free or Reduced Price Meals program, and percent of White not Hispanic students at school. Sample size is slightly reduced when covariates are added in the models.
Effect of Sun Safe Schools Intervention on Policy Implementation
In the intervention schools versus controls using the multiple imputation analysis, principals reported more total practices implemented (Mt=4.2, Mc=2.7), more district sun safety practices implemented (Mt=2.9, Mc=2.1), and a greater number (Mt=0.9, Mc=0.4) and percentage (Mt=23.1%, Mc=9.6%) of practices not in the district’s policy implemented. This pattern of significant effects was replicated for principals in the completer analysis.
In the intervention schools versus controls using the multiple imputation analysis, teachers reported more practices in total implemented (Mt=1.8, Mc=1.2) and a higher percentage of at least one sun safety practice being implemented from the district’s policy (Mt=47.7%, Mc=26.0%). In addition, teachers reported that a higher number of sun safety practices were implemented in the district’s policy in the intervention schools versus controls (Mt=0.9, Mc=0.6) and a higher percentage of practices implemented not in the district’s policy (Mt=11.8%, Mc=5.0%). In the completer analysis alone, teachers also reported a higher proportion of practices implemented in the district’s policy (Mt=23.3%, Mc=11.1%), a greater percentage of at least one sun safety practice implemented not in the district’s policy (Mt=48.2%, Mc=21.1%), and a greater number implemented from outside the district’s policy (Mt=0.4, Mc=0.2). Differences observed for principals were replicated in the combined sample of principals and teachers except for the number of district sun safety policy elements implemented and the percentage of district sun safety policy elements implemented, which was statistically significant in the combined sample in both completer and multiple imputation runs (Table 2).
Effect of Sun Safe Schools Intervention on Specific Policy Practices
Table 3 reports the implementation produced by SSS for specific sun safety practices. The pattern of statistical significance at the p<.05 level was identical between the multiple implementation and completer analyses. According to principals, sun safety was incorporated in class instruction for students (Mt-61.2%, Mc=8.7%) and in-service training for staff (Mt=40.9%, Mc=14.8%). Sun safety information was included in communications to parents, along with requests that parents provide items to protect children from the sun (e.g., sunscreen) (Mt=63.4%, Mc=29.3%). More principals in intervention schools reported that students were wearing sun protective clothing at school than in control schools (Mt=45.5%, Mc=22.7%). More teachers in intervention schools reported in-service training on sun safety (Mt=15.2%, Mc=3.8%) and that school staff were encouraged to model sun safety practices than at control schools (Mt=59.6%, Mc=33.9%). The combined sample of principals and teachers replicated the effects for principals and showed that the intervention schools were more likely to provide sun safety items such as sunscreen for students (Mt=6.1 %, Mc=1.6%) and encourage staff to model sun safety practices (Mt=61.3%, Mc=44.0%) than control schools (Table 3).
Table 3:
Comparison of policy elements implemented (in mean percentages ± sd) at posttest by experimental condition (completer analysis)
| Implementation Measure |
Control | Intervention | Test Statistic (t) |
p | Cohen’s d |
Control | Intervention | Test Statistic (F) |
p | Cohen’s d |
|---|---|---|---|---|---|---|---|---|---|---|
| Multiple Imputation School Principals (n=118 in each imputation (n=60 Control and n=58 Intervention), 5 imputations)1 |
Completer Analysis School Principals (n=108 (n=55 Control and n=53 Intervention))1 |
|||||||||
| UV index is monitored and outdoor activities are modified with regard to risk of harm associated with the index level | 36.9% ± 9.4% | 40.4% ± 9.1% | −0.33 | .74 | .06 | 36.2% ± 9.2% | 38.8% ± 8.9% | 0.06 | .81 | .04 |
| Students wear sun-protective clothing (shirts with sleeves; long pants) and sunglasses | 22.7% ± 7.7% | 45.5% ± 8.5% | −2.11 | .04 | .46 | 22.2% ± 7.0% | 46.4% ± 8.8% | 5.40 | .02 | .48 |
| Students wear hats when outdoors. | 48.7% ± 10.4% | 61.3% ± 9.5% | −1.07 | .29 | .21 | 47.2% ± 9.9% | 61.6% ± 9.4% | 1.46 | .23 | .23 |
| Students wear sunscreen when outdoors. | 24.3% ± 9.2% | 33.1% ± 10.4% | −0.79 | .43 | .14 | 24.7% ± 9.2% | 33.3% ± 10.7% | 0.59 | .44 | .13 |
| Construction of new schools or renovation of existing schools or school grounds include plans that increase shade on school grounds. | 39.3% ± 9.4% | 29.3% ± 8.4% | 0.95 | .34 | −.16 | 39.6% ± 9.4% | 28.8% ± 8.3% | 1.02 | .32 | −.18 |
| Educational objectives related to the health risks of sun exposure are taught in the classroom. | 8.7% ± 5.5% | 61.2% ± 12.7% | −3.88 | <.001 | .89 | 7.2% ± 4.6% | 63.3% ± 12.9% | 19.32 | <.0001 | .97 |
| In-service staff training and continuing education regarding sun safety is delivered to teachers. | 14.8% ± 6.2% | 40.9% ± 9.7% | −2.54 | .01 | .54 | 14.0% ± 5.8% | 40.5% ± 9.5% | 7.63 | <.01 | .56 |
| School communicates the district’s sun safety measures with parents and requests parent provision of sun safety resources, such as sunscreen and sun protective clothing for their child. | 29.3% ± 9.7% | 63.4% ± 10.3% | −2.79 | <.01 | .56 | 27.6% ± 9.6% | 66.0% ± 10.7% | 9.86 | <.01 | .62 |
| School provides sun protection resources, such as sunscreen or UV-protective clothing. | 1.5% ± 1.9% | 9.4% ± 5.6% | −1.63 | .10 | .32 | 1.3% ± 1.5% | 10.5% ± 6.5% | 3.43 | .07 | .37 |
| Staff is encouraged to model recommended sun-safe behaviors. | 41.4% ± 10.3% | 55.7% ± 9.7% | −1.17 | .25 | .25 | 39.3% ± 9.9% | 57.8% ± 10.0% | 2.52 | .12 | .30 |
| Multiple Imputation Teachers (n=114 in each imputation (n=58 Control and n=56 Intervention), 5 imputations)1 |
Completer Analysis Teachers (n=97 (n=48 Control and n=49 Intervention))1 |
|||||||||
| UV index is monitored and outdoor activities are modified with regard to risk of harm associated with the index level | 17.3% ± 13.1% | 24.8% ± 14.9% | −0.92 | .36 | .19 | 13.8% ± 11.0% | 23.9% ± 15.9% | 1.75 | .19 | .25 |
| Students wear sun-protective clothing (shirts with sleeves; long pants) and sunglasses | 11.4% ± 8.7% | 17.1% ± 11.1% | −0.89 | .38 | .18 | 8.4% ± 7.1% | 15.0% ± 11.1% | 1.33 | .26 | .23 |
| Students wear hats when outdoors. | 17.7% ± 15.8% | 23.4% ± 18.7% | −0.78 | .44 | .15 | 14.7% ± 16.0% | 21.4% ± 21.0% | 0.89 | .35 | .19 |
| Students wear sunscreen when outdoors. | 10.3% ± 7.8% | 7.4% ± 5.7% | 0.68 | .50 | −.10 | 8.8% ± 7.4% | 6.2% ± 5.2% | 0.42 | .52 | −.11 |
| Construction of new schools or renovation of existing schools or school grounds include plans that increase shade on school grounds. | 10.3% ± 7.6% | 16.4% ± 10.6% | −1.07 | .28 | .20 | 9.0% ± 7.5% | 15.6% ± 11.4% | 1.39 | .24 | .23 |
| Educational objectives related to the health risks of sun exposure are taught in the classroom. | 9.5% ± 10.2% | 12.2% ± 12.3% | −0.47 | .64 | .09 | 7.9% ± 9.7% | 10.8% ± 12.4% | 0.28 | .60 | .11 |
| In-service staff training and continuing education regarding sun safety is delivered to teachers. | 3.8% ± 4.8% | 15.2% ± 13.7% | −2.27 | .02 | .53 | 1.6% ± 2.4% | 11.9% ± 13.8% | 6.26 | .02 | .69 |
| School communicates the district’s sun safety measures with parents and requests parent provision of sun safety resources, such as sunscreen and sun protective clothing for their child. | 11.9% ± 11.4% | 20.7% ± 15.0% | −1.21 | .23 | .26 | 9.5% ± 7.7% | 19.7% ± 13.3% | 2.28 | .14 | .32 |
| School provides sun protection resources, such as sunscreen or UV-protective clothing. | 2.8% ± 3.8% | 7.7% ± 8.5% | −1.20 | .23 | .36 | 0.1% ± 0.4% | 2.7% ± 8.9% | 2.26 | .14 | .43 |
| Staff is encouraged to model recommended sun-safe behaviors. | 33.9% ± 22.0% | 59.6% ± 19.7% | −2.08 | .04 | .47 | 33.1% ± 19.6% | 67.6% ± 18.4% | 7.24 | <.01 | .60 |
| Multiple Imputation Principals & Teachers Combined (n=232 in each imputation (n=118 Control and n=114 Intervention), 5 imputations)1 |
Completer Analysis Principals & Teachers Combined (n=205 (n=103 Control and n=102 Intervention))1 |
|||||||||
| UV index is monitored and outdoor activities are modified with regard to risk of harm associated with the index level | 21.4% ± 12.6% | 27.1% ± 14.1% | −.92 | .36 | .14 | 20.9% ± 13.2% | 27.7% ± 15.7% | 1.20 | .28 | .16 |
| Students wear sun-protective clothing (shirts with sleeves; long pants) and sunglasses | 13.2% ± 10.5% | 24.8% ± 16.3% | −2.08 | .04 | .33 | 11.5% ± 8.2% | 24.1% ± 14.3% | 6.01 | .02 | .38 |
| Students wear hats when outdoors. | 24.5% ± 14.1% | 32.1% ± 16.5% | −1.29 | .20 | .18 | 22.9% ± 13.8% | 31.8% ± 16.7% | 2.12 | .15 | .22 |
| Students wear sunscreen when outdoors. | 13.3% ± 8.4% | 14.7% ± 9.1% | −.33 | .74 | .03 | 11.7% ± 8.4% | 12.6% ± 8.7% | 0.05 | .83 | .03 |
| Construction of new schools or renovation of existing schools or school grounds include plans that increase shade on school grounds. | 17.4% ± 10.2% | 18.6% ± 10.6% | −.25 | .80 | .03 | 16.7% ± 11.2% | 17.6% ± 11.4% | 0.03 | .87 | .02 |
| Educational objectives related to the health risks of sun exposure are taught in the classroom. | 8.7% ± 6.4% | 24.8% ± 14.2% | −3.26 | <.01 | .50 | 7.6% ± 8.8% | 24.2% ± 22.4% | 12.50 | <.001 | .55 |
| In-service staff training and continuing education regarding sun safety is delivered to teachers. | 6.4% ± 4.7% | 20.4% ± 11.6% | −3.23 | <.01 | .52 | 4.6% ± 3.7% | 18.2% ± 11.9% | 13.73 | <.001 | .58 |
| School communicates the district’s sun safety measures with parents and requests parent provision of sun safety resources, such as sunscreen and sun protective clothing for their child. | 13.6% ± 8.8% | 30.2% ± 15.0% | −2.85 | <.01 | .44 | 11.2% ± 12.3% | 28.3% ± 24.7% | 11.11 | <.01 | .51 |
| School provides sun protection resources, such as sunscreen or UV-protective clothing. | 1.6% ± 2.4% | 6.1% ± 8.7% | −2.01 | <.05 | .37 | 0.7% ± 1.0% | 5.3% ± 6.3% | 10.33 | <.01 | .42 |
| Staff is encouraged to model recommended sun-safe behaviors. | 44.0% ± 20.1% | 61.3% ± 17.5% | −2.19 | .03 | .34 | 47.1% ± 19.9% | 68.3% ± 16.9% | 6.65 | .01 | .41 |
Comparison was adjusted for whether respondent was assessed at pretest and posttest or was new in position at posttest, respondent phenotypic index, total number of students enrolled in school, percent of students in Free or Reduced Price Meals program, and percent of White not Hispanic students at school. Sample size is slightly reduced when covariates are added in the models.
Discussion
The SSS intervention was effective at increasing implementation of sun safety practices in public elementary schools producing effects for a) the total number of practices implemented; b) practices reflecting elements of district policy and; c) practices enacting elements not included in district policy. These changes suggested that the SSS intervention can increase implementation of policy at elementary schools and impact sun safety practices without being restricted to the school district’s adopted policy. It is possible that the presence of district policy indicated to personnel that the district prioritized sun safety, increasing willingness to implement practices, regardless of whether they were explicitly written in the policy.
The practices implemented with greater frequency due to intervention included the education of students on sun safety, students’ use of protective clothing, communication to parents, in-service training of staff, and encouragement of staff to model sun safety. The Community Guide has identified education of children in primary grades as an effective intervention for skin cancer prevention and SSS may facilitate action on this recommendation.31 Likewise, when teachers are convinced to engage in sun safety themselves, the modeling provided may motivate children to practice sun safety.32 Practice implementation for protective clothing use by children is important because clothing can provide superior protection to sunscreen.33
The increase in communication with parents may be particularly important. By conveying the importance of sun protection to parents, parents may be motivated to improve sun protection behavior within the family. Sun safety relies on items such as sunscreen, hats, protective clothing, and sunglasses which are principally the responsibility of parents to provide. School uniforms may also be a source of sun protection for students, but not in the present sample which included only public schools that did not require a school uniform.
While reports by the principals and teachers were largely consistent, differences did emerge. Principals were more knowledgeable of the policy than teachers perhaps due to principals’ responsibility for implementation of district policy at the school and potentially greater access to school district communications regarding policy. Both informants reported increased practices at intervention compared to control schools but with principals reporting more change in the practices not specified in district policy when examined in the multiple imputation analysis. When individual sun safety practices were examined, principals reported more practice implementation than teachers, with medium to large effect sizes for principals and small to medium for teachers.34 Principals may be better positioned to know what is happening throughout a school. Also, principals were aware of and more knowledgeable about the district’s sun safety policy at baseline than teachers,15 which may have made principals more attentive to sun safety practices. Finally, principals received a higher intervention dose and would be more likely to instigate and observe change in sun safety practices than teachers.
The present study had several strengths. A large sample of schools were recruited in multiple districts across a wide geographic area within California. Generalizability might be further enhanced in future studies by expanding the geographic range and perhaps recruiting across multiple states that have adopted state-level mandates for student sun safety. A randomized controlled trial design was utilized, with equivalent experimental conditions at baseline, increasing the ability to draw causal inferences about program effectiveness. The content of district board policies was confirmed using a validated protocol.35 High uptake rates of the SSS intervention were produced and high completion rates were obtained at posttest. The use of an intervention that can be delivered entirely through electronic and mail contacts enhances dissemination potential.
Limitations were also evident. Generalizability is limited to a) a single state with high levels of UV and with a state law requiring schools to allow sunscreen and hat use and b) to elementary schools. Future studies might examine approaches to school level policy implementation in the absence of state law and supportive district policy. The primary outcome was assessed using self-report measures which are open to detection bias. The study was run in a region where a professional association promoted the school policy, potentially increasing the policy’s legitimacy. The implementation measure was a post-only assessment; we performed multiple imputation to account for the small attrition, which produced only small reductions in effect sizes.
Conclusion
Policy initiatives are an important part of school-based efforts to improve skin cancer prevention. To our knowledge, this is one of the first studies to evaluate a program designed to promote school-level implementation of district sun safety policy. The findings suggest that a technical assistance program delivered to principals and teachers can increase sun safety practices related to district-board policy and even generalize to practices that promote sun protection but that are not included in the district-board policy. Our findings further suggest that effective initiatives to enlist schools in community-based skin cancer prevention efforts need to combine the adoption of district sun safety policies with communication to school personnel and resources supporting implementation of sun safety practices. Such implementation is essential if policy is to improve skin cancer prevention for school children and fulfill the Surgeon General’s call to action.11
Supplementary Material
Highlights.
Intervention with elementary school personnel increases sun safety practice use.
Practices in, and not in board policy, were both implemented in schools.
Policy combined with communication to school personnel is most effective approach.
Acknowledgments
Funding: This research was funded by a grant from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (HD074416).
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Cancer Facts & Figures 2020. Atlanta, GA: American Cancer Society;2020. [Google Scholar]
- 2.Armstrong BK, English DR. Cutaneous malignant melanoma In: Schottenfeld D, Franmeni JF, eds. Cancer epidemiology and prevention (2nd ed.). New York: Oxford University Press; 1996:1281–1312. [Google Scholar]
- 3.Marks R, Jolley D, Lectsas S, Foley P. The role of childhood exposure to sunlight in the development of solar keratoses and non-melanocytic skin cancer. Med J Aust. 1990;152:62–66. [DOI] [PubMed] [Google Scholar]
- 4.Godar DE, Urbach F, Gasparro FP, van der Leun JC. UV doses of young adults. Photochem Photobiol. 2003;77(4):453–457. [DOI] [PubMed] [Google Scholar]
- 5.Green AC, Wallingford SC, McBride P. Childhood exposure to ultraviolet radiation and harmful skin effects: epidemiological evidence. Prog Biophys Mol Biol. 2011;107(3):349–355. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Buller DB, Callister MA, Reichert T. Parents' sources of information on skin cancer prevention for their children. Oncol Nurs Forum. 1995;10(22):1559–1566. [PubMed] [Google Scholar]
- 7.Buller DB, Borland R. Skin cancer prevention for children: A critical review. Health Educ Behav. 1999;26:317–343. [DOI] [PubMed] [Google Scholar]
- 8.Eaton DK, Kann L, Kinchen S, Shanklin S, Ross J, Hawkins J. Youth risk behavior surveillance--United States, 2007. MMWR Surveill Summ. 2008;57:1–131. [PubMed] [Google Scholar]
- 9.Cokkinides V, Weinstock M, Glanz K, Albano J, Ward E, Thun M. Trends in sunburns, sun protection practices, and attitudes toward sun exposure protection and tanning among US adolescents, 1998-2004. Pediatrics. 2006;118(3):853–864. [DOI] [PubMed] [Google Scholar]
- 10.Buller D, Cokkinides V, Hall H, Hartman A, Saraiya M, Miller E et al. Prevalence of sunburn, sun protection, and indoor tanning behaviors among Americans: Systematic review from national surveys. Journal of the American Academy of Dermatology. 2011;65:114–123. [DOI] [PubMed] [Google Scholar]
- 11.U.S. Department of Health and Human Services, Office of the Surgeon General. The Surgeon General's Call to Action to Prevent Skin Cancer. U.S. Department of Health and Human Services;Office of the Surgeon General;2014. [PubMed] [Google Scholar]
- 12.Glanz K, Saraiya M, Wechsler H. Guidelines for school programs to prevent skin cancer. Morbidity and Mortality Weekly Report: Recommendations and Reports. 2002;51(RR04):1–18. [PubMed] [Google Scholar]
- 13.Guy GP Jr, Holman DM, Watson M. The important role of schools in the prevention of skin cancer. JAMA dermatology. 2016;152(10):1083–1084. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Buller DB, Reynolds KD, Ashley JL, et al. Motivating public school districts to adopt sun protection policies: a randomized controlled trial. Am J Prev Med. 2011;41(3):309–316. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Buller DB, Reynolds KD, Berteletti J, et al. Accuracy of principal and teacher knowledge of school district policies on sun protection in California elementary schools. Prev Chronic Dis. 2018;15:E07. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Reynolds K, Buller D, Berteletti J, et al. School-Level Factors Associated with Sun Protection Practices in California Elementary Schools. Journal of School Health. In press. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Everett Jones S, Guy GP Jr. Sun Safety Practices Among Schools in the United States. JAMA Dermatol. 2017;153(5):391–397. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Reeder AI, McNoe BM, Iosua EE. Sun protection practices in New Zealand secondary schools: a 2014 baseline study. Preventive medicine reports. 2016;3:257–263. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Reeder AI, Jopson JA, Gray A. Primary school sun protection policies and practices 4 years after baseline--a follow-up study. Health Educ Res. 2012;27(5):844–856. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Rezai L, Thorgaard C, Philip A. Influential factors for sun policy implementation in Danish kindergartens. Scand J Public Health. 2011;39(5):479–483. [DOI] [PubMed] [Google Scholar]
- 21.Dudley DA, Winslade MJ, Wright BJ, Cotton WG, McIver JL, Jackson KS. Rationale and study protocol to evaluate the SunSmart policy intervention: a cluster randomised controlled trial of a primary school-based health promotion program. Bmc Public Health. 2015;15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Wright B, Winslade M, Dudley D, Cotton W, Hamer A. Protect your skin and let the fun begin: The results of an intervention to improve NSW primary schools' implementation of the SunSmart Program. Health Promotion Journal of Australia. 2019;30(2):267–271. [DOI] [PubMed] [Google Scholar]
- 23.Turner D, Harrison SL, Buettner P, Nowak M. School sun-protection policies-does being SunSmart make a difference? Health Education Research. 2014;29(3):367–377. [DOI] [PubMed] [Google Scholar]
- 24.SAS(R) 9.2 Language Reference: Dictionary. Fourth Edition:https://support.sas.com/documentation/cdl/en/lrdict/64316/HTML/default/viewer.htm#a000202926.htm. Accessed February 18, 2020, 2020.
- 25.Fishman GS, Moore LR. A Statistical Evaluation of Multiplicative Congruential Random Number Generators with Modulus 2(31)-1. Journal of the American Statistical Association. 1982;77(377):129–136. [Google Scholar]
- 26.Rogers EM. Diffusion of innovations. 5th ed. New York: Free Press; 2003. [Google Scholar]
- 27.Havelock R The Change Agent's Guide to Innovation in Education. Englewood Cliffs: Education Technology; 1973. [Google Scholar]
- 28.Berteletti J, Buller DB, Massie K, Ashley J, Liu X, Reynolds KD. Sun protection policies in public school districts with elementary schools in California. JAMA dermatology. 2018;154(1):103–105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Shadish W, Cook T, Campbell D. Experimental and Quasi-Experimental Designs for Generalized Causal Inference. Belmont, CA: Wadsworth Cengage Learning; 2002. [Google Scholar]
- 30.Yuan Y Multiple Imputation using SAS Software. Journal of Statistical Software. 2011;45(6):1–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.The Guide to Community Preventive Services: The Community Guide. http://www.thecommunityguide.org/index.html. Accessed November 4, 2018.
- 32.Bandura A Social Foundations of Thought and Action: A Social Cognitive Theory. New Jersey: Prentice Hall, Inc.; 1986. [Google Scholar]
- 33.Buller DB, Cokkinides V, Hall HI, et al. Prevalence of sunburn, sun protection, and indoor tanning behaviors among Americans: review from national surveys and case studies of 3 states. J Am Acad Dermatol. 2011;65(5 Suppl 1):S114–123. [DOI] [PubMed] [Google Scholar]
- 34.Cohen J Statistical power analysis for the behavioral sciences. Hillsdale, New Jersey: Lawrence Erlbaum Associates; 1988. [Google Scholar]
- 35.Reynolds KD, Buller DB, French SA, Buller MK, Ashley JL. School sun-protection policies: measure development and assessments in 2 regions of the United States. J Sch Health. 2012;82(11):499–507. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.