Studies have evaluated the extent to which nutrition and physical activity policies are established and implemented in schools and districts (9-21). Few studies have assessed awareness of policies as an outcome separate from implementation (17) or examined factors that influence policy implementation (9,13,16,18). One study noted that a lack of awareness among principals of district wellness policies impeded implementation (18), but no studies have examined teacher awareness of policies.

The Centers for Disease Control and Prevention (CDC) encourages state education and health agencies to develop coordinated school health programs to promote health-related policies (22). In 2001, the Utah Department of Health (UDOH) created the Gold Medal Schools (GMS) program (19,23) to promote student food and physical activity policies at elementary schools on the basis of CDC guidelines (24,25) and the Utah State Office of Education core curriculum (26).

We created a 27-question online survey (Appendix) by using Zoomerang (MarketTools, San Francisco, California). We adapted survey questions from those currently in use in GMS program performance measurement surveys. We also used findings from the literature and GMS teacher focus groups to guide question development. Before sending it to teachers, we administered the survey to 8 UDOH volunteers who were familiar with school health programs. We made minor changes to ensure it would be relevant, understandable, and easily navigated by teachers. The UDOH institutional review board approved the study, and a CDC human research protection coordinator also reviewed it. We calculated that at least 858 respondents would be necessary to accurately estimate teacher characteristics within 3%.

We asked teachers whether their school had the GMS food policy or GMS physical activity policy. Teachers who responded yes were considered to be aware of the policy; teachers who responded no or "don't know" were considered to be unaware of the policy. We asked teachers who indicated that they were aware of either the food or physical activity policy whether they helped develop each policy. We also asked teachers about implementation of the GMS food and physical activity policies, familiarity with the GMS program and coordinator, opinions about childhood obesity and its prevention, and their school's status for Adequate Yearly Progress (AYP), an academic performance measurement defined by the No Child Left Behind Act (27). We asked about certain characteristics of their class, their school, and themselves, including questions on the number of years they have been a teacher at their current school and at any school. For physical activity policy implementation, we asked teachers to identify a range of how many minutes of structured physical activity their students received each week. We did not include ranges of weekly physical activity less than 90 minutes as response options. (Because the GMS policy stipulates that students in grades kindergarten through 2 need only to receive a minimum of 45 minutes of weekly structured physical activity, we did not include teachers of these grades in the analysis of physical activity policy implementation.) To decrease potential social desirability bias and increase the response rate (by lowering the fear of negative repercussions), we did not collect any potentially identifying information about schools or teachers.

We gathered principal and teacher e-mail addresses from school Internet sites and by request to school districts. We asked principals by postcard and e-mail to forward the survey link to classroom teachers. We also sent teachers the survey link directly by e-mail. We informed teachers that survey participants would have the opportunity to be entered in a drawing for 1 of 200 online-redeemable $25.00 bookstore gift certificates. We stated on the e-mails and introductory survey page that teachers should take the survey only once. We closed the survey 1 month after launch. Teachers implied their consent to participate by advancing to each question on the survey. On the survey closing page, we invited teachers to submit their name via e-mail to the GMS program if they wished to be entered in the drawing.

Of all respondents, 546 (44%) were aware of the food policy at their school, 550 (44%) were aware of their school's physical activity policy, and 318 (26%) were aware of both policies. Among teachers aware of policies, 94 (17%) helped develop the food policy, and 80 (15%) helped develop the physical activity policy. Of all respondents, 404 (33%) reported that they never use food as a reward or punishment in their classroom; of these, 277 (69%) were aware of the food policy. Of the 710 teachers in grades 3 through 6, 333 (47%) were aware of their school's physical activity policy, and 275 (39%) reported that their students participated in 90 minutes or more of structured physical activity per week; of these, 160 (58%) were aware of the physical activity policy. Of the 710 teachers in grades 3 through 6, 88 (12%) reported implementation of both the food and physical activity policies.

Knowing the location of the school's written food and physical activity policies, being reminded of food and physical activity policies at least once per semester, and teaching at a school with a physical education (PE) specialist were each associated with both awareness and implementation of both food and physical activity policies (Table 1). No significant difference existed in awareness between teachers who reported being reminded once per semester versus once per month. Teacher sex, years teaching at the school, and number of students in class were not associated with awareness or implementation of either policy.

Teacher awareness of the food policy was associated with knowing the location of written policies, knowing the identity of the program coordinator, being reminded of policies at least once per semester, having a PE specialist at the school, and teaching at a school that failed AYP in the previous year (Table 2). Teacher implementation of the food policy was associated with being aware of the school's food policy and teaching at a school with 700 students or more. For physical activity policy awareness, a significant interaction existed between knowing the location of written policies and believing students to be more overweight than in the past. Knowing the location of written policies was significantly associated with awareness of the physical activity policy only among teachers who thought students were more overweight than they used to be. Other factors associated with awareness of the physical activity policy were knowledge of the program coordinator, being reminded of policies at least once per semester, having a PE specialist at the school, having more than 5 years of teaching experience, and thinking that the school offers adequate time for physical activity. Teacher-reported implementation of the physical activity policy was significantly associated with being reminded of policies at least once per semester, being aware of the school's physical activity policy, and thinking that the school offered adequate time for physical activity. Neither thinking the school had adequate physical activity facilities nor having faculty meetings at least weekly was associated with awareness or implementation in any of the models.

For school policies to achieve their intended outcome, at least 3 actions are necessary: 1) policies should be established at schools, 2) school personnel should be made aware of the policies, and 3) policies should be implemented by school personnel. After a policy is established, multiple factors at the school can enhance or impede policy awareness and implementation. Our study demonstrates that teacher awareness of policies promotes policy implementation. We also identify simple interventions — such as involving teachers in policy development and keeping them continually apprised of school policies — to improve teacher awareness and implementation of policies.

Two survey findings highlight the need to educate teachers about the childhood obesity epidemic: 1) knowing the location of written policies was associated with physical activity policy awareness only among teachers who thought students were more overweight than they used to be (75% had this opinion); and 2) believing that students were more overweight than they used to be was associated with food policy implementation. Teachers who view childhood obesity as an emerging concern likely place more emphasis on policies intended to prevent obesity in children.

The finding that teachers were more likely to be aware of policies if their school had a PE specialist reinforces existing recommendations for schools to have a certified faculty member dedicated to PE (22). The finding that teaching at a school that failed AYP in the previous year was associated with teacher awareness of food policy was somewhat unexpected. One explanation could be that schools that pass AYP may prioritize academic subjects over childhood obesity prevention policies. Research (17,29) suggests that emphasis on health subjects in schools has decreased since the advent of academic testing mandated by No Child Left Behind (27). Additional factors significantly associated with teacher awareness and implementation of policies included a longer teaching career, larger school size, and thinking that a school had adequate time for student physical activity; these contribute to the understanding of factors that might influence policy effectiveness but do not lend themselves easily to intervention.

Research on factors that influence food and physical activity policy awareness and implementation at elementary schools, particularly among teachers, is lacking. This study identifies several such factors. Future research should focus on additional factors (eg, emphasis on health subjects, student socioeconomic status [15-17,30], school-area population density [15,30]). Our findings apply to Utah elementary schools active in the GMS program, but they may also apply to other elementary schools with childhood obesity prevention policies.

This study has several limitations. First, because we did not collect identifying information about schools, we could not control for school-level factors not included on the survey (eg, student socioeconomic status and school-area population density) or assess potential clustering effects based on different numbers of respondents per school. However, multivariate analysis of previous GMS survey data indicated that eligibility for free or reduced-price lunch and whether the school's county was urban or rural were not significantly associated with teacher awareness of the policies and that there was no significant clustering effect on the estimates by number of respondents per school (unpublished GMS survey data, 2008-2009). Second, although the anonymity of the survey might have encouraged more teachers to respond and to be more candid in their responses, it also prevented us from determining how nonrespondents differed from respondents, which is important, given the 28% response rate. Nonrespondents would arguably be less interested in GMS policies than respondents and be expected to report lower awareness and implementation of food policies. The purpose of this study, however, was to determine factors associated with policy awareness and implementation; it is unclear whether nonrespondents would differ significantly from respondents on these factors. Some evidence suggests that respondents were representative of the target teacher population. For example, the distribution of respondents among grade levels was similar to the distribution of the teachers in the population surveyed. In addition, we ascertained from the identified data collected in the gift certificate drawing, which 49% of respondents entered, that respondents represented at least 79% of the districts targeted by the survey. The third major limitation was that analysis of factors associated with implementation of the physical activity policy was limited to responses from teachers in grades 3 through 6, so findings on physical activity policy may not be generalizable to other grades.

William A. Lanier, Food and Drug Administration, Center for Food Safety and Applied Nutrition. At the time of the study, Dr Lanier was assigned to the Utah Department of Health as an Epidemic Intelligence Service (EIS) Officer, EIS Field Assignments Branch, Centers for Disease Control and Prevention. He is also affiliated with the US Public Health Service. 5100 Paint Branch Pkwy, College Park, MD 21113, Email: william.lanier/at/fda.hhs.gov.