Economic Evaluation of the US Environmental Protection Agency's SunWise

Program: Sun Protection Education for Young Children

Jessica W. Kyle, James K. Hammitt, Henry W. Lim, Alan C. Geller, Luke H.
Hall-Jordan, Edward W. Maibach, Edward C. De Fabo and Mark C. Wagner
Pediatrics 2008;121;e1074-e1084
DOI: 10.1542/peds.2007-1400

The online version of this article, along with updated information and services, is
located on the World Wide Web at:
http://www.pediatrics.org/cgi/content/full/121/5/e1074

PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

publication, it has been published continuously since 1948. PEDIATRICS is owned, published,
and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk
Grove Village, Illinois, 60007. Copyright © 2008 by the American Academy of Pediatrics. All
rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.

Downloaded from www.pediatrics.org. Provided by Mc Master University on November 8, 2009

ARTICLE

Economic Evaluation of the US Environmental

Protection Agency’s SunWise Program: Sun
Protection Education for Young Children
Jessica W. Kyle, BAa, James K. Hammitt, PhDb, Henry W. Lim, MDc, Alan C. Geller, RN, MPHd, Luke H. Hall-Jordan, BAe,
Edward W. Maibach, PhD, MPHf, Edward C. De Fabo, PhDg, Mark C. Wagner, SBa

aICF International, Washington, DC; bCenter for Risk Analysis, Harvard University, Boston, Massachusetts; cDepartment of Dermatology, Henry Ford Hospital, Detroit,

Michigan; dDepartment of Dermatology, Boston University School of Medicine, Boston, Massachusetts; eUS Environmental Protection Agency, Washington, DC; fCenter
of Excellence in Climate Change Communication Research, Department of Communication, George Mason University, Fairfax, Virginia; gLaboratory of Photobiology and
Photoimmunology, Department of Environmental and Occupational Health, School of Public Health and Health Services, George Washington University Medical Center,
Washington, DC

Financial Disclosure: Dr Lim is a consultant for several sunscreen manufacturers (La Roche-Posay, Orfagen, Dow Pharmaceutical Sciences) and has received research grants from Johnson and Johnson; the
other authors have indicated they have no financial relationships relevant to this article to disclose.

What’s Known on This Subject What This Study Adds

Although a number of studies have evaluated the behavioral impacts of school-based To our knowledge, this study is the first to assess the net benefits of a school-based
and community sun safety education programs, few have evaluated the economics of skin cancer prevention program. This study shows that modest behavioral changes
skin cancer prevention programs. Some studies have analyzed the economics of other resulting from SunWise may result in significant reductions in skin cancer incidence
school-based health programs. and mortality.

ABSTRACT
OBJECTIVE. The SunWise School Program is a school-based sun safety education pro-
gram that was developed by the US Environmental Protection Agency and aims to
teach children how to protect themselves from overexposure to the sun. The objec- www.pediatrics.org/cgi/doi/10.1542/
peds.2007-1400
tives of this study were to assess the health benefits of the SunWise School Program
and use economic analysis to determine the program’s net benefits and cost-effec- doi:10.1542/peds.2007-1400
tiveness. The views expressed herein are solely
those of the authors and do not represent
METHODS. Standard cost/benefit and cost-effectiveness analysis methods were used. the official opinion of the US EPA.
Intervention costs were measured as program costs estimated to be incurred by the Key Words
skin cancer, prevention, environmental
US government, which funds SunWise, using 3 funding scenarios. Health outcomes
health, school health, cost/benefit analysis,
were measured as skin cancer cases and premature mortalities averted and quality- cost-effectiveness, schools
adjusted life-years saved. These health outcomes were modeled using an effective- Abbreviations
ness evaluation of SunWise based on pretest and posttest surveys administered to UV— ultraviolet
students who participated in the program and the Environmental Protection Agen- EPA—Environmental Protection Agency
QALY— quality-adjusted life-year
cy’s peer-reviewed Atmospheric and Health Effects Framework model. Costs averted AHEF—Atmospheric and Health Effects
were measured as direct medical costs and costs of productivity losses averted as a Framework
result of SunWise. Net benefits were measured as the difference between costs BCC— basal cell carcinoma
averted and program costs. SCC—squamous cell carcinoma
CMM— cutaneous malignant melanoma
RESULTS. Economic analysis indicated that if the SunWise School Program continues SPF—sun protection factor
SCUP-h—Skin Cancer
through 2015 at current funding levels, then it should avert ⬎50 premature deaths, Utrecht-Philadelphia– human
nearly 11 000 skin cancer cases, and 960 quality-adjusted life-years (undiscounted) HRQoL— health-related quality of life
among its participants. For every dollar invested in SunWise, between approximately Accepted for publication Oct 18, 2007
$2 and $4 in medical care costs and productivity losses are saved, depending on the Address correspondence to Mark C. Wagner,
funding scenario. SB, ICF International, 1725 Eye St, NW, Suite
1000, Washington, DC 2006. E-mail:
CONCLUSIONS. From a cost/benefit and cost-effectiveness perspective, it is worthwhile to mwagner@icfi.com

educate children about sun safety; small to modest behavioral impacts may result in Address reprint requests to Luke H. Hall-
Jordan, BA, Stratospheric Protection
significant reductions in skin cancer incidence and mortality. Division, Office of Atmospheric Programs,
US Environmental Protection Agency, 1200
Pennsylvania Ave, NW (6205J),

S KIN CANCER ACCOUNTS for more than half of all cancers diagnosed in the United
States, and ⬎1 million new cases of nonmelanoma skin cancer are expected to
occur in 2008.1 Incidence rates of melanoma are rising in most of the world.2 More PEDIATRICS (ISSN Numbers: Print, 0031-4005;
Online, 1098-4275). Copyright © 2008 by the
Washington, DC 20460. E-mail: hall-jordan.
luke@epa.gov

than 8400 Americans are expected to die in 2008 from melanoma, the most com- American Academy of Pediatrics
monly lethal form of skin cancer.3 Research has shown an association between
exposure to ultraviolet (UV) radiation and the development of melanoma and nonmelanoma skin cancers,4 and sun

e1074 KYLE et al
Downloaded from www.pediatrics.org. Provided by Mc Master University on November 8, 2009
exposure during childhood seems to be an important
risk factor associated with the development of mela-
noma skin cancer.5–8 Stratospheric ozone depletion has
exacerbated these health effects by allowing increased
UV radiation to reach Earth’s surface.9
Major skin cancer prevention strategies focus on reduc-
ing overexposure to UV radiation through increasing
knowledge and awareness, modifying sun safety practices,
and implementing policy measures and environmental in-
itiatives.10 These strategies can lead to important reduc-
tions in future skin cancer incidence and mortality for
populations at risk. The Task Force on Community Pre-
ventive Services, supported by the Centers for Disease
Control and Prevention, found that there was sufficient
evidence to support education and policy approaches in
primary schools and recreational and tourism settings.11
The SunWise Program, the first national health and
environmental education program for sun safety de-
signed for children in elementary and middle schools,
was developed by the US Environmental Protection
Agency (EPA) to teach children and their caregivers how
to protect themselves from overexposure to the sun.
SunWise aims to reduce the incidence of skin cancer and
other UV-related health problems by changing attitudes
and behaviors concerning sun exposure. The SunWise
School Program is the major programmatic component
of SunWise, and all public and private elementary and
middle schools in the United States are eligible to par-
ticipate. From its inception in 1999 through August
2007, ⬎15 600 schools registered to use SunWise, rep-
resenting ⬎12% of the ⬎123 000 US elementary and
secondary schools.12 Registered schools receive a free
SunWise tool kit with classroom activities for grades
kindergarten through 8, a UV-sensitive Frisbee for
hands-on experiments, story books, posters, videos, pol-
icy guidance, and other materials.
The cross-curricular, standards-based classroom les-
sons offered by SunWise were reviewed by an expert
panel of educators, curriculum specialists, and skin can-
cer researchers before the program was launched to
ensure scientific accuracy, age-appropriateness, and
alignment with national education standards. Lessons
focus on 3 key areas: effects of UV radiation, risk factors
for overexposure, and sun protection habits. Each lesson
consists of developmentally appropriate activities that
combine education about sun protection and the envi-
ronment with other aspects of students’ regular learning
in science, social studies, health, physical education, lan-
guage arts, and mathematics. One advantage of SunWise
is that it requires relatively little classroom time to im-
plement; nearly 90% of nurses and teachers that have
led the program reported that lessons took 1 to 2 hours.13
The SunWise School Program is an effective means of
improving students’ sun protection knowledge, atti-
tudes, and behaviors.13–15 These improved behaviors can
prevent future cases of skin cancer; however, because
resources for funding school-based health programs are
limited, as is the amount of curriculum time that school
administrators can devote to health-related topics, demon-
strating effectiveness may not be sufficient to justify pro-
gram implementation. Evaluations of the cost-effectiveness
Downloaded from www.pediatrics.org. Provided by Mc Master University on November 8, 2009
of school-based programs can provide a component of
that validation, but few studies, to our knowledge, have
analyzed the economics of school-based health pro-
grams, and none has analyzed school-based sun safety
programs. For addressing this gap in the literature, the
objectives of this study were to assess the health benefits
of the school component of the SunWise Program and
use economic analysis to determine the program’s net
benefits and cost-effectiveness.
METHODS
Framework for Economic Analysis
Standard cost/benefit and cost-effectiveness analysis
methods were used to evaluate the costs and health
outcomes of SunWise compared with a no-intervention
alternative. Intervention costs were measured as pro-
gram costs estimated to be incurred during a 17-year
operation of the program (fiscal years 1999 –2015).
Health outcomes were measured as skin cancer cases
and premature mortalities averted and quality-adjusted
life-years (QALYs) saved. Averted costs associated with
skin cancer cases and premature mortalities prevented
were measured as direct medical costs and costs of pro-
ductivity loss averted by SunWise. All costs are reported
in 1999 dollars. Because there is a lag period between
the intervention and averting skin cancer, health out-
comes, QALYs, and costs averted were calculated over
the period 1999 –2100 and were discounted at a rate of
3% as recommended by the US EPA16 and the Panel on
Cost-effectiveness in Health and Medicine.17
Health outcomes and averted costs were calculated
under 3 intervention cost scenarios: (1) the current
funding scenario, in which funding for the SunWise
School Program, including personnel costs, continues at
its current levels from fiscal years 1999 through 2015
(approximately $926 000 per year, or 85% of the total
SunWise Program budget); (2) an increased funding sce-
nario, in which funding for the school component in-
creases from current levels to $1.4 million per fiscal year
from 2008 through 2015; and (3) a low funding sce-
nario, in which no funding is provided from 2008
through 2015.
For each funding scenario, the base-case analysis was
conducted in 6 steps: (1) effectiveness analysis of Sun-
Wise based on pretest and posttest surveys administered
to students who receive the intervention; (2) modeling
to translate sun safety behavioral changes reported by
students into changes in lifetime UV radiation exposure;
(3) modeling using the US EPA’s peer-reviewed Atmo-
spheric and Health Effects Framework (AHEF) model to
translate changes in lifetime UV exposure into an esti-
mated number of averted skin cancer cases and prema-
ture mortalities; (4) estimation of medical care and
productivity costs averted per skin cancer case and pre-
mature mortality prevented; (5) estimation of QALYs
saved; and (6) calculation of the net benefit of SunWise.
Sensitivity analysis was also conducted on key parame-
ters to evaluate the robustness of base-case results.
PEDIATRICS Volume 121, Number 5, May 2008 e1075
Intervention Costs
The cost of the intervention was measured as the oper-
ating costs of the SunWise School Program from the
perspective of the US government as the funder of Sun-
Wise. The annual cost is based on current and projected
program expenditures for infrastructure (Web site man-
agement, tool kit printing, school registration), outreach
and communication (conference attendance, school re-
cruitment), and evaluation and training, as well as 3 to
4 full-time employees, depending on the funding sce-
nario. For the current funding scenario, the annual cost
of the program was assumed to stay constant through
2015. For the increased funding scenario, certain com-
ponents of the overall program cost (eg, Web site man-
agement, conference attendance) were considered fixed
costs, whereas other costs (eg, tool kit printing, registra-
tion data entry) were estimated on the basis of the
number of additional schools assumed to be registered
each year and the per-school cost.
In addition to the operating cost of the program,
other social costs are incurred by the SunWise interven-
tion; however, these costs are not included in this anal-
ysis. These social costs include private costs incurred by
students to comply with SunWise recommendations,
such as the cost to purchase sunscreen or sun-protective
hats, and opportunity costs associated with teachers’ and
students’ time spent on SunWise rather than other class-
room lessons. Other social costs could include changes
in participants’ happiness resulting from behavioral
changes brought about by SunWise, such as a decrease
in self-perception of beauty stemming from less tanned
skin.
Effectiveness Evaluation of the SunWise Program
Children’s one-time participation in SunWise activities
(1–2 hours spent on classroom lessons) includes a self-
administered pretest on students’ knowledge, attitudes,
practices, and intended practices, followed by classroom
teaching, and then a posttest. Pretest surveys are distrib-
uted in classrooms generally during September through
March, and posttests are distributed generally in May
through June, ⬃2 months after classroom teaching. For
the posttests, teachers were instructed to tell students to
report on their postteaching current practices, and the
posttest survey instrument reflects postteaching prac-
tices. Teachers were asked not to have students complete
the posttest unless they had completed the pretest. Sur-
veys were administered and collected each calendar year
from 1999 through 2005. Because different schools were
part of the evaluation year to year, no student completed
⬎1 set of pretest and posttest. The survey was reviewed
and edited by educational and child development ex-
perts, pilot tested by 9 children aged 6 to 7 for readability
and length, and time-tested to ensure completion within
5 to 7 minutes. Age, gender, and hair color data were
also collected.
Of the nearly 12 100 schools registered as of Decem-
ber 2005, ⬃10% were randomly selected to participate
in the survey process, and ⬃60% of these selected
schools completed both the pretest and the posttest. A
total of 13 791 pretests and 10 299 posttests were com-
e1076 KYLE et al
Downloaded from www.pediatrics.org. Provided by Mc Master University on November 8, 2009
pleted. Respondents were between the ages of 5 and 15
years (median: 10 years). Because students completed
the surveys anonymously, linking individual students’
answers from both surveys was not possible. As a result,
the measure of the effectiveness of SunWise is the dif-
ference between pretest and posttest for the percentage
of students who provided specific responses (instead of
the change in individual students’ responses). The sig-
nificance of the difference between pretest and posttest
was determined using the ␹2 test for 2 ⫻ 2 contingency
tables. Data through 2002 were published elsewhere,14
and survey results through 2005 were presented at the
2006 International Union Against Cancer World Cancer
Congress.18
Number of SunWise Program Participants
The number of children who participated in SunWise in
each fiscal year from 1999 through 2006 was calculated
on the basis of the number of schools and classrooms
that are registered with SunWise and the average class
size for US elementary schools.19 The number of students
who would participate in SunWise in each fiscal year
from 2007 through 2015 was projected on the basis of
historical participation plus projected future participa-
tion. It was assumed that 3000 additional schools would
be registered in 2007 and 3500 schools in 2008. Under
the current funding scenario, 3500 additional schools
were assumed to be registered annually in 2009 through
2015. For the increased funding scenario, the number of
schools registered was assumed to be twice the number
of schools registered annually under the current funding
scenario. The low funding scenario assumed that no new
schools would be registered from 2008 through 2015
and that the number of already registered schools would
decrease to 0 over 4 years (100% of schools implement
SunWise in 2008, 50% in 2009, 25% in 2010, and 0%
from 2011–2015). In each registered school, ⬃2 class-
rooms are assumed to implement SunWise each year,
and only the students in those classrooms are assumed
to participate. Because the same classrooms are assumed
to implement SunWise every year, students are as-
sumed to participate in SunWise only once over their
lifetime. We also assumed that 74% of teachers in reg-
istered schools actually use the SunWise materials, based
on teacher survey results.20 The total numbers of Sun-
Wise students from 1999 through 2015 under the low
funding, current funding, and increased funding scenar-
ios are 3.4 million, 12 million, and 17 million students,
respectively.
Health Outcomes
Health outcomes were calculated for 3 types of skin
cancers: basal cell carcinoma (BCC), squamous cell car-
cinoma (SCC), and cutaneous malignant melanoma
(CMM). Averted skin cancer incidences and premature
mortalities were calculated in 4 steps: (1) modeling an
annual baseline UV exposure dose for students before
participation in SunWise, (2) modeling a SunWise UV
exposure dose for students for the 3 years immediately
after the intervention, (3) comparing that reduction in
UV exposure (baseline UV dose minus SunWise UV
dose) to lifetime UV exposure, and (4) translating that
percentage reduction in cumulative lifetime UV expo-
sure to reduced incidence of skin cancer using the AHEF
model. Registered schools are assumed to implement the
program once per year from 1999 to 2015. Health out-
comes are calculated for each year’s intervention on the
basis of the number of schools registered and then cu-
mulated to estimate the total number of skin cancer
cases and premature mortalities averted by implemen-
tation of SunWise from 1999 through 2015.
For development of a model for estimating students’
baseline and SunWise UV exposure, ambient solar UV
radiation at the Earth’s surface by latitude and month of
the year was first computed using the Tropospheric Ul-
traviolet-Visible radiation model (3.9a).21–25 The accu-
racy of this model has been demonstrated in several
comparisons to direct measurements of UV at the Earth’s
surface,26–29 and the model has been used in many sci-
entific evaluations of ozone depletion.30–37
Data from Godar et al38 estimating the percentage of
ambient UV exposure that children are exposed to by
gender, season, and geographic region were used to
extrapolate students’ potential UV exposure on the basis
of time spent outdoors. These data were based on the
National Human Activity Pattern Survey, which re-
corded the daily minute-by-minute activities of ⬃2000
young adults over 2 years.38 The demographics of Sun-
Wise students (geographic distribution and race) are
assumed to reflect those of the general US population
and to vary over time to reflect changes projected by
the US Census Bureau; these demographic changes
are built into the AHEF model used to predict health
outcomes.
To develop a baseline for children’s UV exposure, we
made several assumptions to adjust the potential
amount of UV exposure from time spent outdoors to
account for the frequency with which students reported
practicing sun protection behaviors (wearing sunscreen,
hats, and long-sleeved shirts) and the effectiveness of
those practices in reducing UV exposure. The reduction
in UV exposure derived from each sun protection behav-
ior was combined to develop a total reduction in UV
exposure, and the difference in this reduction on the
basis of students’ reports before and after intervention
formed the basis for the health effects analysis.
For calculation of these UV exposure reductions, it
was assumed that students who reported that they prac-
ticed a sun protection behavior “all of the time,” “some-
times,” or “never,” exhibited that behavior 75%, 50%,
and 0%, respectively, of the time that they spend out-
side. For long-sleeved shirts39–41 and hats,42–44 Ultraviolet
protection factors of 25 and 2 were selected, respectively.
No measure of lower body coverage was included in the
analysis. We assumed that students applied sun protec-
tion factor (SPF) 15 sunscreen, but only at 25% of the
recommended thickness (2 mg/cm2).45–49 On the basis of
the exponential relationship between SPF and the thick-
ness applied,50 an effective SPF of 2 was calculated. Ef-
fectiveness of sunscreen use was modeled as the prod-
uct of this protection factor and the percentage of
Downloaded from www.pediatrics.org. Provided by Mc Master University on November 8, 2009
body surface area to which sunscreen is applied. Stu-
dents who reported that they applied sunscreen to 6 to
8, 3 to 5, and 0 to 2 body parts (eg, face, arms, legs)
were assumed to be protecting 75%, 50%, and 10% of
their body surface area, respectively. Each body part
was weighted equally.
Students are assumed to participate in SunWise only
once in their lifetime and to retain the sun safety behav-
iors that they report learning at a decreasing rate over
the 3 years after the intervention. We assumed that in
the first year after participation in SunWise, 100% of
students still practice learned sun safety behaviors; in the
second year, 50% of students display these behaviors;
25% in the third year; and in the fourth year and be-
yond, students revert to sun safety behaviors practiced
before participation in SunWise.
The difference between students’ UV exposure before
and after SunWise was compared with an estimate of
lifetime UV exposure to calculate a percentage reduction
in lifetime UV exposure associated with a one-time par-
ticipation in SunWise. Lifetime UV exposure was esti-
mated by first calculating annual UV exposure for girls
and boys aged 1 to 18 years on the basis of (1) annual
ambient UV radiation at the Earth’s surface by latitude,
(2) the percentage of ambient UV exposure that children
receive annually, and (3) sun protection behavior re-
ported in the SunWise pretest and posttest surveys. An-
nual exposure for ages 1 to 18 years was totaled, and
data from Godar et al51 on the percentage of UV expo-
sure received by 18 years of age was used to extrapolate
lifetime UV exposure.
The AHEF model was used to translate the cumula-
tive percentage change in UV exposure into skin cancer
cases and premature mortalities averted. The AHEF is a
peer-reviewed model that is typically used to evaluate
human health impacts associated with changes in emis-
sions of ozone-depleting substances.52,53 For this analysis,
only the AHEF module that translates changes in
ground-level UV (in this case, changes in actual UV
exposure) into changes in health outcomes was used.
This module applies calculated dose-response relation-
ships to the baseline skin cancer incidence/mortality and
change in UV exposure to calculate averted incidence/
mortality in each year through 2100.
The dose-response relationships used in the AHEF
model measure the degree to which changes in UV ex-
posure weighted by the appropriate action spectrum
cause incremental changes in health effects. An action
spectrum describes the relative effectiveness of energy at
different UV wavelengths in producing a particular bio-
logical response. The AHEF relies on the Skin Cancer
Utrecht-Philadelphia– human (SCUP-h) action spectrum
(derived on the basis of the induction of SCC in hairless
mice and corrected for human skin transmission) to
predict incidence of SCC, BCC, and CMM (because a
mammalian action spectrum for CMM remains to be
determined).52
Costs Averted
Averted costs of BCC, SCC, and CMM prevented were
calculated as the number of cases averted multiplied by
PEDIATRICS Volume 121, Number 5, May 2008 e1077
 TABLE 1 Total Cost per Case of Nonfatal Skin Cancer and Mortality
(1999 $)
Parameter Medical Productivity
Cost, $ Loss Cost, $
Nonfatal skin cancer case
BCC 825a 959b
SCC 825a 3698b
CMM – –
Skin cancer mortality – –
Totals may not sum due to independent rounding.
a Chen et al,54 adjusted to 1999 dollars using the medical care component of the Consumer Price
Index.
b Calculated by authors, based on US EPA55 and US Bureau of Labor Statistics.56
c US EPA,55 adjusted to 1999 dollars using the medical care component of the Consumer Price
Index.
d US EPA,16 based on the central estimate recommended for benefits analysis from a review of
26 risk value of a statistical life studies.
the medical and productivity loss cost per case. Cost per
case is for cancer care only and excludes the costs of
unrelated care, such as increased costs for treating other
medical conditions later in life that might have occurred
after the predicted skin cancer mortality. Total averted
costs were calculated over 1999 through 2100 and dis-
counted to 1999 using a discount rate of 3%. The med-
ical costs and productivity loss per case are shown in
Table 1.
Medical costs per case of BCC and SCC were based on
Chen et al54; that study used data from the Medicare
Current Beneficiary Survey (1999 –2000) to estimate
medical treatment costs associated with BCC and SCC in
different practice settings. To determine an average med-
ical treatment cost per case, we calculated weighted
averages on the basis of the percentage of episodes man-
aged in each setting.
Productivity loss costs were based on a US EPA anal-
ysis supporting the Regulatory Impact Analysis: Protec-
tion of Stratospheric Ozone.55 We calculated cost per
case by multiplying EPA’s estimates of the loss of work as
a result of illness and caregiving performed by others for
the patient for BCC and SCC by the national mean
annual wage for 1999.56 For CMM, we used EPA’s esti-
mate of the total medical cost and productivity loss per
Total Cost per Case/
Mortality, $ case. The value of a statistical life was based on US EPA’s
Guidelines for Preparing Economic Analyses.16
1785
4523 QALY Losses Averted
28 857c For estimation of QALY losses averted, total life-years
6 100 000d saved were calculated and then adjusted to account for
health-related quality of life (HRQoL). We assumed that
the duration of each nonfatal case of skin cancer was 1
year. For premature mortalities avoided, the number of
life-years saved was calculated by comparing the age at
which each premature mortality was projected to have
occurred (in the absence of the SunWise Program) with
the average life expectancy for a person at that age.57
The life-years saved were weighted to consider
HRQoL. For premature mortalities avoided, the average
HRQoL at the age at which the mortality would have
occurred (“without-condition” HRQoL, based on a scale
from 0 to 1) was compared with the HRQoL associated
with death (“with-condition” HRQoL, or 0). For nonfatal
cases, the average HRQoL at the age at which the case
would have occurred was compared with the HRQoL
associated with having skin cancer. The difference be-
tween with-condition and without-condition age-spe-
cific HRQoL was multiplied by the duration of the mor-
bidity to estimate QALY losses averted.
The age-specific without-condition HRQoL was based
on the mean EuroQol-5D index scores for US adults as
derived from the 2000 –2002 Medical Expenditure Panel
Survey.58 Because a EuroQol-5D index score was not pro-
vided for people who were younger than 18 years, we
applied the mean index for adults aged 18 to 29 years for
age cohorts aged 10 to 19 years. For the with-condition
HRQoL for melanoma and nonmelanoma skin cancer, ex-
pert assigned values from Freedberg et al,59 as interpreted
by the Tufts-New England Medical Center Institute for
Clinical Research and Health Policy Studies (data are avail-
able from the corresponding author), were used.

TABLE 2 Sensitivity Analysis Upper and Lower Bounds

Variable Base-Case Value Sensitivity Analysis
Upper Bound Lower Bound
Behavioral retention rate Year0 ⫽ 100%, Year1 ⫽ 100%, Year0 ⫽ 100%, Year1 ⫽ 100%, Year0 ⫽ 100%, Year1 ⫽ 50%,
Year2 ⫽ 50%, Year3 ⫽ 25%, Year2 ⫽ 100%, Year3 ⫽ 50%, Year2 ⫽ 25%, Year3 ⫽ 0%,
Year4 ⫽ 0% Year4 ⫽ 25% Year4 ⫽ 0%
Total SunWise classrooms (1999–2015)a Current funding ⫽ 766 798 Current funding ⫽ 1 043 452 NA
Increased funding ⫽ 1 066 521 Increased funding ⫽ 1 451 312
Low funding ⫽ 218 585 Low funding ⫽ 297 448
Percentage of time spent on behaviors 75 100 50
reportedly practiced “all the time”
Percentage of lifetime UV exposure before Northern girls ⫽ 23% Northern girls ⫽ 46% Northern girls ⫽ 12%
age 18 Northern boys ⫽ 21% Northern boys ⫽ 43% Northern boys ⫽ 11%
Southern girls ⫽ 23% Southern girls ⫽ 47% Southern girls ⫽ 12%
Southern boys ⫽ 23% Southern boys ⫽ 46% Southern boys ⫽ 11%
7% discount rate, % 3 7
All estimates have been independently rounded. As such, upper and lower bounds may not sum. NA indicates not applicable.
a This value does not represent the number of unique classrooms that have registered, but rather, the total number of classrooms that have participated in the SunWise program. Thus, if a classroom

registers for the program in 1999 and continues to implement the program each year until 2015, that classroom is counted in this total 17 times.

e1078 KYLE et al
Downloaded from www.pediatrics.org. Provided by Mc Master University on November 8, 2009
 TABLE 3 SunWise Program: Evaluation of Student Responses (All Ages), 1999 –2005
Parameter Pretest Posttest Difference, %
(n ⫽ 13 791), % (n ⫽ 10 299), %
Knowledge
Sun can hurt one’s skin 74 79 5a
Wearing hat and shirt are ways to protect one from 61 72 11a
the sun
Sun protection and UV index 25 49 24a
The right number SPF to use 49 73 24a
Attitudes
Suntan is good for my skin 29 22 ⫺6a
People look healthier with a suntan 41 38 ⫺3a
Practices
Wear sunscreen all the time 25 26 1
Apply sunscreen to ⱖ5 body parts 71 76 4a
Wear hats all the time 14 16 3a
Wear long-sleeved shirt all the time 3 4 1a
Wear sunglasses all the time 20 21 1b
Intended practices
Do you think you will put on sunscreen this summer? 45 49 4a
I will try to play in the shade instead of the sun 67 73 6a
All estimates have been independently rounded. As such, the posttest response minus the pretest response may not equal the difference. The
significance of the difference between pretest and posttest responses was tested by using the ␹2 test for 2 ⫻ 2 contingency tables.
a Level of statistical significance is P ⬍ .001.

b Level of statistical significance is P ⬍ .05.

Survey data source: Geller AC. The Environment Protection Agency’s SunWise Program (1999 –2005) [abstract]. In: Proceedings of the UICC World
Cancer Congress 2006. Bologna, Italy: Medimond; 2006.

Sensitivity Analysis RESULTS

To test the robustness of the base-case results, we con-
ducted a univariate sensitivity analysis over a reason-
able range of values for key parameters: behavioral
retention rate (the rate at which students retain
learned sun safety behaviors over the years after the
intervention), number of classrooms participating in
SunWise, the percentage of time that students actually
perform a behavior that they reported practicing “all
the time,” and the percentage of UV exposure received
before age 18.51
Table 2 presents upper and lower bound estimates for
each of the key parameters. Upper and lower bounds for
percentage of lifetime UV received before age 18 were
based on a twofold increase and decrease in the base-
case value. Because the number of classrooms that par-
ticipate in SunWise may be underreported (because only
teachers who also sign up to report UV measurements
report the number of participating classrooms), only an
upper bound was tested at a 50% increase in the base-
case value.
Evaluation of the SunWise Program
As shown in Table 3, the greatest improvement between
pretests and posttests was shown for the knowledge
variables; changes in students’ attitudes, practices, and
intended practices were less marked, but still statistically
significant. Only the measures of sunscreen, hats, and
long-sleeved shirts practices are used as inputs for mod-
eling the change in children’s UV exposure associated
with participation in SunWise.
Economic Analysis Results
As shown in Table 4, if the school component of the
SunWise Program continues through 2015 at current
funding levels, then it should avert ⬎50 premature deaths
and almost 11 000 skin cancer cases among its participants.
More than 70 premature deaths and ⬎15 200 skin cancer
cases could be averted if funding for SunWise were in-
creased to $1.4 million per year in 2008 –2015. In addition,

TABLE 4 Averted Nonfatal Cases, Mortalities, and QALYs Associated With the SunWise Program, Fiscal Year 1999 –2015
Funding Scenario Nonmelanoma Skin Cutaneous Malignant Total Cases Total Premature QALY Losses Averted
Cancer Melanoma Averted Mortalities Averted
Cases Premature Cases Premature Undiscounted Discounted
Averted Mortalities Averted Mortalities at 3%
Averted Averted
Current funding scenario 10 696 19 264 33 10 960 51 960 159
Increased funding scenario 14 877 26 367 45 15 244 71 1335 217
Low funding scenario 3049 5 75 9 3124 15 274 52
These averted cases and mortalities are based on baseline rates from the AHEF model for light-skinned individuals and are blended across different age cohorts. Totals may not sum due to
independent rounding.

PEDIATRICS Volume 121, Number 5, May 2008 e1079

Downloaded from www.pediatrics.org. Provided by Mc Master University on November 8, 2009
 TABLE 5 Base-Case Cost/Benefit Analysis Results (1999 $) incremental cost-effectiveness associated with moving
Funding Scenario Averted Intervention Net Benefit from 1 funding level to a higher level is nearly constant,
Costs Costs and the average cost-effectiveness for SunWise increases
with the scale of the program.
Current funding scenario 43 391 914 12 194 833 31 197 080
Increased funding scenario 59 345 457 14 772 974 44 572 484 The results of the sensitivity analysis (Table 6) indi-
Low funding scenario 14 078 062 7 211 715 6 866 347 cate that changes in the values of key variables affect
All estimates have been independently rounded As such, averted costs minus intervention
health and economic benefits. All sensitivity cases result
costs may not equal net benefit. in positive net benefits, with the exception of the sce-
narios in which a 7% discount rate is used. Net benefits
are also not achieved in the low funding scenario when
between ⬃270 and 1300 QALY losses (undiscounted) a lower bound for percentage of lifetime UV exposure
could be averted by SunWise. before age 18 is used. The results of the sensitivity anal-
Table 5 presents the results of the base-case cost/ ysis on other variables were found to have minimal
benefit analysis. Net of program costs, SunWise could influence on benefits.
generate between approximately $7 million and $45
million in net savings, depending on the level of future DISCUSSION
funding provided to SunWise. For every dollar invested We evaluated the net benefits and cost-effectiveness of a
in SunWise, between $1.95 and $4.02 in medical care school-based intervention that educates children about
costs and productivity losses could be saved. how to protect themselves from overexposure to the sun
Program costs include both fixed costs and variable to reduce the incidence of skin cancer and other UV-
costs that depend on how many classrooms are regis- related health problems. The results of our study suggest
tered. In contrast, the benefits are nearly proportional to that from the perspective of a cost/benefit and cost-
the number of classrooms enrolled. Consequently, the effectiveness analysis, it is worthwhile to educate chil-

TABLE 6 Sensitivity Analysis Results

Variable Total Cases Total Premature Averted Costs, Intervention Net Benefit, QALY Losses
Averted Mortalities Averted 1999 $a Costs, 1999 $a 1999, $a Averteda
Current funding scenario
Behavioral retention rate
Upper bound 14 945 70 58 373 901 12 194 833 46 179 067 214
Lower bound 6974 33 27 960 467 12 194 833 15 765 634 102
SunWise classrooms
Upper bound 14 914 70 59 094 203 12 194 833 46 899 370 272
Percentage of lifetime UV exposure
before age 18
Upper bound 21 920 103 86 783 827 12 194 833 74 588 994 318
Lower bound 5480 26 21 695 957 12 194 833 9 501 124 79
7% discount rate 10 960 51 4 571 495 9 042 975 ⫺4 471 480 24
Increased funding scenario
Behavioral retention rate
Upper bound 20 787 97 79 835 747 14 772 974 65 062 773 292
Lower bound 9701 45 38 240 459 14 772 974 23 467 485 140
SunWise classrooms
Upper bound 20 744 97 80 803 634 14 772 974 66 030 660 372
Percentage of lifetime UV exposure
before age 18
Upper bound 30 487 143 118 690 915 14 772 974 103 917 941 434
Lower bound 7622 36 29 672 729 14 772 974 14 899 755 109
7% discount rate 15 244 71 6 107 119 10 599 436 ⫺4 492 317 33
Low funding scenario
Behavioral retention rate
Upper bound 4260 20 18 938 815 7 211 715 11 727 099 69
Lower bound 1988 9 9 071 487 7 211 715 1 859 772 33
SunWise classrooms
Upper bound 4251 20 19 204 193 7 211 715 11 992 478 88
Percentage of lifetime UV exposure
before age 18
Upper bound 6248 29 28 156 125 7 211 715 20 944 410 103
Lower bound 1562 7 7 039 031 7 211 715 ⫺172 684 26
7% discount rate 3124 15 1 733 926 6 034 593 ⫺4 300 667 9
All estimates have been independently rounded. As such, averted costs minus intervention costs may not equal net benefit.
a Using a 3% discount rate, except where otherwise noted.

e1080 KYLE et al
Downloaded from www.pediatrics.org. Provided by Mc Master University on November 8, 2009
dren about sun protection behavior. Our results show
that small to modest behavioral effects can produce cost-
effective results; the sun safety education delivered by
the SunWise School Program translates into significant
reductions in the risk for developing skin cancer, as well
as averted costs to society.
To our knowledge, this study is the first to assess the
net benefits and cost-effectiveness of a school-based sun
safety intervention. Although a number of studies have
evaluated the behavioral impacts of school-based and
community sun safety education programs,13–15,60–67 our
study is 1 of few economic evaluations of skin cancer
prevention programs, school-based or otherwise. Carter
et al68 modeled the cost-effectiveness of a hypothetical
national skin cancer prevention program in Australia,
based on the existing SunSmart campaign, and con-
cluded that such a program would provide excellent
value for money.
Complementing the assessment of sun protection
habits reported in this study, Geller et al14 found that
children who participated in SunWise experienced an
11% reduction in the sunburning rate (from 66% of
students reporting sunburns at pretest to 55% at post-
test). An 8% decrease from pretest to posttest in the rate
of frequent sunburning (ⱖ3 sunburns per summer) was
also reported. This study followed ⬎500 of the same
students over the course of 2 summers. Because a quan-
titative relationship between sunburn and development
of skin cancer is not known, it was not possible to use
these results in this analysis.
Our study shows that the medical costs and produc-
tivity losses averted by SunWise outweigh the cost of the
program by between 2 and 4 times. These results are
comparable to those of economic analyses of other
school-based health education programs. For example, 1
study examined Safer Choices, a school-based sexually
transmitted disease and unintended pregnancy preven-
tion program for high school students, and calculated a
benefit-cost ratio of 2.65.69 Another study evaluated
Planet Health, a school-based intervention designed to
reduce obesity in youth, and estimated averted medical
and productivity costs equal to $40 991 and 2-year pro-
gram costs of $33 677, which translates to a benefit-cost
ratio of 1.2.70
Our study has several limitations. First, student self-
reporting of sun protection behaviors forms the basis for
the baseline and SunWise scenario UV exposure models.
The sun protection practices reported in SunWise stu-
dent surveys, however, are consistent with those in
other studies of US children, reported by either the child
or the parent.71–73 Verbal reports are used most fre-
quently in sun protection studies; for example, 76 of 81
skin cancer prevention studies cited in the Guide to
Community Services evidence review relied on verbal
reports.11 Although moderate to high levels of agree-
ment between self-report and objective verification of
sunscreen use and protective clothing have been shown,
such assessment methods are impractical in school set-
tings; correlations between self-report and dosimeter
readings have been shown to be fair, although still sta-
tistically significant.74
Downloaded from www.pediatrics.org. Provided by Mc Master University on November 8, 2009
Second, like any complex modeling framework, the
AHEF model— used to estimate health effects in this
study— uses data inputs and computational procedures
that introduce uncertainty to the results. Much of the
uncertainty in the AHEF model is associated with the
choice of the SCUP-h action spectrum, which was de-
rived on the basis of the induction of SCC in hairless
mice and corrected for human skin transmission. Past
peer reviews of the AHEF have agreed that SCUP-h is
the best choice of action spectrum and found the uncer-
tainty to be in an acceptable range.52
Third, not all health outcomes associated with re-
duced lifetime UV exposure, including reducing the
incidence of cataracts and actinic keratosis, minimizing
the effect of photoimmunosuppression,75–81 and mini-
mizing the development of photoaging, are modeled in
this analysis. Averted costs associated with reduced in-
cidence of these additional health outcomes were not
considered, although the large majority of the health
benefit associated with SunWise is believed to be cap-
tured. Averted cataract incidence has accounted for
⬍1% of the total health benefits (ie, averted costs asso-
ciated with cataracts and skin cancer cases/mortalities)
in other studies.55
Fourth, limitations of the analysis may have led to
either overestimations or underestimations of its impact.
Factors other than SunWise, such as community pro-
grams or parent influence, may have resulted in im-
proved sun protection behavior from pretest to posttest.
The absence of a randomly selected control group makes
it difficult to determine definitively whether the mea-
sured changes in students’ sun protection behaviors are
attributable entirely to the SunWise intervention; how-
ever, Geller et al13,14 reported that comparably aged
children who served as non–randomly assigned control
subjects to SunWise experienced no positive changes in
sun protection practices during the 1-year study period.
Conversely, other assumptions may have led to conser-
vative estimates of the averted skin cancer incidence and
costs in this study. For example, in calculating UV pro-
tection resulting from children’s reported sun protective
behaviors, data were not available to adjust sunscreen
coverage for clothing worn; thus, UV protection (and
hence benefits) may be underestimated if, for instance,
children are wearing long-sleeved shirts but not sun-
screen on their arms. In addition, the program impact
may be conservative because the full reach of SunWise
(ie, on the sun protection behavior of teachers and
school nurses teaching the program and participants’
families) was not quantified, although it could be signif-
icant. As of August 2007, 19 840 teachers had registered
to use SunWise. Teacher surveys undertaken for Sun-
Wise indicated that 77% of teachers have made sun
protection habit changes as a result of teaching the pro-
gram.20 Future efforts should focus on overcoming such
limitations to strengthen the cost-effectiveness evalua-
tion of SunWise.
Fifth, private costs for participants to comply with
SunWise recommendations and for teachers to imple-
ment the Program were not included in this analysis. If
the private costs of students’ participation were in-
PEDIATRICS Volume 121, Number 5, May 2008 e1081
cluded, then the net benefit of the program could de-
crease by approximately $5 million to $20 million during
the period 1999 to 2015 (present value discounted at
3%); SunWise would still generate between $2 million
and $24 million in net benefit. These calculations are
based on the behavioral changes measured in the Sun-
Wise student surveys, assuming that sunscreen is reap-
plied every 2 hours spent outside, that 1 oz of sunscreen
is used for every application, that an average 8-oz bottle
of sunscreen costs $8, and that a sun-protective hat costs
$10. Another private cost of the program may be the
opportunity cost of teachers’ time spent on SunWise;
assuming teachers spend 1 hour on SunWise at an av-
erage hourly wage of $33.19,82 these costs could range
from approximately $2 million to $9 million during the
period 1999 to 2015 (at a 3% discount rate); however,
because 1 of the advantages of SunWise is that its activ-
ities are designed to be integrated into normal classroom
lessons, we do not expect teachers to incur this oppor-
tunity cost.
CONCLUSIONS
This study suggests that, after participation in the Sun-
Wise School Program, students make small to modest
behavioral changes that may result in significant reduc-
tions in skin cancer incidence and mortality, which, in
turn, render the program cost saving. These findings
have important implications for pediatricians, research-
ers, policy-makers, and educators. For pediatricians,
communicating ways to reduce adverse sun exposure to
patients and their parents could have sizable impacts on
reducing future skin cancer incidence. For the research
community, economic evaluation and assessment of be-
havioral modifications, along with development of UV
studies for dose-response relationships and a CMM
action spectrum, should continue to play a part in as-
sessing school-based programs to help policy-makers de-
termine which programs reduce student risk behaviors
at what cost. For school administrators, the cost of Sun-
Wise is 0, the time commitment is ⬍2 hours, and, as this
study has shown, the health benefits may be potentially
substantial. Future evaluations of SunWise should seek
to improve the quality of the evidence on the effective-
ness of the intervention and could also consider the
impact of other program components, namely the Sun-
Wise Cities/Communities Program, as well as investigate
the change in impact if students receive SunWise lessons
more than once.
ACKNOWLEDGMENTS
This study was prepared under US EPA contracts EP-W-
06-008 and EP-06-H-002787.
We thank Drusilla Hufford, Margaret Sheppard, Ross
Brennan, Jim Democker, and Linda Rutsch of the US
EPA for insight regarding the operation of the SunWise
Program and comments on early versions of the manu-
script. We also thank David Berv of ICF International for
research contributions to this study.
e1082 KYLE et al
Downloaded from www.pediatrics.org. Provided by Mc Master University on November 8, 2009
REFERENCES
1. National Cancer Institute. Common cancer types. Available
at: www.cancer.gov/cancertopics/commoncancers. Accessed
March 10, 2008
2. Kannan K, Sharpless NE, Xu J, O’Hagan R, Bosenberg M, Chin
L. Components of the Rb pathway are critical targets of UV
mutagenesis in a murine melanoma model. Proc Natl Acad Sci
USA. 2003;100(3):1221–1225
3. American Cancer Society. Cancer facts and figures 2008. Avail-
able at: www.cancer.org/downloads/STT/2008 CAFFfinal
secured.pdf. Accessed March 7, 2008
4. International Agency for Research on Cancer, World Health
Organisation. Solar and Ultraviolet Radiation. Vol 55. Geneva,
Switzerland: International Agency for Research on Cancer;
2002
5. Noonan FP, Recio JA, Takayam H, et al. Neonatal sunburn and
melanoma in mice. Nature. 2001;413(6853):271–272
6. Whiteman DC, Whiteman CA, Green AC. Childhood sun ex-
posure as a risk factor for melanoma: a systematic review of
epidemiologic studies. Cancer Causes Control. 2001;12(1):69 – 82
7. Harrison SL, MacLennan R, Speare R, Wronski I. Sun exposure
and melanocytic naevi in young children in Townsville,
Queensland. Lancet. 1994;344(8936):1529 –1532
8. Zanetti R, Franceschi S, Rosso S, Colonna S, Bidoli E. Cutane-
ous melanoma and sunburns in childhood in a Southern Eu-
ropean population. Eur J Cancer. 1992;28A(6 –7):1172–1176
9. United Nations Environment Programme. Environmental Effects
of Ozone Depletion and Its Interactions With Climate Change: 2006
Assessment. Nairobi, Kenya: United Nations Environmental
Programme; 2006
10. Glanz K, Lew RA, Song V, Cook VA. Factors associated with
skin cancer prevention practices in a multiethnic population.
Health Educ Behav. 1999;26(3):344 –359
11. Glanz K, Mayer JA. Reducing ultraviolet radiation exposure to
prevent skin cancer methodology and measurement. Am J Prev
Med. 2005;29(2):131–142
12. US Department of Education, National Center for Education
Statistics. Digest of Education Statistics, 2005. Washington, DC:
Education Department, Institute of Education Sciences, Na-
tional Center for Education Statistics; 2006. Publication NCES
2006 – 030
13. Geller AC, Rutsch L, Kenausis K, Selzer P, Zhang Z. Can an
hour or two of sun protection keep the sunburn away? Eval-
uation of the Environmental Protection Agency’s SunWise
School Program. Environ Health. 2003;2(1):1–9
14. Geller AC, Rutsch L, Kenausis K, Zhang Z. Evaluation of the
SunWise School Program. J Sch Nurs. 2003;19(2):93–99
15. Geller AC, Cantor M, Kenausis K, et al. The Environmental
Protection Agency’s National SunWise School Program:
1999 –2000. J Am Acad Dermatol. 2002;46(5):683– 689
16. US Environmental Protection Agency. EPA Guidelines for Pre-
paring Economic Analyses. Washington, DC: Office of the Admin-
istrator, US Environmental Protection Agency; 2001. Report
240-R-00-003
17. Gold MR, Siegel JE, Russel LB, Weinstein MC. Cost-effectiveness
in Health and Medicine. New York, NY: Oxford University Press;
1996
18. Geller AC. The Environmental Protection Agency’s SunWise
Program (1999 –2005) [abstract]. In: Proceedings of the UICC
World Cancer Congress 2006. Bologna, Italy: Medimond; 2006
19. US Department of Education, National Center for Education
Statistics. Digest of Education Statistics, 2002. Washington, DC:
Education Department, Institute of Education Sciences, Na-
tional Center for Education Statistics; 2003. Publication NCES
2003– 060
20. Geller AC. SunWise school education in the United States.
 Presented at: UV Alert Stakeholders Meeting; March 9, 2006;
Washington, DC
21. Madronich S. The atmosphere and UV-B radiation at ground
level. In: Young AR, Bjorn LO, Moan J, Nultsch W, eds. Envi-
ronmental UV Photobiology. New York, NY: Plenum Press; 1993:
1–39
22. Madronich S. UV radiation in the natural and perturbed atmo-
sphere. In: Trevini M, ed. UV-B Radiation and Ozone Depletion:
Effects on Humans, Animals, Plants, Microorganisms, and Materials.
Boca Raton, FL: Lewis Publisher; 1993:17– 69
23. Madronich S, de Gruijl FR. Skin cancer and UV radiation.
Nature. 1993;366(6450):23
24. Madronich S, McKenzie RL, Bjorn LO, Caldwell MM. Changes
in ultraviolet radiation reaching the Earth’s surface. J Photo-
chem Photobiol. 1998;46(1–3):5–19
25. Madronich S, Weatherhead E, Flocke S. Trends in UV radia-
tion. Int J Environ Sci. 1996;51(3):183–198
26. Shetter RE, McDaniel AH, Cantrell CA, Madronich S, Calvert
JG. Actinometer and Eppley radiometer measurements of the
NO2 photolysis rate coefficient during MLOPEX. J Geophys Res.
1992;97(D10):10349 –10359
27. Kirk JT, Hargreaves BR, Morris DP, et al. Measurements of
UV-B radiation in two freshwater lakes: an instrument inter-
comparison. Arch Hydrobiol. 1994;43:71–99
28. Lantz KO, Shetter RE, Cantrell CA, Flocke SJ, Calvert JG,
Madronich S. Theoretical, actinometric, and radiometric deter-
minations of the photolysis rate coefficient of NO2 during
MLOPEX II. J Geophys Res. 1996;101(D9):14613–14630
29. Gao W, Slusser JR, Gibson JH, et al. Direct-sun column ozone
retrieval by the ultraviolet multi-filter rotating shadow-band
radiometer and comparison with those from Brewer and Dob-
son spectrophotometers. Appl Opt. 2001;40(19):3149 –3155
30. World Meteorological Organization. Scientific Assessment of Ozone
Depletion: 1989. Geneva, Switzerland: World Meteorological Orga-
nization Global Ozone Research and Monitoring Project; 1990.
Report 20
31. World Meteorological Organization. Scientific Assessment of Ozone
Depletion: 1991. Geneva, Switzerland: World Meteorological Orga-
nization Global Ozone Research and Monitoring Project; 1992.
Report 25
32. World Meteorological Organization. Scientific Assessment of Ozone
Depletion: 1994. Geneva, Switzerland: World Meteorological Orga-
nization Global Ozone Research and Monitoring Project; 1995.
Report 37
33. World Meteorological Organization. Scientific Assessment of Ozone
Depletion: 1998. Geneva, Switzerland: World Meteorological Orga-
nization Global Ozone Research and Monitoring Project; 1999.
Report 44
34. United Nations Environment Programme. Environmental Effects
of Ozone Depletion. Nairobi, Kenya: UNEP Environmental Effects
Panel Report, United Nations Environmental Programme;
1989
35. United Nations Environment Programme. Environmental Effects
of Ozone Depletion. Nairobi, Kenya: UNEP Environmental Effects
Panel Report, United Nations Environmental Programme;
1991
36. United Nations Environment Programme. Environmental Effects
of Ozone Depletion. Nairobi, Kenya: UNEP Environmental Effects
Panel Report, United Nations Environmental Programme;
1994
37. United Nations Environment Programme. Environmental Effects
of Ozone Depletion. Nairobi, Kenya: UNEP Environmental Effects
Panel Report, United Nations Environmental Programme;
1998
38. Godar DE. UV doses of American children and adolescents.
Photochem Photobiol. 2001;74(6):787–793
Downloaded from www.pediatrics.org. Provided by Mc Master University on November 8, 2009
39. Hatch KL, Osterwalder U. Garments as solar ultraviolet radia-
tion screening materials. Dermatol Clin. 2006;24(1):85–100
40. Gambichler T, Rotterdam S, Altmeyer P, Hoffmann K. Protec-
tion against ultraviolet radiation by commercial summer
clothing: need for standardized testing and labelling. BMC Der-
matol. 2001;1:6
41. Dummer R, Osterwalder U. UV transmission of summer cloth-
ing in Switzerland and Germany. Dermatology. 2000;200(1):
81– 82
42. Gies P, Javorniczyk J, Ray C, Henderson S. Measurements of
the UVR protection provided by hats used at school. Photochem
Photobiol. 2006;82(3):750 –754
43. Wong JC, Airey DK, Fleming RA. Annual reduction of solar UV
exposure to the facial area of outdoor workers in Southeast
Queensland by wearing a hat. Photodermatol Photoimmunol Pho-
tomed. 1996;12(3):131–135
44. Diffey BL, Cheeseman J. Sun protection with hats. Br J Derma-
tol. 1992;127(1):10 –12
45. Neale R, Williams G, Green A. Application patterns among
participants randomized to daily sunscreen use in a skin cancer
prevention trial. Arch Dermatol. 2002;138(10):1319 –1325
46. Autier P, Boniol M, Severi G, Doré JF; European Organization
For Research and Treatment of Cancer Melanoma Co-
Operative Group. Quantity of sunscreen used by European
students. Br J Dermatol. 2001;144(2):288 –291
47. Azurdia RM, Pagliaro JA, Diffey BL, Rhodes LE. Sunscreen
application by photosensitive patients is inadequate for protec-
tion. Br J Dermatol. 1999;140(2):255–258
48. Wulf HC, Stender IM, Lock-Andersen J. Sunscreens used at the
beach do not protect against erythema: a new definition of SPF
is proposed. Photodermatol Photoimmunol Photomed. 1997;13(4):
129 –132
49. Bech-Thomsen N, Wulf HC. Sunbather’s application of sun-
screen is probably inadequate to obtain the sun protection
factor assigned to the preparation. Photodermatol Photoimmunol
Photomed. 1993;9(6):242–244
50. Faurschou A, Wulf HC. The relation between sun protection
factor and amount of sunscreen applied in vivo. Br J Dermatol.
2007;156(4):716 –719
51. Godar DE, Urbach F, Gasparro FP, van der Leun JC. UV doses
of young adults. Photochem Photobiol. 2003;77(4):453– 457
52. US Environmental Protection Agency. Human health benefits
of stratospheric ozone protection: peer reviewed report—final
report; 2006. Available at: www.epa.gov/ozone/science/
AHEFDEC2003D3.pdf. Accessed September 13, 2007
53. US Environmental Protection Agency. Human health effects of
ozone depletion from stratospheric aircraft: final report—
2001. Available at: http://gltrs.grc.nasa.gov/reports/2001/CR-
2001-211160.pdf. Accessed September 13, 2007
54. Chen J, Housman T, Williford P, Teuschler H. Estimating cost
of treatment episode of nonmelanoma skin cancer in elderly
[abstract]. J Am Acad Dermatol. 2005;52(suppl):3, P147
55. US Environmental Protection Agency. Regulatory Impact
Analysis: Protection of Stratospheric Ozone, E. Final Report. Wash-
ington, DC: US Environmental Protection Agency; 1988
56. US Bureau of Labor Statistics, Department of Labor. Occupa-
tional Employment Statistics (OES) survey: 1999 national OES
estimates. Available at: www.bls.gov/oes. Accessed September
13, 2007
57. Arias E. United States life tables, 2003. Natl Vital Stat Rep.
2006;54(14):1– 40
58. Sullivan PW, Ghushchyan V. Preference-based EQ-5D index
scores for chronic conditions in the United States. Med Decis
Making. 2006;26(4):410 – 420
59. Freedberg KA, Geller AC, Miller DR, Lew RA, Koh HK. Screen-
ing for malignant melanoma: a cost-effectiveness analysis.
J Am Acad Dermatol. 1999;41(5 Pt 1):738 –745
PEDIATRICS Volume 121, Number 5, May 2008 e1083
60. Milne E, Jacoby P, Giles-Corti B, Cross D, Johnston R, English
DR. The impact of the kidskin sun protection intervention on
summer suntan and reported sun exposure: was it sustained?
Prev Med. 2006;42(1):14 –20
61. Buller DB, Reynolds KD, Yaroch A, et al. Effects of the Sunny
Days, Healthy Ways curriculum on students in grades 6 to 8.
Am J Prev Med. 2006;30(1):13–22
62. Stankeviciute V, Zaborskis A, Petrauskiene A, Valiukeviciene
S. Skin cancer prevention: children’s health education on pro-
tection from sun exposure and assessment of its efficiency.
Medicina (Kaunas). 2004;40(4):386 –393
63. Milne E, English DR, Johnston R, et al. Reduced sun exposure
and tanning in children after 2 years of a school-based inter-
vention (Australia). Cancer Causes Control. 2001;12(5):387–393
64. Hewitt M, Denman S, Hayes L, Pearson J, Wallbanks C. Eval-
uation of “Sun-safe”: a health education resource for primary
schools. Health Educ Res. 2001;16(5):623– 633
65. Milne E, English DR, Johnston R, et al. Improved sun protec-
tion behaviour in children after two years of the Kidskin in-
tervention. Aust N Z J Public Health. 2000;24(5):481– 487
66. Lowe JB, Balanda KP, Stanton WR, Gillespie A. Evaluation of
a three-year school-based intervention to increase adolescent
sun protection. Health Educ Behav. 1999;26(3):396 – 408
67. Olson AL, Gaffney C, Starr P, Gibson JJ, Cole BF, Dietrich AI.
SunSafe in the middle school years: a community-wide inter-
vention to change early-adolescent sun protection. Pediatrics.
2007;119(1). Available at: www.pediatrics.org/cgi/content/
full/119/1/e247
68. Carter R, Marks R, Hill D. Could a national skin cancer primary
prevention campaign in Australia be worthwhile? An eco-
nomic perspective. Health Promot Int. 1999;14(1):73– 83
69. Wang LY, Davis M, Robin L, Collins J, Coyle K, Baumler E.
Economic evaluation of safer choices: a school-based human
immunodeficiency virus, other sexually transmitted diseases,
and pregnancy prevention program. Arch Pediatr Adolesc Med.
2000;154(10):1017–1024
70. Wang LY, Yang Q, Lowry R, Wechsler H. Economic analysis of
a school-based obesity prevention program. Obes Res. 2003;
11(11):1313–1324
71. Cokkinides VE, Weinstock MA, Cardinez CJ, O’Connel MA.
Sun-safe practices in U.S. youth and their parents: role of
e1084 KYLE et al
Downloaded from www.pediatrics.org. Provided by Mc Master University on November 8, 2009
caregiver on youth sunscreen use. Am J Prev Med. 2004;26(2):
147–151
72. Johnson K, Davy L, Boyett T, Weathers L, Roetzheim RG. Sun
protection practices for children: knowledge, attitudes, and parent
behaviors. Arch Pediatr Adolesc Med. 2001;155(8):891– 896
73. Hall HI, Jorgensen CM, McDavid K, Kraft JM, Breslow R.
Protection from sun exposure in US white children ages 6
months to 11 years. Public Health Rep. 2001;116(4):353–361
74. Dwyer T, Blizzard L, Gies PH, Ashbolt R, Roy C. Assessment of
habitual sun exposure in adolescents via questionnaire: a com-
parison with objective measurement using polysulphone
badges. Melanoma Res. 1996;6(3):231
75. Noonan FP, Muller HK, Fears TR, Kusewitt DF, Johnson TM,
De Fabo EC. Mice with genetically determined high suscepti-
bility to ultraviolet (UV)-induced immunosuppression show
enhanced UV carcinogenesis. J Invest Dermatol. 2003;121(5):
1175–1181
76. Prater MR, Gogal RM Jr, De Fabo EC, Longstreth J, Holladay
SD. Immunotoxic effects of cis-urocanic acid exposure in
C57BL/6N and C3H/HeN mice. Photochem Photobiol. 2003;
77(4):383–389
77. IJland SA, Noonan FP, Ceryak S, et al. Urocanic acid does not
photobind to DNA in mice irradiated with immunosuppressive
doses of UVB. Photochem Photobiol. 1998;67(2):222–226
78. Noonan FP, De Fabo EC. Immunosuppression by ultraviolet B
radiation: initiation by urocanic acid. Immunol Today. 1992;
13(7):250 –254
79. De Fabo EC, Noonan FP, Frederick JE. Biologically effective
doses of sunlight for immune suppression at various latitudes
and their relationship to changes in stratospheric ozone. Pho-
tochem Photobiol. 1990;52(4):811– 817
80. De Fabo EC, Noonan FP. Mechanism of immune suppression
by ultraviolet irradiation in vivo: I— evidence for the existence
of a unique photoreceptor in skin and its role in photoimmu-
nology. J Exp Med. 1983;158(1):84 –98
81. Hrushesky W, Sothern R, Wop J, Du Quiton J, Boon M.
Season, sun, sex, and cervical cancer. Cancer Epidemiol Biomar-
kers Prev. 2005;14(8):1940 –1947
82. US Department of Labor, Bureau of Labor Statistics. National
Compensation Survey, All U.S. Secondary School Teachers’ Hourly
Wage. Washington, DC: US Department of Labor; 2005
 Economic Evaluation of the US Environmental Protection Agency's SunWise
Program: Sun Protection Education for Young Children
Jessica W. Kyle, James K. Hammitt, Henry W. Lim, Alan C. Geller, Luke H.
Hall-Jordan, Edward W. Maibach, Edward C. De Fabo and Mark C. Wagner
Pediatrics 2008;121;e1074-e1084
DOI: 10.1542/peds.2007-1400
Updated Information including high-resolution figures, can be found at:
& Services http://www.pediatrics.org/cgi/content/full/121/5/e1074
References This article cites 55 articles, 12 of which you can access for free
at:
http://www.pediatrics.org/cgi/content/full/121/5/e1074#BIBL
Subspecialty Collections This article, along with others on similar topics, appears in the
following collection(s):
Allergy & Dermatology
http://www.pediatrics.org/cgi/collection/allergy_and_dermatolog
y
Permissions & Licensing Information about reproducing this article in parts (figures,
tables) or in its entirety can be found online at:
http://www.pediatrics.org/misc/Permissions.shtml
Reprints Information about ordering reprints can be found online:
http://www.pediatrics.org/misc/reprints.shtml

Downloaded from www.pediatrics.org. Provided by Mc Master University on November 8, 2009