The Sunscreen for Kindergarteners (SKIN) Study Trial Protocol

Abstract

Background:

Exposure to ultraviolet radiation (UVR) is the major modifiable risk factor for skin cancers. The majority of lifetime UVR exposure occurs before age 20, underscoring an important window for risk reduction. Incorporation of skills-based sunscreen education into school health curricula may foster the development of consistent and effective use of sunscreen among children and youth. We describe the study protocol for a first-of-its-kind study that examined the feasibility of bringing skills-based sunscreen education into kindergarten classrooms.

Methods:

Participants were 96 kindergarten students across four classrooms in a single elementary school. A single-blind open-label trial design was used to evaluate the feasibility of incorporating a song-based, video-guided intervention for independent application of sunscreen into the kindergarten curriculum. Students first completed a 10-day no-intervention baseline period, followed by a 10-day intervention period, and then a 10-day randomized follow-up period where students were randomly assigned to continue with the intervention or to revert to the no-intervention condition.

Outcomes:

Feasibility metrics associated with study process, resources, management, scientific outcomes and safety were gathered. The primary outcome was pre-to-post intervention changes in student engagement in the sunscreen task. The secondary outcome was pre-to-post intervention changes in the proportion of exposed skin to which a student applies sunscreen. Teacher and student perceptions of intervention value and utility were also evaluated.

Discussion:

This is the study protocol for a clinical trial designed to determine the feasibility of implementing a skills-based sunscreen curriculum in kindergarten classrooms. Next steps include evaluation of the intervention for efficacy and effectiveness.

Introduction

Skin cancer remains the most common form of cancer in the United States (U.S.).^1–3 One in five Americans will develop skin cancer by age 70,⁴ and approximately 7,000–9,000 die from melanoma each year.^5,6 Exposure to ultraviolet radiation (UVR) from the sun is the major modifiable risk factor for skin cancers, including melanoma.^7–10 When used consistently and effectively, sunscreen and sun protective clothing can significantly reduce deleterious effects of UVR exposure, prevent sunburns, and lower lifetime skin-cancer risk.^11–15 However, consistent engagement in sun safe behaviors and the effective use of sunscreen are uncommon.^16,17 Efforts to develop brief, effective interventions that increase proper use of sunscreen are urgently needed.

Children are important targets for such interventions. Research suggests that the majority of lifetime UVR exposure occurs before age 20.¹⁸ Moreover, results of multiple meta-analyses suggest that the presence of just one blistering sunburn in childhood can nearly double lifetime risk of melanoma.^19–21 Furthermore, regular sunscreen use in the first 18 years of life has been estimated to reduce the lifetime incidence of non-melanoma skin cancers by 78%.^22,23 Educational interventions that foster habit formation regarding sun safe behaviors are critical to decreasing sunburn risk and reducing the lifetime risk of skin cancer.

Targeting skills-based sunscreen intervention and training opportunities in the classroom setting is particularly important. It is common for elementary students who live in mild climate states in the U.S. to spend several hours of school time outdoors each week, many of which are during peak UVR exposure hours.²⁴ The past decade has seen a sharp uptick in adaptations to the school environment (e.g., the addition of sunshade structures to school playgrounds) designed to reduce students’ daily UVR exposure.^25,26 This, coupled with the proliferation of knowledge-based sun-safety curricula for elementary education health programming, represents an important initial effort to reduce UVR exposure during the school day.^27–29

Access to sunscreen during the school day, however, remains severely curtailed. A minority of all U.S. schools allow time for students to apply sunscreen during the school day.³⁰ Only 13% of U.S. schools have sunscreen available for student use,³⁰ and, to our knowledge, there are no published curricula that include comprehensive skills-based training for systematic, independent application of sunscreen. The “Sunbeatables™” is an existing educational program designed to improve sun safety behaviors among early elementary students.³¹ However, this program does not specifically emphasize practice in independent application of sunscreen. Efforts to develop and evaluate competency-based sun safety curricula that emphasize skills for independent, effective application of sunscreen among youth are needed. Such interventions would nicely complement other trends in competency-based health education (e.g., hand washing, dental hygiene) targeting youth.

As a first step, we have developed a song-based, video-guided intervention designed to foster skills needed for systematic and effective self-application of sunscreen among kindergarten students. This paper describes the study protocol associated with a small, single-blind, open-label feasibility trial designed to examine the feasibility of incorporating a 2.5-minute song-based, video-guided sunscreen application intervention into daily use in four kindergarten classrooms.

Materials and Methods

Study Overview

This study assessed the feasibility of conducting a randomized controlled trial to evaluate the efficacy of a song-based, video-guided intervention designed to teach kindergarteners to independently apply (i.e., students self-apply) sunscreen as a short, in-classroom activity. Feasibility data (i.e., study process, safety, school-wide and classroom resources that would contribute to the ease of incorporating the sunscreen task, and ease of classroom management) was gathered continuously throughout the preparatory, baseline, intervention, and follow-up phases. Preliminary data regarding the intervention’s impact, including students’ ability and willingness to productively engage with the sunscreen task, pre-post intervention changes in students’ competency for sunscreen application (i.e., ability to attend to all areas of exposed skin with sunscreen) was also gathered to guide the revision of the intervention for a larger clinical trial.

Design

This was a single-blind, open- label clinical trial conforming to an A/B design with a randomized follow-up to either condition A or B (Figure 1).

Figure 1.

Study Design. All four classrooms received no intervention (Condition A/instrumental audio only) throughout baseline- -weeks 1 and 2. Next, all moved to the full intervention (Condition B/audio with lyrics and video) throughout the intervention phase - -weeks 3 and 4. Finally, all four classrooms of 24 students each were randomized by classroom to receive either no intervention (Condition A/audio only) or the full intervention (Condition B/audio with lyrics and video) on weeks 5 and 6 of the study. Four students from each class were observed weekly for secondary outcomes data.

We assessed the feasibility of implementing a song-based, video-guided intervention that was designed to teach systematic and independent application of sunscreen among four kindergarten classes (96 students) at a single school. Participating students completed two consecutive weeks (10 school days) of baseline assessment in which they participated in an unguided sunscreen application task. Here they were provided a tube of sunscreen and instructed by the classroom teacher (with use of a script) to begin the task of applying sunscreen. To ensure uniformity across classrooms, students were instructed to engage in this task during a 2.5-minute period in which the teacher played an instrumental audio recording (Condition A; no intervention/instrumental audio).

Next, all participating students entered the intervention phase (Condition B; audio with lyrics and video intervention), in which they were exposed to the full intervention, a song-based, video-guided intervention video “I Wear Sunscreen Every Day” in class, daily, for the next two consecutive weeks (10 school days). Here, the song (same tune as the instrumental audio in Condition A) featured lyrics that guided the students through the sunscreen task. Upon completion of the (Condition B) intervention period, classrooms were randomly assigned to one of two follow-up conditions: 1) return to baseline (no intervention/instrumental audio only, Condition A); or 2) continue daily use of the song-based, video-guided intervention (audio with lyrics and video intervention, Condition B) for two more consecutive weeks (10 school days) (Figure 1).

Key feasibility metrics included salient process, resource, management, scientific and safety factors. Successful intervention engagement, learning outcomes (i.e., pre-post intervention increases in competency with sunscreen application), and maintenance effects (i.e., durability of post-intervention skill changes throughout the follow-up period) were evaluated. Information on student/teacher perceptions of the acceptability of the audio with lyrics and video intervention condition was gathered, as was information about teachers’ satisfaction with the intervention, and their perceptions of its convenience and value.

The protocol for this RCT follows the Standard Protocol Items for Randomized Trials (SPIRIT) statement guidelines and is registered with the Clinical Trials website (ClinicalTrials.gov identified: NCT03752736). A completed SPIRIT statement can be found online (https://www.spirit-statement.org). The dataset for this trial is not publicly disclosed.

Setting

The study took place within four kindergarten classrooms of a public elementary school in Northern California.

Study Population

Participants were recruited from January 7, 2019 – February 15, 2019 (trial was implemented March-April, 2019) through written advertisements sent home with the students, electronic mail announcements from the teachers, and in-person recruitment efforts made onsite by the study team.

Eligibility:

Participants were drawn exclusively from the general education kindergarten population of a single elementary school. This included all eligible general education kindergarten students (N = 96) and their teachers (N = 4). Eligibility criteria for kindergarten students was as follows: a) enrollment in general education kindergarten at the selected school; b) passing of a cursory, teacher-lead screening that assessed psychological readiness, (i.e., no diagnosed pica or other disorders that would preclude the safe use of sunscreen); emotional readiness, (i.e., child was developmentally able to safely handle a sunscreen bottle); and behavioral readiness, (i.e., child was unlikely to intentionally disrupt class via misuse of sunscreen); c) confirmation by the parent(s) to have no contraindications to daily use of sunscreen, including no known allergy to sunscreen; d) agreement to participate from the child; and f) a signed parental consent.

The flow chart (CONSORT diagram) for this study is presented in Figure 2.

Figure 2.

CONSORT flow diagram. Out of 96 students in the school, 86 started and finished the randomized study.

Exclusionary criteria included failure to meet any one of the criteria delineated above. In addition, otherwise eligible students whose parents consented on their behalf were excluded if one or more of the following conditions were met: 1) student developed a skin reaction to sunscreen being used during the pre-intervention “test” period (see below) or throughout the study; 2) student refused to participate or expressed a desire to not participate despite parental consent; 3) student became disruptive such that s/he prevented others from participating in the study without incident. English fluency was not required for participation.

Procedure

The project was introduced to the parents through an announcement by the kindergarten teachers at Back-to-School night. Then, prior to the winter vacation, a reminder notification and informational advertisements were sent out by the teachers via electronic mail to the parents. The week before winter vacation, the students received printed materials in their “take-home folders.” Next, a series of “parent overview” sessions, 1-hour in length, were held each morning for a period of three weeks prior to study baseline. These overview sessions allowed parents to be fully informed about the study and participatory requirements prior to providing written consent for their child(ren). At the time of consent, parents completed a questionnaire about their child’s sunburn history and sun-protective behaviors both during the school year and in the summertime. Teachers also provided written consent to participate after their own study overview and consent session.

Based on discussion with the teachers, the study team chose a commercially available, mineral-based, tear free, sun protection factor level 50 (SPF 50) sunscreen, designed and marketed for child use. Two-ounce bottles were purchased for each study participant with funds from the Stanford Society of Physicians Scholars Grant. The study team received no support, financial or otherwise, from the manufacturer of the study sunscreen. Further, there was no correspondence between the study team and the sunscreen manufacturer. Parents could choose to send a different, preferred sunscreen product from home. If they chose this option, they were required to inform the teacher in writing and ensure that the product was neither a spray nor a roll-on formulation.

During a brief ten-minute classroom presentation one week prior to baseline, the study team discussed the project and reviewed developmentally appropriate material on sun safety skills (seeking shade, wearing sunglasses, hats, protective clothing, and sunscreen) for outside play. Students also learned how to safely use sunscreen (e.g., do not get it in eyes or mouth). They also learned a proper eye wash technique to use if sunscreen gets in their eyes. Each student with parental consent received their sunscreen bottle and was shown how to open the cap and squeeze a small amount of sunscreen into their palm. Students were instructed to apply the sunscreen to one spot on their non-dominant forearm. They brought home a study-related handout which explained to their parents that a didactic about sunscreen application occurred in the classroom that day. Parents were advised to observe and report any redness or skin irritation on the forearm of their child within 24 hours of the in-class presentation. This facilitated an opportunity to detect skin reactions to the study sunscreen prior to the onset of the trial. Students who did not have parental consent were not given sunscreen and were instead provided with a quiet task to do in the reading corner of the classroom.

Randomization, Random Allocation Concealment, and Blinding

Per the study design, all classrooms conformed to the same schedule for the first four weeks of the study: two weeks of baseline assessment without intervention (Condition A; instrumental audio/no intervention), followed by two weeks of active intervention with daily in-classroom use of the “I Wear Sunscreen Every Day” audio with lyrics and video intervention (Condition B).) Assignment for post-intervention follow-up was done by classroom using a 1:1 computer-generated randomization scheme. Two classrooms were assigned to active treatment follow-up in which they received two more weeks of daily use of the audio with lyrics and video intervention (Condition B). The other two classrooms were assigned to no-intervention follow-up (Condition A, instrumental audio/no intervention). Teachers were informed of the outcome of their classroom randomization status as they administered the assigned intervention daily.

Efforts to preserve the blind included the following: a) delivering the same intervention across the four classrooms on the days that the students were rated for effective sunscreen application by members of the study team; b) training the entire team (study staff, teachers, and data analysts) to adhere to a strict evaluation protocol that ensured that the outcome assessors, study statistician, and PI remained blinded to follow-up condition assignment throughout the study.

Study Conditions

Baseline (instrumental audio/no intervention, Condition A).

During the two-week baseline period, kindergarten students (all classes) were offered six and a half minutes of class time daily, just before the noon-time recess, for the sunscreen project. Here they were allotted two minutes to get ready (e.g., sunscreen is brought out by teacher and distributed), two and a half minutes for to engage in the task of self-application of sunscreen (the sunscreen task), and two more minutes for clean-up and putting away sunscreen. For uniformity, teachers read a brief explanatory script stating that students should return to their assigned seats at their worktables, find their personal tube of sunscreen, and prepare to put on their sunscreen. The teacher then played a 2.5-minute instrumental audio recording and told the children to complete their sunscreen application by the time the song ended. This audio recording was an instrumental version of the “I Wear Sunscreen Every Day” song, which is identical in length to the video intervention but includes no verbal instructions or guidance aside from “Get some sunscreen in your palms, kids!” No further instructions and no coaching/correction about sunscreen application were offered to the students. This was an inert condition designed to be identical in length and teacher attention as the active intervention condition (described below).

Intervention (audio with lyrics and video intervention, Condition B).

During the two-week intervention period, all 96 kindergarten students were administered the 2.5-minute intervention (i.e., “I Wear Sunscreen Every Day” video) in class daily, immediately prior to release for their midday recess. Teachers read a brief explanatory script encouraging the students to “watch the children in the video closely”, listen to the lyrics of the song (which guided sunscreen application), and follow along with the video’s instructions regarding the systematic approach sunscreen application. The teacher script also instructed the children to complete the sunscreen task by the time the video song ends, put their sunscreen away, and sit down when they are done so that the teacher would know that they are finished.

Follow-Up Conditions (A or B).

In the two-week period immediately following the intervention phase, kindergarten classrooms were randomly assigned to return to Condition A (instrumental audio/no intervention) or to continue with Condition B (audio with lyrics and video intervention).

Teacher and Clinical Assessor Training

Teacher Training.

In preparation for this study, teachers initially completed six, one-hour training sessions that: a) provided them with documents from the study protocol manual; b) familiarized them with the baseline, intervention, and follow-up conditions; c) provided them with the required teacher instructions (script) to give to the class; and d) helped them practice the basics of scoring their class using the study standard worksheets (classroom maps). Training began three months prior to the commencement of study recruitment and occurred on successive weeks during the before-school meeting time allotted for teacher preparation. During these training sessions, the study team emphasized the importance of adherence to the study protocol and script, as well as consistency and timeliness with respect to reporting difficulties and/or adverse events. Strategic plans for management of the protocol with classroom substitutes were also discussed. Finally, the importance of “keeping the blind” during the follow-up period was emphasized, and teachers were informed that they should not discuss the follow-up condition (video or no video) assigned to their classroom with members of the study staff.

Upon completion of these training sessions, teachers completed a testing session in which they rated videos of children performing the sunscreen task using a classroom-based map document (Figure 3).

Figure 3.

Classroom map instrument used to assess the primary outcome. Each circle corresponds to one student. Teachers used this sheet every day of the intervention to determine each student’s performance according to the rubric, as well as to provide qualitative feedback.

Inter-rater reliability for scoring children’s performance with the sunscreen task was high across raters, with the percent agreement ranging from 96 – 100% across raters for the successive videos. Finally, once the trial began, weekly, one-hour preparatory meetings between the teachers and the study team were conducted to ensure proper execution of each study phase. “Wrap-up” sessions facilitated by study team members and attended by teachers were offered at the end of each study phase. These sessions provided support, coaching, and a brief “check-in” regarding questions, concerns, fidelity, and compliance with study procedures, as well as any study-related “events” that warranted consideration.

Clinical Team Assessor Training.

Selected clinical members of the study team received training pertinent to scoring effectiveness of sunscreen application for a select subset of students (Figure 1). This process included formal training in a two-part scoring rubric. Part one focused on completeness of coverage and was designed to identify body sites where the students applied sunscreen. Using a body map scoring sheet (Figure 3), clinical team members assigned a “yes/pass” (designated by a checkmark) for body site application if the student touched this area with sunscreen during the 2.5-minute allotted timeframe, and they assigned a “no/fail” (designated by an “X”) for body site application if the student did not touch this site with sunscreen during the same timeframe. If the body part was more than 50% covered by clothing, that area was not expected to be touched by sunscreen and was designated by a “C” for “covered”. Training emphasized teaching clinical team members to use the body map scoring sheet to identify body parts covered by clothing/shoes prior to the music starting. Part two of the scoring rubric focused on the child’s systematic application of sunscreen and is designed to distinguish between children who take an orderly approach (i.e., head to toe) to sunscreen application from those who choose body parts at random for sunscreen application. Students who began with the head/face and work downward toward the feet were scored “yes/pass” for systematic approach; else, they were scored “no/fail.”

Clinical team members, when they had completed their training on the scoring rubric, observed a series of three distinct training video clips that showed a group of elementary school students applying sunscreen using the “I wear sunscreen everyday” audio with lyrics and video intervention. This practice allowed them to hone their scoring skills, ask questions, and eliminate points of confusion regarding what constitutes effective sunscreen application. As a final part of the observer training, clinical team members who were designated assessors were then tested using three video clips showing different groups of elementary students independently applying sunscreen. Inter-rater reliability for scoring of the video clips was calculated between assessors, and training was terminated when agreement was at or above 89% for two consecutive trials.

Feasibility Metrics

Conforming to current guiding protocols for feasibility studies^{c.f., 32} data pertinent to key study processes, resources, management processes, scientific outcomes and assessments, and safety metrics was systematically gathered throughout the study preparation period, as well as all three study phases: baseline, intervention and follow-up. These metrics are delineated in Table 1. Scientific elements of feasibility, e.g., formal assessment of primary, secondary and tertiary outcomes, the assessment of acceptability, satisfaction and perceptions of intervention utility and convenience, and quality assurance metrics related to the implementation of the study, are described in greater detail below.

Table 1.

Feasibility Metrics

Category	Metric	How this was monitored and assessed	Importance to Phase II/III Trials
Process
	Recruitment and retention	Number/proportion of eligible students recruited into the trial. Number/proportion of eligible students who refuse enrollment. Number/proportion of enrolled students who attrit after day one of study baseline. Documentation of identified barriers to recruitment/retention. Documentation of identified factors associated with attrition.	This will provide information regarding optimization of recruitment and retention practices. This will provide insight to potentially modifiable barriers to recruitment and retention.
	Protocol adherence rate	Number and proportion of days for each study phase (and across all study phases) in which teachers failed to administer the sunscreen intervention.	This will provide insight regarding the feasibility of administering a classroom intervention daily for a sustained period of time with kindergarten students.
	Collaboration and “buy-in”	Documentation of any difficulties associated with collaboration between classroom teacher, school/district administration regarding the project. Identification of any barriers to collaboration and buy in.	This will provide insight regarding building collaborative and trusting relationships between schools, districts and research teams.
	Study implementation barriers	Documentation of any identified teacher-level barriers (e.g., difficulty with the study script, materials, audio/video materials or equipment or unavoidable/environmental factors that impacted fidelity to the study protocol). Documentation of student-level barriers to implementing the trial including behavioral difficulties, lack of interest/cooperation, broader practical difficulties of assessing young children’s skill gains with fidelity.	This will provide insight to factors that need attention/revision for future trials.
	Adherence to regulatory guidance and practice	Documentation of any barriers to maintaining student sunscreen or other study materials in compliance with the school/district/study policy guidelines and protocols. Documentation of any barriers associated with appropriate maintenance of student/teacher data in compliance with the study protocol.	This will provide insight to real-world factors that may impact the ethical conduct of a trial in a classroom setting. Gathering data on such factors during a feasibility trial may guide the development of a study protocol that can address these factors in future studies.
Resources
	Classroom/School/District resources required	Classroom space and equipment required for the research. Teacher time, availability, and willingness required to participate. Collaborative relationship with school, district. Willingness of parents to enroll students in the trial. Willingness of students to participate.	This will provide insight into the real-world aspects of conducting clinical trials in public school settings. This data may facilitate the development of realistic study protocols that anticipate teacher/classroom/school and district needs for research participation. This may also provide insight into facilitators and barriers for students/parents to engage in school-based trials.
	Financial resources needed for the research.	Documentation of actual expenditures of the study and other resources (e.g., audio/video equipment, teacher stipends, etc.) that might be useful or necessary in a larger study.	This data will guide the realistic planning for a next-steps study including Phase II/III trials.
	Teacher training needs	Documentation of the training needs of teachers to implement the study protocol (including intervention correctly), teacher preferences, length of study overview meetings, frequency and length of intervention training sessions, including number/length of sessions required for mastery of the scoring of student engagement in study tasks.	This data will provide a realistic picture into the preparation needs and help guide realistic planning for a full-scale trial.
	School/district training needs	Documentation of the number of desired/required meetings, emails, discussions and phone calls required to get district buy-in and cooperation, school buy in and cooperation, and school/district readiness to implement the protocol with fidelity.	This will help inform future research projects, including our planned trial, with respect to the nuances of school/district level policies around research and management of sunscreen within school settings. It will also highlight communications/communication styles that foster strong collaborations between research teams and school/district settings.
	Study team training needs	Documentation of the number of didactic trainings and meetings dedicated to informing and educating the study team about school/district policies, norms, expectations for researchers while on site, management of student data and all other aspects of interfacing with a non-clinical and real world setting in which research is being conducted.	This information will highlight the time and resources required to ensure that a clinical research team is well-prepared to conduct a trial within a public-school setting.
	Translation resources	Documentation of the type, amount, and cost associated with translation resources that enable parents who do not speak or read English to consent for their child to participate in the study, receive appropriate study updates and communicate with the study team/Institutional Review Board (IRB) if/when necessary.	This information will guide planning of inclusive study protocols that anticipate the needs of diverse students and their parents.
Management
	IRB Approval procedures	Documentation of communications with the IRB and study staff, and track time from submission of the protocol to approval.	This information will guide planning of a future study with appropriate time for IRB review and approval.
	Study staff preparation time for study launch	Documentation of the time required for all preparatory actions to the trial, including all meetings with the study staff, IRB, etc. Documentation of all communications with students’ parents ahead of enrollment and throughout the implementation of the study.	This information will help to provide a realistic picture to guide the planning of future studies.
	Communication between study staff and teachers/school/district	Documentation of all communications between teachers, students, parents and the study team throughout the trial.	This information will help to provide a realistic picture to guide the planning of future studies.
	Time and accuracy of data collection/entry	Documentation of how long data collection and data entry take, any difficulties, inaccuracies and how they were collected. Confirmation of completion of data entry, record time to collect, enter and check using the double entry method.	This will shed light on the utility of our data collection process, including the usability of the classroom-maps designed for scoring student engagement in the sunscreen task. This information will help guide the planning of a future trial, both in terms of design of outcome measures, scoring systems, and other processes for evaluating students, but also in terms of planning for resources (time/money) for data collection, entry, cleaning and the like.
	Reporting and handling of adverse events, serious adverse events and clinical emergencies.	Documentation of the study team’s use and handling of standard university protocol for reporting of all adverse events, serious adverse events and clinical emergencies.	This will inform future studies regarding safety procedures to be implemented.
Scientific
	Participant engagement in the sunscreen task.	Documentation of the “success rate” or, the proportion of student participants successfully engaging in the sunscreen task. This data was gathered daily, by classroom teachers, throughout each phase of the study. (Primary Outcome)	This is a key metric of feasibility, as understanding how likely and capable kindergarten students are of consistently engaging appropriately in the sunscreen task (with and without the audio with lyrics and video intervention) is crucial to planning the full-scale trial.
	Caveats to Participation (Intervention Spoilers)	Documentation of factors that impact successful student participation (intervention spoilers) in the sunscreen task, including adverse events, serious adverse events, clinical emergencies (e.g., sunscreen reactions, sunscreen in eyes, nose, mouth), behavioral disruptions, and/or student loss of interest in the intervention.	This information will help to inform the iterative refinement of the intervention, intervention script, teacher training to implement the intervention and broader planning for safety in the Phase II/III trial(s).
	Treatment effects	Documentation of pre-post intervention (secondary outcomes) and post-intervention to follow-up (maintenance effects/tertiary outcomes) changes in competence with sunscreen application. Student competence with sunscreen application is defined as proportion of exposed skin areas touched by sunscreen during the sunscreen task. This outcome was assessed via direct observation of students by clinical study team members.	This preliminary data will help shape our understanding of how the audio with lyrics and video intervention impacts students’ competence with self-application of sunscreen. It will also give us an opportunity to examine the dose-response relationship to the intervention (e.g., do those who are assigned to the audio with lyrics and video intervention at follow-up show greater gains in competency relative to those who only had the intervention for two weeks?). Taken together, this information will help us to develop appropriate hypotheses for the full-scale trial. These preliminary outcomes will also help to determine effect size and contribute towards the determination of appropriate sample size for future studies.
	Acceptability/participant experience/perception	Documentation of student and teacher satisfaction with the audio with lyrics and video intervention, student/teacher/parent comments/complaints about participation in the study. Assessment of teacher perceptions of the audio with lyrics and video intervention’s convenience, value, and willingness to use again/recommend this intervention to a friend/colleague.	This data provides important insight into the perceptions of participants and stakeholders as to the value of this intervention, and the overall experience of participation in a research trial that is embedded in the classroom curriculum. This will be highly valuable information for planning the Phase II/Phase III trial(s).
	Trial integrity	Documentation of key quality assurance measures taken, including efforts: a) to preserve the blind; b) minimize missing data; and c) manage and monitor data throughout the trial.	This information will help with the design and implementation of a future trial(s) by helping to shape the quality assurance protocol(s) that are suitable to those efforts.
Safety		Followed standard university policy to record all adverse events, serious adverse events and clinical emergencies. The Principal Investigator asked teachers about any adverse events/serious adverse events or clinical emergencies during weekly study meetings throughout all phases of the trial and made notes if/when necessary. All adverse events/serious adverse events and clinical emergencies were thoroughly documented within the study protocol logbook. Reportable events (none occurred) were to beimmediately disclosed to the Institutional Review Board.	This information will help to determine the safety and feasibility of implementing this audio with lyrics and video intervention in future studies.

Outcome Assessment

Primary Outcome:

The primary outcome for our study was productive engagement with the 2.5-minute self-application sunscreen task. This outcome was assessed daily on all participants by classroom teachers throughout the baseline (no intervention), video intervention (intervention) and follow-up phases. Productive engagement incorporated student performance on three distinct tasks: 1) starting the activity on time (i.e., student has opened the sunscreen bottle and is applying sunscreen within the first 40 seconds of the 2.5-minute task); 2) maintaining the activity for the duration of the 2.5 minute task period (i.e., actively applying sunscreen on exposed skin); and 3) completing the activity appropriately and on time, meaning the student is able to close the bottle, wipe off hands, and put the bottle away by the conclusion of the 2.5-minute task period. For our purposes, productive engagement was a binary (yes/pass or no/fail) variable in which a “yes/pass” score reflects a student’s fulfillment of all three criteria, else “no/fail”. To ensure uniform assessment of the starting task criterion, a “beep” was embedded into the audio only (no-intervention) condition and video (intervention) condition at the 40-second mark to alert teachers to the conclusion of the “start of task” period.

Primary Endpoint:

The primary endpoint was defined as pre-post intervention changes in percentages of students achieving “yes/pass” scores for productive engagement.

Secondary/Tertiary Outcomes:

The secondary outcome was defined as pre-post intervention changes in the percentage of exposed body parts to which students applied sunscreen during the 2.5-minute sunscreen task. A “yes/pass” designation for effective application of sunscreen required that all exposed body parts received a “yes/pass” score, else “no/fail.” A complete delineation of the scoring rubric for this outcome is presented in Figure 3. The tertiary outcome (maintenance effects) was defined as post-intervention to follow-up changes in the percentage of exposed body parts to which students applied sunscreen during the 2.5-minute sunscreen task.

Secondary/Tertiary Endpoints:

The secondary endpoint was defined as pre-post intervention changes in the percentages of exposed body sites touched with sunscreen during the 2.5-minute sunscreen task. The tertiary endpoint was defined as the post-intervention to follow-up changes in the percentages of exposed body sites touched with sunscreen during the 2.5-minute sunscreen task.

Intervention Detractors

Productive Engagement Spoilers (descriptive).

As described above, one component of productive engagement in the sunscreen task related to a student’s ability to refrain from distracting actions during the 2.5-minute sunscreen task. Teachers provided qualitative data about the nature and scope of the “spoiler” or disruptive action (e.g., will not stop singing the song; poured sunscreen on the desk) for each student who do not pass this component of the productive engagement assessment on each observation day. They also provided general qualitative descriptions of how their class was doing with the intervention (e.g., any children crying) to inform the study team more fully regarding the classroom impact of this intervention.

Classroom Mess.

Teachers also provided a count of the number of desks that needed to be cleaned by an adult (i.e., wiped off) after the 2.5-minute sunscreen task on each day of study baseline, intervention, and follow-up. They also provided other, open-ended descriptive narratives (e.g., description of the mess) of any other issues related to disturbance of the classroom environment that was a result of the students accessing sunscreen in the classroom.

Acceptability, Satisfaction and Perceptions of Intervention Utility

Teacher perceptions of the audio with lyrics and video intervention were measured through single item variables (satisfaction, value, convenience, willingness to use again/recommend to a friend) scored on an 11-point scale (range: 0 – 10) in which lower scores reflect negative perceptions and higher scores more positive ones.

Student perceptions of the audio with lyrics and video intervention acceptability were measured via a single question with a binary “yes/no” response choice. Specifically, kindergarten students were asked if they like the “I Wear Sunscreen Every Day” video intervention enough to recommend that it be used for next year’s kindergarten “e.g., Should your teacher do this again next year?”. Responses were tallied in a “vote” where students placed “yes” responses (a happy face card) or “no” responses (a sad face card) into the ballot box.

Quality Assurance Measures

Preserving the Blind.

Study staff responsible for student observations remained blinded to treatment allocation, as did study staff responsible for data entry. Observing staff refrained from discussing random assignment allocation with classroom teachers and observed students without the use of the video regardless of the class condition to preserve the blind. Study staff entering data had no contact with the classroom environment and remained unaware of the classroom follow-up condition assignment of the students/teachers as they entered data.

Minimization of Missing Data.

Screening and baseline data were closely monitored by study staff to minimize the potential for missing data values. Study staff worked with translators to help parents (of participating students) who struggled to complete the baseline assessment, e.g., who could not complete the form on their own. Students, their parents, and teachers were blinded to randomization allocation assignment until the first day of follow-up, reducing the chance of post randomization attrition.

Treatment Fidelity.

Adherence to the study protocol during baseline, intervention, and follow-up periods was maintained in the following manner. First, preparatory and wrap-up study meetings were held with teachers during each phase of the trial. Second, teachers completed the classroom map-based scoring every day, which documented students’ daily engagement in the study task. Third, planned and surprise visits to the classroom during the study period allowed for confirmation that the study procedures were being carried out as planned. Finally, the sunscreen bottles were weighed before distribution at baseline and weekly through all study phases (baseline/Condition A; intervention/Condition B; and Follow-up) so that decrements in bottle weight over the course of the study could be documented as additional, confirmatory evidence that study materials (e.g., sunscreen) were being consistently used.

Data Management and Confidentiality.

Data-entry conformed to the gold standard: double data entry into a secure online database. Diagnostic descriptive analyses (e.g., range, mean, median, mode, standard deviations) were performed prior to our study analyses to identify outliers or abnormalities in the data and as a final check for entry errors (e.g., the detection of the “impossible” value). Data was de-identified using unique study ID numbers that were generated for each student and teacher. A key connecting these IDs to the participant (either student or teacher) was stored separately from the data and consent forms until the completion of all components of the study and study analyses. Data was reported in aggregate, and individual students are not identifiable in any scholarly or educational products that will result from this work. Electronic data was initially stored on encrypted devices and then transferred to a secure online database. Paper data is stored in two locked cabinets in a locked office accessible only to one of the principal investigators. These measures are in accordance with the highest ethical, privacy, and security standards outlined by the sponsoring university and consistent with district guidelines.

Data Monitoring Committee.

Our study did not require the use of a data monitoring committee. Rather, teachers, parents, and students were encouraged to report any problems that arise, and the study staff were present at the school frequently to address any possible problems. Moreover, this study featured a behavioral intervention with sunscreen, and while adverse outcomes were possible (e.g., getting sunscreen in one’s eyes), our outcomes did not yield information that required close monitoring by a data safety committee as would be appropriate in a pharmacologic trial.

Stopping Guidelines.

Interim analyses conducted throughout the trial helped the study team stat prepared to adhere to ethical stopping guidelines if any of the following cases materialized: 1) students were sufficiently disruptive that the intervention video severely distracted from classroom time or functioning; 2) students’ skill with sunscreen was sufficient at baseline such that further intervention was deemed unnecessary; and 3) frequency and count of adverse events (e.g., allergic reactions) were also reviewed at the completion of baseline to determine if the trial should be stopped.

Adverse Events

Adhering to standard research guidance and policies pertinent to the conduct of human subjects’ research, an adverse events protocol was developed and approved prior to the commencement of the study. Adverse events excluded expected outcomes, including standard allergic or skin reactions to sunscreen, stinging of the eyes or other transient injuries associated with inappropriate sunscreen use, and developmentally appropriate outcomes (e.g., putting your sunscreen on your desk mate and causing distress) pertinent to kindergarten students. Adverse events were to be reported to the Principal Investigator, Institutional Review Board, and logged into the master study file. Serious adverse events, should they occur, regardless of their relationship to the study protocol or not, were to be immediately reported to the study Principal Investigator, and the Institutional Review Board in accordance with study protocol.

Analysis Plan

Descriptive statistics (e.g., means, standard deviation, percent, frequencies, and interquartile range) were used to characterize participant characteristics (age, gender, skin type), recent sunburn exposure, and sun safety habits including average sun exposure during peak hours (summer, school year), use of protective clothing (hats, long sleeves, sunglasses), and family use of sunscreen. For families that reported non-use of sunscreen, reasons for non-use were characterized by frequency and ranked in order of most to least common. For families that reported use of sunscreen prior to enrollment in our study, times when they apply and intervals for re-application of sunscreen were also characterized by frequency and ranked from most to least common.

Contextual variables, including spoilers to productive engagement and classroom mess, were listed, and, where appropriate, characterized by descriptive statistics (e.g., frequency counts, rank order). Descriptive statistics were also used to characterize acceptability metrics, including teacher and student perceptions of the intervention (satisfaction, convenience, value), and weekly changes in sunscreen bottle weights.

Primary (proportion of students productively engaged in the intervention) and secondary (proportion of students who are successful with the “inclusion of exposed body parts”) endpoints were evaluated via logistic regression designed to estimate the pre-post treatment change in the proportion of students who achieve a “yes/pass” score for productive engagement (primary outcome) in the sunscreen application task. To accomplish this, we estimated the average daily proportion of students who scored “yes/pass” for these endpoints during the second week of the baseline and intervention periods using the generalized estimation equation (GEE) method to account for repeated observations of the same student and classroom environment. Data from the first week of the study period was used as a covariate. Given the descriptive nature of a feasibility trial, no additional covariates (e.g., age, prior use of sunscreen) were entered into the models. The “pass” rates were calculated using the generalized estimation equation (GEE) method to account for repeated observations of the same student and classroom. Ninety-five percent confidence intervals will be derived for successful rates in the two periods. A similar approach will be used to address tertiary (maintenance effects) endpoints.

Discussion

This study evaluated the feasibility of incorporating a song-based, video-guided sunscreen application intervention into kindergarten health education. To the best of our knowledge, this represents the first effort to develop and evaluate a skills-based sunscreen education intervention for young elementary students. It is the first study to explicitly examine kindergarteners’ skills related to independent application of sunscreen without adult assistance. If effective, this intervention could have a significant benefit for the students, their teachers, and, most broadly, elementary health education. First, our intervention may foster habit formation and skill development (effective self-application of sunscreen) that could have far reaching benefits in terms of lifetime reduction in UVR exposure for our kindergarten participants. Second, teachers’ understanding of optimally effective strategies for applying sunscreen as well as their understanding of students’ learning needs and skill development trajectories associated with safe and effective use of sunscreen are expected to increase because of participation in this study. This, in turn, could foster broader and more effective efforts to meaningfully incorporate sunscreen education into elementary education classroom settings.

There were few expectations of a negative impact on students or teachers associated with this study. Disruption to the classroom environment may have resulted from the intrusion of a study team into the classroom. Students might have gotten sunscreen in their eyes or mouth. Moreover, parents might have found some aspects of the study intrusive (e.g., baseline questions about skin type, family sunscreen habits) or harbored concerns about the confidentiality of their student’s study related data. We anticipated some concerns by parents, teachers, and school administrators about the potential for such a study to negatively impact the school day, over-burden teachers, or introduce novel safety concerns (e.g., safe storage of large quantities of sunscreen in the classroom) as well. However, we ensured that all such concerns were addressed before the study began and planned numerous opportunities for discussions with the school district, school administration, and parents of kinder student participants. As this study featured a randomized follow-up condition in which students in some classes were allocated to a “continued video intervention condition” while students in other classes reverted back to using an inert intervention (instrumental audio only) during sunscreen application, it was conceivable that disappointment or jealousy in students who were no longer able to access the video, and/or that assignment to the inert condition (instrumental audio only) would result in more limited impact on skill development in those students. However, all participants were given unrestricted access to the intervention video at the study conclusion.

The results of this study will be disseminated in the form of published scholarly works (e.g., journal articles) and formal conference presentations. Our goal is to reach a variety of relevant groups including educators, dermatologists, pediatricians, policymakers and other clinical stakeholders (e.g., at the American Academy of Dermatology meeting, Society for Pediatric Dermatology meeting and American Academy of Pediatrics meetings). Parents of elementary school students are also relevant stakeholders, and our team will endeavor to find parent friendly publication outlets (including district wide newsletters) that will optimize the dissemination of this work. Finally, findings were disseminated to study participants and their parents through a specially arranged meeting for that purpose.

Next steps include efforts to revise the intervention based upon “lessons learned” from this feasibility trial, pilot the revised intervention, and ultimately prepare for a formal efficacy and effectiveness trial of this intervention. Hypotheses for a full-scale trial will relate to evaluating the dose-response relationship between use of the song-based, video-guided intervention and skill gains (e.g., ability to consistently apply sunscreen to all exposed areas of skin using a systematic or head-to-toe approach) so that formal recommendations about how to implement this skill-based protocol in real world classroom settings may be made. The evaluation of learning loss, or what happens when young students participate in a skills-based learning intervention, make skill-gains (e.g., with sunscreen competency) but then discontinue the intervention, was beyond our scope. We were unable to examine long term outcomes or changes in skills for sunscreen application in this cohort across the entire school year, however, future studies may be able to examine the learning / skill acquisition trajectory over months, or even years. This, coupled with efforts to examine the association between the use of sunscreen interventions and skin health (e.g., frequency of sunburns) in elementary students, represent important next steps. Efforts to adapt this intervention to be developmentally appropriate for other age groups (e.g., older elementary school students) are also needed, as are implementation and dissemination studies that highlight the most effective means of spreading this intervention efficiently throughout U.S. elementary schools.

Figure 4.

Body map instrument used to quantify the secondary outcome. Each numbered area corresponds to a unique skin region evaluated for the following binary values: 1) sunscreen applied to the region (marked with a check) or 2) sunscreen not applied to the region (marked with an “X”). Gray areas correspond to pre-determined regions that were not evaluated because they are almost always covered by clothing or hair. Duplicate regions were also grayed-out. Of the 28 regions under evaluation, those that were covered by clothing on the day of the evaluation were also marked with a “C” by the observers.