Source: UNIVERSITY OF CALIFORNIA, BERKELEY submitted to
TECHNOLOGY AND DESIGN INNOVATION TO SUPPORT 21ST CENTURY SCHOOL NUTRITION
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
TERMINATED
Funding Source
Reporting Frequency
Annual
Accession No.
1005599
Grant No.
2015-68001-23236
Project No.
CALW-2014-08389
Proposal No.
2014-08389
Multistate No.
(N/A)
Program Code
A2101
Project Start Date
Mar 15, 2015
Project End Date
Mar 14, 2021
Grant Year
2019
Project Director
Madsen, K.
Recipient Organization
UNIVERSITY OF CALIFORNIA, BERKELEY
(N/A)
BERKELEY,CA 94720
Performing Department
Public Health - CHHD
Non Technical Summary
This integrated education, Extension, and research project addresses the need for school-based solutions to address childhood obesity by testing innovative 21st-century student-centered strategies based on behavioral economics to increase student participation in the school lunch program and reduce plate waste. Building upon several successful pilots, the project will be led by university-based child obesity and nutrition researchers in collaboration with Cooperative Extension and school district administration and staff. Using a cluster-randomized controlled trial design, we will assess the 2-year impact of implementing a 3-pronged intervention involving a SmartMeal technology platform, distributed points of sale, and staff promotion of school meals. We will assess impacts on school lunch participation and intake of fruits and vegetables by students from 12 middle and high schools compared to 12 control schools. All schools have diverse student bodies, with 70% of students eligible for free and reduced-price meals on average. We also will determine if the initial investment in the intervention will be returned by increased revenue from school meals. Finally we will examine sustainability by monitoring adoption of the intervention by control schools and continuance of the intervention by intervention schools the year after the trial. We will disseminate findings by publishing a minimum of two journal articles, distributing a policy brief, engaging eXtension's Healthy Food Choices in Schools Community of Practice, giving oral presentations to stakeholders and the scientific community, and creating an online project tool kit so that other schools can replicate the intervention.
Animal Health Component
0%
Research Effort Categories
Basic
(N/A)
Applied
75%
Developmental
25%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
7036010302033%
7246010302034%
8057410303033%
Goals / Objectives
Our long-term goals are twofold: Goal 1) to improve dietary intake and reduce obesity among low-income youth, and Goal 2) to enhance the financial and social sustainability of school foodservices. We have developed the following specific objectives to address each goal.Extension Objectives. Through a partnership with CE and SFUSD and engagement of students, each of the following objectives will support the accomplishment of Goal 1.1. Finalize the student-centered content for the SmartMeal e-application by June 2017.2. Finalize a staff wellness curriculum that educates teachers and staff about school meals by June 2017.3. By August 2015, formalize a partnership between San Francisco and CE that supports continued nutrition education for students, staff, and teachers in SFUSD.Education Objectives. With CE and SFUSD partners, the following objectives will support the accomplishment over 2 years of Goal 1 at 12 randomly assigned intervention middle and high schools compared to 12 randomly assigned control middle and high schools.4. Increase nutrition knowledge, with an emphasis on school meals, by middle and high school students.5. Increase the proportion of school staff who encourage students to eat the school lunch and who model healthful eating behaviors.Research Objectives. Using a cluster-randomized controlled design, the following research objectives support the accomplishment of Goals 1 and 2.6. After two years of implementation among middle and high school students, determine the ability of the intervention, compared to the control condition, to:a. Increase student participation in the school lunch program [Goals 1 and 2].b. Increase student consumption of fruits and vegetables at lunch and daily [Goal 1].c. Increase the variety of vegetables that students consume at lunch and daily [Goal 1].d. Decrease plate waste at school lunch [Goal 2].e. Reduce obesity [Goal 1].7. After two years, determine the intervention's financial sustainability [Goal 2].8. Over one year, assess the replicability and sustainability of the intervention [Goal 2].
Project Methods
To accomplish our research objectives, we will conduct a 2-year cluster-RCT involving 12 middle schools and 12 high schools in San Francisco, half of which will receive the SmartMeal design intervention, and half of which will act as controls. In the subsequent year, we will study the continued application of the intervention in the original intervention schools, and adoption of the intervention by control schools, to determine how best to scale up the intervention. Using a cluster-randomized controlled trial design, we will assess the 2-year impact of implementing the 3-pronged intervention: SmartMeal technology platform, distributed points of sale, and staff promotion of school meals. We will assess impacts on school lunch participation and intake of fruits and vegetables by students from 12 middle and high schools compared to 12 control schools. All schools have diverse student bodies, with 70% of students eligible for free and reduced-price meals on average. We also will determine if the initial investment in the intervention will be returned by increased revenue from school meals. Finally we will examine sustainability by monitoring adoption of the intervention by control schools and continuance of the intervention by intervention schools the year after the trial. We will disseminate findings by publishing a minimum of two journal articles, distributing a policy brief, engaging eXtension's Healthy Food Choices in Schools Community of Practice, giving oral presentations to stakeholders and the scientific community, and creating an online project tool kit so that other schools can replicate the intervention.Measures include:Aggregate Plate Waste at School Lunch: Researchers quantify (by weight) amounts of fruits and vegetables prepared, served, and uneaten by students for 2 days each Spring. Fruits and vegetables served alone and in mixed dishes (limited to dishes containing at least 1/8 cup fruit or vegetables) will be assessed. Aggregate measures quantify waste and dietary intake of individual meal components,81 but capture entire student body, rather than relying on random sample of students.School Foodservice Manager Interview: On-site, 1 foodservice manager per school (n=24) will be interviewed to assess satisfaction with SmartMeal intervention and changes in school food environment that may influence school meal intake or participation. Foodservice Financial Records: The District Foodservice Director will work closely with research staff to document foodservice costs and revenue at the intervention and control schools. Student Survey: Proctored by researchers, completed by students in 7th and 9th (n~6600), ~10 minutes. Questions adapted from Healthful Eating Active Communities Student Surveyand Block Kids Questionnaire, a food frequency questionnaire that includes 13 types of fruits/vegetables as well as numerous commonly consumed mixed dishes containing vegetables.Analyses: Our primary outcome is the difference in change in average daily participation in school lunch over two years between intervention and control schools. We will use a linear regression model in order to adjust for school enrollment, FRP enrollment, and race/ethnic composition. We will also examine changes over one year, when only the front-end of the SmartMeal app is in place. We will use mixed effects models with a group-by-time (intervention vs. control) interaction term as the primary predictor and: 1) change in plate waste (servings of fruits and vegetables by weight) to assess change in fruit and vegetable consumption at lunch and 2) students' reported servings of fruits and vegetables (Block Kids) to assess daily F/V.

Progress 03/15/15 to 03/14/21

Outputs
Target Audience:In 2014, we selected 12 middle and 12 high schools to participate in the study and assigned half (6 middle and 6 high schools) to intervention and half to the control condition. The intervention ended in Spring 2018. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? This study provides intensive training in nutrition-related research methodology for one doctoral student, two masters students and several undergraduate students in public health. Specifically, this project provides training in: research design, survey design, plate waste assessment methodology, study management, and training in ethical research. Significant efforts have been undertaken to plan robust professional development for the teachers in SFUSD. Recognizing the limited time available for teacher professional development, which is most often optional for teachers, SFUSD contracted with a cinematographer to develop an educational video about the school food program that teachers can access at their convenience. The video features SFUSD leadership, principals, teachers, food service workers and students and covers topics ranging from the changes resulting from the Healthy, Hunger-Free Kids Act to the additional points of sale for school meals, and can be broken into shorter segments to appeal to teachers' busy school schedules. SFUSD and UCB collaborated with UCANR to develop educational flyers that correspond to the video segments for distribution to teachers. Twelve flyers were produced and disseminated between August 2016 and May 2018. UCANR hired a Teacher Outreach Coordinator whose job was to visit the 12 intervention schools to support teachers in their efforts to promote the school meal program to students. Finally, each of the 12 intervention schools hosted events for teachers throughout the 2017-18 school year to provide updates about school meals and offer meal taste tests for teachers with the goal of reshaping their perceptions about the school food program. Each of these interventions was informed by teacher input gathered by SFUSD and UCB staff. How have the results been disseminated to communities of interest?As previously reported, the project partners hosted an inaugural Advisory Board (AB) meeting on November 13, 2015. AB members (see list below) received project updates via email in Spring and Fall 2016, Summer 2019, and Winter 2020. The research team convened the AB in December 2019 to share results and gather feedback about how to disseminate findings to influence the field. Further, the team invited policy experts to join the AB in discussing school meal policy implications. AB members include: Becky Lee, IDEO Crystal FitzSimons, Food Research & Action Center Deanna Hoelscher, School of Public Health, University of Texas Elaine Waxman, Urban Institute Katie Kettman, SPUR Kim Frinzell, Nutrition Services, California Department of Education Marisa Neelon, University of California Cooperative Extension Sondra Parmer, Alabama, Cooperative Extension System Sylvia Allegretto, Institute for Research on Labor & Employment, UC Berkeley Tia Shimada, California Food Policy Advocates Zoe Neuberger, Center on Budget and Policy Priorities? What do you plan to do during the next reporting period to accomplish the goals?April 2021:Present lessons learned paper at SOPHE conference. September 2021:Submit manuscript on teacher survey outcomes.

Impacts
What was accomplished under these goals? Obj. 1 Activities:The SmartMeal app was determined infeasible in December 2017.Data collected:14 interviews with SFUSD staff were conducted to assess implementation barriers.Results:SFUSD did not have staff with information technology (IT) or app development expertise involved in early app planning.Discussion:Early inclusion of IT staff may have resulted in clearer, more accurate implementation plan. Obj. 2 Activities:Partners developed and implemented a professional development program to educate SFUSD teachers and staff about school meals.Data collected:Video views and meal tasting attendance were tracked.Results:Presentations at all-staff meetings reached 750 teachers and staff (non-unique). Over 5500 teachers and staff (non-unique) were engaged through taste tests.Discussion:A website is accessible district-wide to provide a continued avenue for sharing nutrition education resources. Obj. 3 Activities:CE and SFUSD hired a Teacher Outreach Coordinator (TOC). The TOC aligned work with SFUSD's Wellness Policy to ensure sustainability. CE and SFUSD developed a toolkit to share teacher outreach best practices with other districts.Data collected:The TOC tracked communication with teachers, administrators, and support staff and worked with SFUSD's School Health Programs to identify school-based Wellness Champions.Results:Relationships were established with 43 teachers, 26 administrators, and 23 support staff. 38 staff were identified as Wellness Champions.Discussion:After the intervention ended, SFUSD hired the TOC full-time to continue related nutrition education outreach activities. Obj. 4 Activities:Due to suspension of the Smart Meals app, students did not receive nutrition education via an app. However, teacher education efforts provided teachers with ideas and strategies to incorporate nutrition education in the classroom. The TOC delivered nutrition education to 360 students.Data collected:Survey data was collected in 2016-2018. 11,805 student surveys were included in the final sample.Results:Among 8th grade respondents, there was a relative increase in perceptions that school that school lunch tastes good (0.19; 95% CI: 0.07, 0.31, p<0.01) in intervention versus comparison schools. No changes were seen in perceptions of fullness or healthfullness of school lunch or lunch line length for any grade.Discussion:Results suggest that the intervention was modestly effective in improving perceptions of lunch quality in grade 8 only, and ineffective in improving perceptions of convenience in all grades. Obj. 5 Activities:A teacher survey was administered in spring 2016-2018.Data collected:The final analytic sample included 556 teacher surveys from 2016 and 533 from 2018.Results:A relative increase in teacher-reported frequency of eating with students in the cafeteria was seen in intervention schools (difference-in-change 1.48; 95% CI: 0.42, 2.53, p<0.01). Students at intervention schools had a relative increase in agreement that adults at their schools encouraged them to eat school lunch (difference- in-change 0.10; 95% CI: 0.01, 0.18, p< 0.05). No between-group differences were seen in teachers' perceptions of school meals or participation in school meals.Discussion:The intervention appeared to have a small yet significant impact on teachers eating in the cafeteria with students and on teacher encouragement of school lunch (among those who received a high intervention dose). Obj. 6a Activities:School lunch participation data was collected in school years 14/15-17/18.Data collected:SFUSD provided daily school lunch participation data and monthly average daily attendance for the intervention and control schools in school years 2014/15 (baseline) - 2017/18 (follow-up).Results:From baseline to follow-up, lunch participation dropped 4.1% in intervention and 5.1% in comparison schools (difference-in-difference 1.0%, 95% CI 0.5, 1.4). Among students eligible for free or reduced-price meals, participation decreased 1.8% in intervention and 4.9% in comparison schools (difference-in-difference 3.1%, 95% CI: 2.5%, 3.7%), with a larger difference-in-difference seen in high schools (5.0%, 95% CI: 4.2%, 5.9%) than middle schools (1.8%, 95% CI: 0.8%, 2.6%).Discussion:While the intervention demonstrated a modest but significant relative increase in school lunch participation, the effect was not sufficient to halt large district-wide declines in participation during this study period. Obj. 6b Activities:Survey data was collected in 2016-2018. 11,805 student surveys were included in the final sample.Data collected:Self-reported daily and lunchtime fruit and vegetable data were collected via student surveys (as described in Objective 4).Results: Significant relative increases in fruit (difference-in-change 6.4%; 95% CI: 0.5%, 12.4%, p<0.05) and vegetable consumption (7.1%; 95% CI: 1.5%, 12.6%, p<0.05) at lunch were seen in grade 10 only. In grade 9 only, a relative increase in daily fruit intake was seen in intervention schools versus comparison schools (difference-in-change 0.10; 95% CI: 0.03, 0.16, p<0.01).Discussion:We observed only modest changes in perceptions of school lunch and fruit and vegetable consumption that were not consistent across all grades, suggesting that additional efforts are needed to improve school-lunch uptake. Obj. 6c Activities:Survey data was collected in 2016-2018.Data collected:Self-reported vegetable variety data were collected via student surveys (as described in Objective 4).Results:No significant differences were seen in daily vegetable variety in both the overall sample and when restricting to those who typically eat school lunch.Discussion:The intervention did not appear to impact vegetable variety. Obj. 6dActivities:The research team conducted plate waste assessments in Spring 2016 and 2018.Data collected:At baseline, 3,158 students participated in lunch data collection across all 24 schools. At follow-up, 2,966 students participated in lunch data collection.Results:From baseline to follow-up, overall plate waste in intervention schools increased 2% (from 37% to 39%), while overall plate waste in comparison schools decreased 8% (from 46% to 38%; p <0.001).Plate waste of fruits in intervention schools increased 7% (from 49% to 56%), while plate waste of fruits in comparison schools decreased 4% (from 59% to 55%); p <0.001). Plate waste of vegetables in intervention schools decreased 6% (from 47% to 41%), however plate waste of vegetables in comparison schools decreased significantly more (18%, from 59% to 41%; p <0.001).Discussion:Preliminary results suggest that the intervention did not have a positive impact on reducing student plate waste during school lunch. Obj. 6e As only modest changes in participation and no improvements in plate waste were seen, we will not measure the intervention's impact on obesity. Obj. 7 As only modest changes were seen in school lunch participation and no improvements in plate waste were seen, we will not explore the financial sustainability of the intervention. Obj. 8 Activities:SFUSD staff have tracked how intervention components have been sustained after the intervention ended in spring 2018.Data collected:SFUSD reported on intervention sustainability in fall 2019.Results:TOC outreach efforts have been sustained and expanded to other school staff and parents. When staffing allows, SFUSD continues to operate mobile carts and is considering expanding mobile carts to 2 more schools. Many of the intervention schools have transitioned to using a different cart model that is more nimble. Vending machine maintenance issues have prevented many vending machines from operating and will likely not be expanded to other schools.Discussion:The research team and SFUSD created a toolkit outlining considerations for implementation of the intervention. This can be used by other school districts to replicate the intervention.

Publications

  • Type: Journal Articles Status: Published Year Published: 2020 Citation: Machado SS, Ritchie LD, Thompson HR, Madsen KA. The Impact of a Multi-Pronged Intervention on Students' Perceptions of School Lunch Quality and Convenience and Self-Reported Fruit and Vegetable Consumption. Int J Environ Res Public Health. 2020 Aug 18;17(16):5987. doi: 10.3390/ijerph17165987.
  • Type: Journal Articles Status: Published Year Published: 2020 Citation: Thompson HR, Ritchie L, Park E, Madsen KA, Gosliner W. Effect of Removing Chocolate Milk on Milk and Nutrient Intake Among Urban Secondary School Students. Prev Chronic Dis. 2020 Aug 27;17:E95. doi: 10.5888/pcd17.200033.
  • Type: Journal Articles Status: Submitted Year Published: 2020 Citation: Thompson HR, Gosliner W, Madsen K, Ritchie L. Impact of a multi-component school lunch intervention on secondary students school lunch plate waste in an urban school district.
  • Type: Journal Articles Status: Under Review Year Published: 2020 Citation: Machado S, Brewster A, Shapiro V, Ritchie L. Magee K. Madsen K. Implementation Leadership in School Nutrition: A Qualitative Study. Journal of Nutrition Education and Behavior. (under review)


Progress 03/15/19 to 03/14/20

Outputs
Target Audience:In 2014, we selected 12 middle and 12 high schools to participate in the study and assigned half (6 middle and 6 high schools) to intervention and half to the control condition. The intervention ended in Spring 2018. Thus, data was not collected during 2018/2019 school year. Changes/Problems:The funded partners have worked diligently to overcome a myriad of unanticipated challenges that delayed implementation of the intervention and its subsequent evaluation as described in the last progress report. As reported previously, the project team decided to postpone implementation of all intervention components to Fall 2016 due to significant delays with vendor contracts, staffing, and technology. The mobile carts were all launched in Spring 2017 after many challenges with identifying sufficient staffing and receiving approval from the Health Department. After working with the vending machine vendors for over a year to resolve multiple technical challenges, the vending machines were all launched in Fall 2017. Due to additional challenges with staffing and vendors, the teacher intervention was delayed to January 2017. Finally, the SmartMeal app was suspended due to insurmountable challenges with the vendor and the capacity at the district to provide the required ongoing support (estimated at 40 hours per month) to maintain the app. In addition to the technical and contractual challenges described below, staff turnover in SFUSD School Nutrition Services at the school and leadership level that posedchallenges to project continuity. In March 2016, the interim SNS director gave notice that she would be leaving the district. The district conducted a search for a replacement and the position was filled in January 2017. The new Director left her position in December 2017, after only 11 months of service. The district conducted another national search, this time with the assistance of an executive recruiting firm, and the position was Director position was filled in June 2018. Three different people have filled the Future Dining Experience Project Manager position since the inception of the grant which has been accompanied by typical delays as new staff get on-boarded. Additionally, while the district hired additional staff to support the project, funding was allocated only for a year, leaving a single person to manage the logistics, training, monitoring, and technical challenges of intervention. At the school level, staff retirements, attrition, and unanticipated medical leaves, have left several of the schools without sufficient capacity to staff the mobile carts and, in some cases, fill and manage the vending machines. Moreover, dining staff has required more support and technical assistance from central office staff during the launch than originally anticipated. This limited the district's ability to launch the intervention more quickly. Vending Machines All vending machines were installed in the 12 intervention schools in August 2016. After successfully installing all the machines, SFUSD experienced a significant glitch in their point-of-sale system that prevented the vending machines from connecting to the software. As a result, implementation was delayed by approximately a month while technicians worked around the clock to resolve the problem. Several machines were launched in Fall 2016 and were plagued with technical problems including hardware malfunctions, electrical surges causing machines to randomly restart, and unknown software errors that resulted in students inaccurately being denied meals. Due to the unreliability of the machines SFUSD temporarily removed the launched machines from service in November 2016 and began working with the vending machine distributor to resolve the various problems. Between November 2016 and May 2017 SNS worked in close coordination with the four vendors involved and SFSUD electricians and IT network specialists to pinpoint the cause of each problem and develop and test solutions. In January and February 2017 capacitor harnesses were installed to help divert electrical needs, preventing the machines from randomly restarting. During vendor technician visits in March and May, new hardware including pinpads and sliding doors were installed on several machines. And finally, in June and July 2017 new software installed on all machines went through rigorous testing and was found to eliminate the remaining random errors. All vending machines were launched within the first week of the 2017-2018 school year. The district created marketing materials and worked closely with students and staff at the intervention sites to build awareness of the vending machines. SNS also updated instructional language with dining staff and student input, including translating into multiple languages to make the vending machines more accessible for English as a Second Language (ESL) students. In addition, menu signage with photographs, menu item descriptions and allergen information were created and posted with corresponding menu labels on each of the meals to help students better identify the meals available each day. Ongoing maintenance and troubleshooting through the 2017-2018 school year continued to put a strain on SNS resources. In spring 2018 two machines were temporarily taken out of service due to long repair timelines and a lack of electrical knowledge among SNS staff. In the summer of 2018, additional SNS central office staff members were trained in basic operations of the vending machines including machine settings and basic troubleshooting procedures in an effort to expand knowledge so that response times to vending machine issues could decrease. SNS completed the intervention phase of the project at the end of the 2017-2018 school year. Vending machines continued to operate at all twelve sites in the 2018-2019 school year, however, machine hardware and electrical issues continue to arise and SNS continues to struggle with a lack of internal expertise and slow response times from vendors, resulting in long periods of time where students do not have access to meals from the vending machines. Mobile Carts After overcoming challenges with the Department of Public Health, staffing shortages, and identifying locations for the mobile carts at each of the schools, all twelve were launched by February 2017. Improvements to operations and service locations were made for the start of the 2017-2018 school year as all mobile carts relaunched during the first two days of school. Throughout the 2017-2018 school year staffing shortages continued to pose a significant challenge for the mobile carts and maintenance of equipment enduring heavy use and traveling long distances from the kitchen to the mobile cart location, began toposing a problem. During the 2018-2019 school year, any mobile cart that has sufficient staff to operate has remained open. Approximately 7-9 mobile carts are in operation. SNS is currently evaluating lunch participation, mobile cart operating costs, and staffing models, to determine the financial feasibility of continued operations in each location. In addition, SNS is exploring additional mobile cart models that could streamline operations and decrease maintenance costs over time. SmartMeal App After a series of challenges and significant delays with the vendor, SFUSD terminated its contract with Xtopoly for SmartMeal app development. It contracted instead with PrimeroEdge, the software firm that supports the district's point of sale system at its middle and high schools. Primero offers an app called SchoolCafe, which the district targeted to launch in August 2017. Significant delays in the development of the app and repeated malfunctioning resulted in a stripped down version of the app that ultimately wasn't able to engage with students and meet the needs of the district. Thus, after over 3 years of working to develop a SmartMeal app, SFUSD determined the intervention infeasible in December 2017. The time and resources needed to maintain SmartMeal technology app after the completion of the grant remained a concern for the district throughout the development process. Continuous software upgrades require continuous development and time and technological expertise not currently available to the district. What opportunities for training and professional development has the project provided?This study provides intensive training in nutrition-related research methodology for one doctoral student, two masters students and several undergraduate students in public health. Specifically, this project provides training in: research design, survey design, plate waste assessment methodology, study management, and training in ethical research. Significant efforts wereundertaken to plan robust professional development for the teachers in SFUSD. Recognizing the limited time available for teacher professional development, which is most often optional for teachers, SFUSD contracted with a cinematographer to develop an educational video about the school food program that teachers can access at their convenience. The video features SFUSD leadership, principals, teachers, food service workers and students and covers topics ranging from the changes resulting from the Healthy, Hunger-Free Kids Act to the additional points of sale for school meals, and can be broken into shorter segments to appeal to teachers' busy school schedules. SFUSD and UCB collaborated with UCANR to develop educational flyers that correspond to the video segments for distribution to teachers. Twelve flyers were produced and disseminated between August 2016 and May 2018. UCANR hired a Teacher Outreach Coordinator whose job was to visit the 12 intervention schools to support teachers in their efforts to promote the school meal program to students. Finally, each of the 12 intervention schools hosted events for teachers throughout the 2017-18 school year to provide updates about school meals and offer meal taste tests for teachers with the goal of reshaping their perceptions about the school food program. Each of these interventions was informed by teacher input gathered by SFUSD and UCB staff. How have the results been disseminated to communities of interest?As previously reported, the project partners hosted an inaugural Advisory Board (AB) meeting on November 13, 2015. AB members (see list below) received project updates via email in Spring and Fall 2016 and summer 2019. The research team convened the AB in December 2019 to share results and gather feedback about how to disseminate findings to influence the field. Further, the team invited policy experts to join the AB in discussing school meal policy implications. AB members include: Becky Lee, IDEO Crystal FitzSimons, Food Research & Action Center Deanna Hoelscher, School of Public Health, University of Texas Elaine Waxman, Urban Institute Katie Kettman, SPUR Kim Frinzell, Nutrition Services, California Department of Education Marisa Neelon, University of California Cooperative Extension Sondra Parmer, Alabama, Cooperative Extension System Sylvia Allegretto, Institute for Research on Labor & Employment, UC Berkeley Tia Shimada, California Food Policy Advocates Zoe Neuberger, Center on Budget and Policy Priorities What do you plan to do during the next reporting period to accomplish the goals?May 2020: Disseminate online toolkit September 2020: Submit manuscripts on 1) student survey outcomes; 2) teacher survey outcomes; 3) plate waste outcomes; and 4) lessons learned

Impacts
What was accomplished under these goals? Obj. 1 Activities: The SmartMeal app was determined infeasible in December 2017. Data collected:14interviews with SFUSD staffwere conducted to assess implementation barriers. Results: SFUSD did not have staff with information technology (IT) or app development expertise involved in early app planning. Discussion: Early inclusion of IT staff may have resulted in clearer, more accurate implementation plan. Obj.2 Activities: Partners developed and implemented aprofessional development program to educate SFUSD teachers and staff about school meals. Data collected: Video views and meal tasting attendance weretracked. Results: Presentations at all-staff meetings reached750 teachers and staff (non-unique).Over 5500 teachers and staff (non-unique) were engaged through taste tests. Discussion:A websiteis accessible district-wide to provide a continued avenue for sharing nutrition education resources. Obj. 3 Activities: CE and SFUSDhired a Teacher Outreach Coordinator (TOC). The TOC aligned work with SFUSD's Wellness Policyto ensure sustainability. CE and SFUSD developed a toolkit to share teacher outreach best practices with other districts. Data collected: The TOCtracked communication with teachers, administrators, and support staff and worked with SFUSD's School Health Programs to identify school-based Wellness Champions. Results: Relationships wereestablished with 43 teachers, 26 administrators, and 23 support staff. 38 staff were identified as Wellness Champions. Discussion:After the intervention ended, SFUSD hired the TOC full-time to continue related nutrition education outreach activities. Obj. 4 Activities: Due to suspension of the Smart Meals app, students did not receive nutrition education via an app. However, teacher education efforts providedteachers with ideas and strategies to incorporate nutrition education in the classroom. The TOC delivered nutrition education to 360 students. Data collected: Survey data was collected in 2016-2018. 11,805 student surveys were included in the final sample. Results: Among 8th grade respondents, there was a relative increase in perceptions that school that school lunch tastes good (0.19; 95% CI: 0.07, 0.31, p<0.01) in intervention versus comparison schools. No changes were seen in perceptions of fullness orhealthfullness of school lunch or lunch line length for any grade. Discussion: Results suggest that the intervention was modestly effective in improving perceptions of lunch quality in grade 8 only, and ineffective in improving perceptions of convenience in all grades. Obj. 5 Activities: Ateacher survey was administered in spring 2016-2018. Data collected: The final analytic sample included 556 teacher surveys from 2016 and 533 from 2018. Results: A relative increase in teacher-reported frequency of eating with students in the cafeteria was seen in intervention schools (difference-in-change 1.48; 95% CI: 0.42, 2.53, p<0.01). Students at intervention schools had a relative increase in agreement that adults at their schools encouraged them to eat school lunch (difference- in-change 0.10; 95% CI: 0.01, 0.18, p< 0.05). No between-group differences were seen in teachers' perceptions of school meals or participation in school meals. Discussion:The intervention appeared to have a small yet significant impact on teachers eating in the cafeteria with students and on teacher encouragement of school lunch (among those who received a high intervention dose). Obj. 6a Activities: School lunch participation data was collected in school years 14/15-17/18. Data collected: SFUSD provided daily school lunch participation data and monthly average daily attendance for the intervention and control schools in school years 2014/15 (baseline) - 2017/18 (follow-up). Results: From baseline to follow-up, lunch participation dropped 4.1% in intervention and 5.1% in comparison schools (difference-in-difference 1.0%, 95% CI 0.5, 1.4). Among students eligible for free or reduced-price meals, participation decreased 1.8% in intervention and 4.9% in comparison schools (difference-in-difference 3.1%, 95% CI: 2.5%, 3.7%), with a larger difference-in-difference seen in high schools (5.0%, 95% CI: 4.2%, 5.9%) than middle schools (1.8%, 95% CI: 0.8%, 2.6%). Discussion: While the intervention demonstrated a modest but significant relative increase in school lunch participation, the effect was not sufficient to halt large district-wide declines in participation during this study period. Obj. 6b Activities: Survey data was collected in 2016-2018.11,805 student surveys were included in the final sample. Data collected: Self-reported daily and lunchtime fruit and vegetable data were collected via student surveys (as described in Objective 4). Results: Significant relative increases in fruit (difference-in-change 6.4%; 95% CI: 0.5%, 12.4%, p<0.05) and vegetable consumption (7.1%; 95% CI: 1.5%, 12.6%, p<0.05) at lunch were seen in grade 10 only. In grade 9 only, a relative increase in daily fruit intake was seen in intervention schools versus comparison schools (difference-in-change 0.10; 95% CI: 0.03, 0.16, p<0.01). Discussion: We observed only modest changes in perceptions of school lunch and fruit and vegetable consumption that were not consistent across all grades, suggesting that additional efforts are needed to improve school-lunch uptake. Obj. 6c Activities: Survey data was collected in 2016-2018. Data collected: Self-reported vegetable variety data were collected via student surveys (as described in Objective 4). Results: No significant differences were seen in daily vegetable variety in both the overall sample and when restricting to those who typically eat school lunch. Discussion: The intervention did not appear to impact vegetable variety. Obj. 6d Activities: The research team conducted plate waste assessments in Spring 2016 and 2018. Data collected: At baseline, 3,158 students participated in lunch data collection across all 24 schools. At follow-up, 2,966 students participated in lunch data collection. Results: From baseline to follow-up, overall plate waste in intervention schools increased 2% (from 37% to 39%), while overall plate waste in comparison schools decreased 8% (from 46% to 38%; p <0.001).Plate waste of fruits in intervention schools increased 7% (from 49% to 56%), while plate waste of fruits in comparison schools decreased 4% (from 59% to 55%); p <0.001). Plate waste of vegetables in intervention schools decreased 6% (from 47% to 41%), however plate waste of vegetables in comparison schools decreased significantly more (18%, from 59% to 41%; p <0.001). Discussion: Preliminary results suggest that the intervention did not have a positive impact on reducing student plate waste during school lunch. Obj. 6e As only modest changes in participation and no improvements in plate waste were seen, we will not measure the intervention's impact on obesity. Obj. 7 As only modest changes were seen in school lunch participation and no improvements in plate waste were seen, we will not explore the financial sustainability of the intervention. Obj. 8 Activities: SFUSD staff have tracked how intervention components have been sustained after the intervention ended in spring 2018. Data collected: SFUSD reported on intervention sustainability in fall 2019. Results: TOC outreach efforts have been sustained and expanded to other school staff and parents.When staffing allows,SFUSD continues to operate mobile carts and is considering expanding mobile carts to 2 more schools. Many of the intervention schools have transitioned to using a different cart model that is more nimble. Vending machine maintenance issues have prevented many vending machines from operating andwill likely not be expanded to other schools. Discussion: The researchers, CE and SFUSD created a toolkit outlining considerations for implementation of the intervention. This can be used by other school districts to replicate the intervention.

Publications

  • Type: Journal Articles Status: Published Year Published: 2019 Citation: Machado S, Ritchie L, Thompson H, et al. Multi-pronged intervention to increase secondary student participation in school lunch: Design and rationale. Contemporary Clinical Trials. 2019;78:133-139. doi:10.1016/j.cct.2019.01.013
  • Type: Journal Articles Status: Published Year Published: 2020 Citation: Thompson H, Gosliner W, Ritchie L, et al. The impact of a multi-pronged intervention to increase school lunch participation among secondary school students in an urban public school district. Childhood Obesity. Ahead of print.


Progress 03/15/18 to 03/14/19

Outputs
Target Audience:In 2014, we selected 12 middle and 12 high schools to participate in the study and assigned half (6 middle and 6 high schools) to intervention and half to the control condition. The following target audience was involved in data collection in school year 2017-18: School enrollment: 23,877 students. Race/ethnicity: 25% Latino, 43% Asian, 6% African American, 12% non-Hispanic white, and 14% other. Student survey participation: 2035 students in grade 7 and 8 and 4380 students in grades 9 and 10. Student lunchtime plate waste assessment participation: 2,966 students. Changes/Problems:As reported last year, the project team postponed implementation of all intervention components due to delays with vendor contracts, staffing, and technology. The mobile carts were all launched in Spring 2017,vending machines werelaunched in Fall 2017, and the teacher intervention was implemented inJanuary 2017. Finally, the SmartMeal app was suspended due to insurmountable challenges with the vendor and district capacityto provide the required ongoing support (estimated at 40 hours/month) to maintain the app.The following narrative summarizes the challenges to each intervention component encountered thus far and the efforts of the funded partners to launch the program with fidelity to maximize its impact. Vending Machines All vending machines were installed in the 12 intervention schools in August 2016. After successfully installing all the machines, SFUSD experienced a significant glitch in their point-of-sale system that prevented the vending machines from connecting to the software. As a result, implementation was delayed by approximately a month while technicians worked to resolve the problem. Several machines were launched in Fall 2016 and were plagued with technical problems including hardware malfunctions, electrical surges causing machines to randomly restart, and unknown software errors that resulted in studentsbeing denied meals. Due to the unreliability of the machines SFUSD temporarily removed the machines from service in November 2016 and began working with the vending machine distributor to resolve the various problems. Between November 2016 and May 2017 Student Nutrition Services (SNS) worked in close coordination with the four vendors involved (Betson Enterprises - Vending machine distributor, Crane Merchandising - Vending machine manufacturer, VendIT - Pinpad manufacturer, and PrimeroEdge - POS software company) and SFSUD electricians and IT network specialists to pinpoint the cause of each problem and develop and test solutions. In January and February 2017 capacitor harnesses were installed to help divert electrical needs, preventing the machines from randomly restarting. During vendor technician visits in March and May, new hardware including pinpads and sliding doors were installed on several machines. And finally, in June and July 2017 new software installed on all machines went through rigorous testing and was found to eliminate the remaining random errors. All vending machines were launched within the first week of the 2017-18 school year. After the first launch of the vending machines in Fall 2016, SFUSD realized the importance of a promotional campaign to ensure that students are aware of the availability of the meals during extended times throughout the school day. To that end, the district created marketing materials and worked closely with students and staff at the intervention sites to build awareness of the vending machines. SNS also updated instructional language with dining staff and student input, including translating into multiple languages to make the vending machines more accessible for English as a Second Languagestudents. In addition, menu signage with photographs, menu item descriptions and allergen information were created and posted with corresponding menu labels on each of the meals to help students better identify the meals available each day. Ongoing maintenance and troubleshooting through the 2017-18 school year continued to put a strain on SNS resources. In spring 2018 two machines were temporarily taken out of service due to long repair timelines and a lack of electrical knowledge among SNS staff. In the summer of 2018, additional SNS central office staff members were trained in basic operations of the vending machines including machine settings and basic troubleshooting procedures in an effort to expand knowledge so that response times to vending machine issues could decrease. SNS completed the intervention phase of the project at the end of the 2017-18 school year. Vending machines have continued to operate at all twelve sites in the 2018-19 school year, however, machine hardware and electrical issues continue to arise and SNS continues to struggle with a lack of internal expertise and slow response times from vendors, resulting in long periods of time where students do not have access to meals from the vending machines. Mobile Carts After overcoming challenges with the Department of Public Health, staffing shortages, and identifying locations for the mobile carts at each of the schools, all twelve were launched by February 2017. Improvements to operations and service locations were made for the start of the 2017-18 school year as all mobile carts relaunched during the first two days of school. Throughout the 2017-18 school year staffing shortages continued to pose a significant challenge for the mobile carts and maintenance of equipment enduring heavy use and traveling long distances from the kitchen to the mobile cart location, began toposing a problem. During the 2018-19 school year, any mobile cart that has sufficient staff to operate has remained open. Approximately 7-9 mobile carts are in operation. SNS is currently evaluating lunch participation, mobile cart operating costs, and staffing models, to determine the financial feasibility of continued operations in each location. In addition, SNS is exploring additional mobile cart models that could streamline operations and decrease maintenance costs over time. SmartMeal App After a series of challenges and significant delays with the vendor, SFUSD terminated its contract with Xtopoly for SmartMeal app development. It contracted instead with PrimeroEdge, the software firm that supports the district's point of sale system at its middle and high schools. Primero offers an app called SchoolCafe, which the district targeted to launch in August 2017. Significant delays in the development of the app and repeated malfunctioning resulted in a stripped down version of the app that ultimately wasn't able to engage with students and meet the needs of the district. Thus, after over 3 years of working to develop a SmartMeal app, SFUSD determined the intervention infeasible in December 2017. The time and resources needed to maintain SmartMeal technology app after the completion of the grant remained a concern for the district throughout the development process. Continuous software upgrades require continuous development and time and technological expertise not currently available to the district. Staffing Staff turnover in SFUSD SNS at the school and leadership level that has continued to pose real challenges to project continuity. In March 2016, the interim SNS director gave notice that she would be leaving the district. The district conducted a national search for a replacement and the position was filled in January 2017. The new Director left her position in December 2017, after only 11 months of service. The district conducted another national search, this time with the assistance of an executive recruiting firm, and the position was Director position was filled in June 2018. Three different people have filled the Future Dining Experience Project Manager position since the inception of the grant which has been accompanied by typical delays as new staff get on-boarded. Additionally, while the district hired additional staff to support the project, funding was allocated only for a year, leaving a single person to manage the logistics, training, monitoring, and technical challenges of intervention. At the school level, staff retirements, attrition, and unanticipated medical leaves, have left several of the schools without sufficient capacity to staff the mobile carts and, in some cases, fill and manage the vending machines. Moreover, dining staff has required more support and technical assistance from central office staff during the launch than originally anticipated. This limited the district's ability to launch the intervention more quickly. What opportunities for training and professional development has the project provided?This study provides intensive training in nutrition-related research methodology for one doctoral student, two masters students and several undergraduate students in public health. Specifically, this project provides training in: research design, survey design, plate waste assessment methodology, study management, and training in ethical research. Significant efforts have been undertaken to plan robust professional development for the teachers in SFUSD. Recognizing the limited time available for teacher professional development, which is most often optional for teachers, SFUSD contracted with a cinematographer to develop an educational video about the school food program that teachers can access at their convenience. The video features SFUSD leadership, principals, teachers, food service workers and students and covers topics ranging from the changes resulting from the Healthy, Hunger-Free Kids Act to the additional points of sale for school meals, and can be broken into shorter segments to appeal to teachers' busy school schedules. SFUSD and UCB collaborated with CE to develop educational flyers that correspond to the video segments for distribution to teachers. Nine flyers were produced and disseminated between August 2016 and May 2018. CE hired a Teacher Outreach Coordinator (TOC) whose job was to visit the 12 intervention schools to support teachers in their efforts to promote the school meal program to students. Finally, each of the 12 intervention schools hosted events for teachers throughout the 2017-18 school year to provide updates about school meals and offer meal taste tests for teachers with the goal of reshaping their perceptions about the school food program. Each of these interventions was informed by teacher input gathered by SFUSD and UCB staff. Finally, at the end of the project, SFUSD created two new full-time roles to support nutrition education: one of which has been filled by the staff member previously in the TOC role. Post-intervention, the grant coordinator for this project has also been hired into a full-time SNS management role atSFUSD. How have the results been disseminated to communities of interest?As previously reported, the project partners hosted an inaugural Advisory Board (AB) meeting on November 13, 2015. AB members (see list below) received project updates via email in Spring and Fall 2016. The research team will convene the AB in Spring 2019 to share results and gather feedback about how to disseminate findings to influence the field.Finally, the Co-Project Director presented at a poster session at the AFRI conference in August 2017 and the Society for Nutrition Education and Behavior conference in July 2018. The UCB team submitted a manuscript describing the intervention rationale and baseline data toContemporary Clinical Trialsin October 2018. Name Title Affiliation May C. Wang, MA, MPH, DrPH Associate Professor Department of Community Health Sciences UCLA Fielding School of Public Health Deanna M. Hoelscher, PhD, RD, LD John P. McGovern Professor in Health Promotion Director, Michael & Susan Dell Center for Healthy Living The University of Texas Health Science Center School of Public Health Carol Chase Huegli Associate Director Nutrition Services at the California Department of Education Pat Crawford Senior Director of Research and CE Specialist Nutrition Policy Institute Wendi Gosliner Nutrition Education and Obesity Prevention Research and Evaluation Unit Nutrition Policy Institute Sondra Parmer, PhD Program Coordinator, Nutrition Education Program Alabama Cooperative Extension Virginia Bolshakova, PhD County Director and Elkus Ranch Director UC Cooperative Extension San Mateo/San Francisco Counties Coe Stafford Design Director IDEO Paula Jones Director of Food Systems San Francisco Department of Public Health Fred Espinosa Food Service Manager San Diego Unified School District Kim Coates Director of Office of School Health Programs San Francisco Unified School District What do you plan to do during the next reporting period to accomplish the goals?The Project Timeline below reflects key activities for the next reporting period. Year 5 Project Timeline Date Activity Owner Deliverable/Milestones April Advisory Board Meeting UCB Minutes from meeting to inform relevant means of communicating results. Apr/May Complete online toolkit UCB/ SFUSD Communicate best practices and lessons learned for implementation of each intervention component Dec Complete papers on 1) meal participation; 2) student survey; 3) teacher outreach; 4) plate waste; and 5) lessons learned UCB Communicate study results.

Impacts
What was accomplished under these goals? Objective 1: 1) Key activities: After 3 years of working to develop a SmartMeal app, that component of the intervention was determined infeasible in December 2017. 2) Data collected: In summer 2018, 13 in-depth interviews with key project personnel were conducted to assess barriers to implementation of the app. 3) Summary statistics/discussion: Nothing to report. 4) Outcomes/accomplishments: Data from the staff interviews will be used to write a lessons learned paper. Objective 2: 1) Key activities: The funded partners developed and implemented a professional development program to educate SFUSD teachers and staff about school meals. 2) Data collected: Tracked attendance at all-staff meetings where Teacher Outreach Coordinator (TOC) presented information about school meal programs and provided tastings to staff at all 12 intervention schools. Sent 11-minute educational video directly to administrators for distribution to all teachers and staff. Viewing was tracked for individual and group settings. 3) Summary statistics/discussion: During the 2017-18 school year, presentations at all-staff meetings reached a total of 750 teachers and staff.Over 5500 teachers and staff were engaged through school meal taste. Nine educational flyers were developed and distributed to teachers at all 12 intervention schools. As of 2018, the education video had 206 plays:3 of which werestaff-wide screenings averaging 65 viewers, and 1 was screened school-wide reaching 150 teachers. 4) Outcomes/accomplishments: In order to sustain staff wellness activities after the intervention period, SFUSD restructured employee roles such that a manager embedded within SFUSD took over the TOC role when the TOC position contract ended in early October 2018. The manager is now responsible for bringing teacher outreach activities to scale district-wide. Objective 3: 1) Key activities: CE and SFUSD worked collaboratively to develop monthly flyers for teachers and to hire a TOC who provides ongoing school-based nutrition education in the 12 intervention schools. The TOC worked to align her outreach work with SFUSD's Wellness Policy section 9 for Staff Wellness to ensure sustainability. CE and SFUSD are working together to develop an online toolkit to share information on teacher outreach development and sustainability with other districts. 2) Data collected: The TOC contacted and tracked communication with teachers, administrators, and support staff and worked with SFUSD's School Health Programs to identify school-based Wellness Champions. 3) Summary statistics/discussion: Relationships have been established with 43 teachers, 26 administrators, and 23 support staff across the 12 intervention schools to provide ongoing nutrition education support. Thirty-eight school staff have been identified as Wellness Champions. 4) Outcomes/accomplishments: Established relationships by meeting with teachers, had a presence at school-based events, and visited classroom and meal times to support building of school food culture. Created a website that is now accessible district-wide to provide a continued avenue for sharing nutrition education resources: including CE specific links, a user-friendly display of Student Nutrition Services' content, and a set of activities for teachers to integrate school food topics into classroom engagement with students. After the intervention period ended, SFUSD hired the TOC as a full-time SFUSD employeeto continue related nutrition education work. Objective 4: 1) Key activities: Due to suspension of the Smart Meals e-application, students did not receive nutrition education. However, teacher education efforts are providing teachers with strategies to incorporate classroom nutrition education. The TOCdelivered nutrition education to 360 students in 11 classes, student club meetings, and school-wide workshops. 2) Data collected: In spring 2018, 6829 8th-10th grade students (64% of those enrolled) in the 24 study schools completed a survey. 3) Summary statistics/discussion: Compared to their peers at control schools, students at intervention schools had a significantly greater increase in their agreement that school meals make them feel full. 4) Outcomes/accomplishments: Nothing to report. Objective 5: 1) Key activities: Ateacher survey was administered in spring 2016 and 2018 to assess teacher perceptions and promotion school meals. 2) Data collected: In spring 2018, 689 teachers in the 24 study schools completed a survey, representing 60% of teachers. 3) Summary statistics/discussion: Data show that compared to their peers at control schools, students at intervention schools had a significantly greater increase in their agreement that adults at their schools encourage them to eat lunch. Further, compared to control schools, teachers in intervention schools had a significantly greater increase in their agreement that school meals taste good and a greater increase in odds of eating lunch in the cafeteria with students. 4) Outcomes/accomplishments: As mentioned in a previous section, a manager within SFUSD took over TOC responsibilities to see that outreach activities are scaled-up district-wide. Objective 6a: 1) Key activities: Baseline data collected in 2016. Data were collected again in spring 2017 and 2018. 2) Data collected: Data on where students eat lunch most frequently was obtained via6929 student surveys in Spring 2018. 3) Summary statistics/discussion: Average daily lunch participation in school year (SY) 2017-18, based on SFUSD records, was 7,065 students in middle schools and 9,789 students in high schools; 31% Latino, 8.0% African American, 46% Asian, 15% other. Preliminary analyses show that during SY 2017-18 intervention schools served a total of 21,065 meals from vending machines. Meals from vending machines accounted for 6.4% of all meals served in intervention schools. Initial analyses show an increase in the number of meals served on days when mobile carts were operational. Further analyses are ongoing. 4) Outcomes/accomplishments: There was no difference between groups in the proportion of students reporting eating the school lunch the day prior to the survey. Students in intervention schools had significantly higher odds of getting their meal from a mobile cart compared to students in control schools. Objective 6b: 1) Key activities: Survey data was collected in spring 2016, 2017, and 2018. 2) Data collected: Approximately 6900 student surveys were completedin spring 2018. 3) Summary statistics/discussion: In 2018, 66% of students reported eating green salad, vegetables, and/or fruit (not including fruit juice) at lunch the day before the survey. Among those eating school meals the proportion was significantly higher at 79%. 4) Outcomes/accomplishments: Students in intervention schools did not report increases in daily fruit or vegetable consumption relative to control students. Objective 6c: 1) Key activities: The research team conducted follow-up plate waste assessments in Spring 2018. 2) Data collected: In the 2018 plate waste assessment 2966 students participated. 3) Summary statistics/discussion: Results will be forthcoming after 2018 plate waste data is analyzed. 4) Outcomes/accomplishments: Nothing to report. Objective 6d: Same as for objective 6c. Objective 6e: Nothing to report. Objective 7: 1) Key activities: SFUSD istracking intervention costs, including maintenance and labor. 2) Data collected: Nothing to report. 3) Summary statistics/discussion: Nothing to report. 4) Outcomes/accomplishments: Nothing to report. Objective 8: 1) Key activities: In 2018, 24 interviews with dining staff and 13 interviews with administrators were conducted to assess perceptions of the intervention and implementation barriers/facilitators. 2) Data collected: Interviews were conducted with 37staff. 3) Summary statistics/discussion: Interview data will be used to write alessons learned manuscript. 4) Outcomes/accomplishments: Nothing to report.

Publications

  • Type: Journal Articles Status: Under Review Year Published: 2018 Citation: Machado S. Ritchie L. Thompson H. Reed H. Ibarra-Castro A. Neelon M. Madsen K. Multi-Pronged Intervention to Increase Secondary Student Participation in School Lunch: Design and Rationale. Contemporary Clinical Trials.
  • Type: Conference Papers and Presentations Status: Accepted Year Published: 2018 Citation: Ritchie L. School Meals in the 21st Century: Using Technology & Design Innovation. Society for Nutrition Education and Behavior Annual Conference. July 2018.


Progress 03/15/17 to 03/14/18

Outputs
Target Audience:In 2014, we selected 12 middle and 12 high schools to participate in the study and assigned half (6 middle and 6 high schools) to intervention and half to the control condition. The following target audience was involved in follow-up data collection in the 2016-17 school year: School enrollment: 6,183 students in grades 7-8; 7,135students in grades 9-10. Race/ethnicity: 25% Latino, 48% Asian, 7% African American, 10% non-Hispanic white, 9% other. Student survey participation (spring 2017): 4,209 students in grade 7 and 8 and 4,511 students in grades 9 and 10 Changes/Problems:The funded partners have worked diligently in the last year to overcome the a myriad of unanticipated challenges (summarized below) that delayed implementation of the intervention and its subsequent evaluation as described in the last progress report. These challenges are similar to those faced by many large, urban school districts seeking to implement a project of this scope, and thus provide important insight for districts seeking to implement similar interventions. As reported last year, the project team decided to postpone implementation of all intervention components to Fall 2016 due to significant delays with vendor contracts and technology. The mobile carts were all launched in Spring 2017 after many challenges with identifying sufficient staffing and receiving approval from the Health Department. After working with the vending machine vendors for over a year to resolve multiple technical challenges, the vending machines were all launched in Fall 2017. Due to additional challenges with staffing and vendors (particularly getting the finalized teacher video), the teacher intervention was delayed to January 2017. Finally, the SmartMeal app was suspended due to insurmountable challenges with the vendor and the capacity at the district to provide the required ongoing support (estimated at 40 hours per month) to maintain the app. The following narrative summarizes the challenges to each intervention component encountered thus far, the programmatic and evaluation changes those challenges have precipitated, and the efforts of the funded partners to launch the program with fidelity to maximize its impact. In addition to the technical and contractual challenges described below, staff turnover in SFUSD School Nutrition Services at the school and leadership level that has posed real challenges to project continuity. In March 2016, the interim SNS director gave notice that she would be leaving the district. The district has conducted a national search for a replacement and the position has been revised and posted numerous times. The position was filled in January 2017and the new Director left her position in December 2017, after only 11 months of service. Three different people have filled the Future Dining Experience Project Manager position since the inception of the grant which has been accompanied by typical delays as new staff get on-boarded. Additionally, while the district hired additional staff to support the project, funding was allocated only for a year, leaving a single person to manage the logistics, training, monitoring, and technical challenges of intervention. At the school level, staff retirements, attrition, and unanticipated medical leaves, have left several of the schools without sufficient capacity to staff the mobile carts and, in some cases, fill and manage the vending machines. Moreover, dining staff has required more support and technical assistance from central office staff during the launch than originally anticipated. This limited the district's ability to launch the intervention more quickly. Vending Machines All vending machines were installed in the 12 intervention schools in August 2016. After successfully installing all the machines, SFUSD experienced a significant glitch in their point-of-sale system that prevented the vending machines from connecting to the software. As a result, implementation was delayed by approximately a month while technicians worked around the clock to resolve the problem. Several machines were launched in Fall 2016 and were plagued with technical problems including hardware malfunctions, electrical surges causing machines to randomly restart, and unknown software errors that resulted in students inaccurately being denied meals. Due to the unreliability of the machines SFUSD temporarily the launched machines out of service in November 2016 and began working with the vending machine distributor to resolve the various problems. Between November 2016 and May 2017 SNS worked in close coordination with the four vendors involved (Betson Enterprises - Vending machine distributor, Crane Merchandising - Vending machine manufacturer, VendIT - Pinpad manufacturer, and PrimeroEdge - POS software company) and SFSUD electricians and IT network specialists to pinpoint the cause of each problem and develop and test solutions. In January and Febuary 2017 capacitor harnesses were installed to help divert electrical needs, preventing the machines from randomly restarting. During vendor technician visits in March and May, new hardware including pinpads and sliding doors were installed on several machines. And finally, in June and July 2017 new software installed on all machines went through rigorous testing and was found to eliminate the remaining random errors. All vending machines were launched within the first week of the 2017-2018 school year. After the first launch of the vending machines in Fall 2016, SFUSD realized the importance of a promotional campaign to ensure that students are aware of the availability of the meals during extended times throughout the school day. To that end, the district created marketing materials along with updated instructional signage. In addition, menu signage with photographs, menu item descriptions and allergen information were created and posted with corresponding menu labels on each of the meals to help students better identify the meals available each day. Mobile Carts After overcoming challenges with the Department of Public Health, staffing shortages, and identifying locations for the mobile carts at each of the schools, all twelve were launched by February 2017. Improvements to operations and service locations were made for the start of the 2017-2018 school year as all mobile carts relaunched during the first two days of school. Staffing shortages continue to pose a significant challenge for the mobile carts and maintenance of equipment enduring heavy use and traveling long distances from the kitchen to the mobile cart location, has posed a recent problem. SmartMeal App After a series of challenges and significant delays with the vendor, SFUSD terminated its contract with Xtopoly for SmartMeal app development. It contracted instead with PrimeroEdge, the software firm that supports the district's point of sale system at its middle and high schools. Primero offers an app called SchoolCafe, which the district targeted to launch in August 2017. Significant delays in the development of the app and repeated malfunctioning, resulted in a striped down version of the app that ultimately wasn't able to engage with students and meet the needs of the district. Thus, after over 3 years of working to develop a SmartMeal app, SFUSD determined the intervention infeasible in December 2017. The time and resources needed to maintain SmartMeal technology app after the completion of the grant remained a concern for the district throughout the development process. Continuous software upgrades require continuous development and time and technological expertise not currently available to the district. What opportunities for training and professional development has the project provided?This study provides intensive training in nutrition-related research methodology for one doctoral student, one graduate student and several undergraduate students in public health. Specifically, this project provides training in: research design, survey design, plate waste assessment methodology, study management, and training in ethical research. Significant efforts have been undertaken to plan robust professional development for the teachers in SFUSD. Recognizing the limited time available for teacher professional development, which is most often optional for teachers, SFUSD contracted with a cinematographer to develop an educational video about the school food program that teachers can access at their convenience. The video features SFUSD leadership, principals, teachers, food service workers and students and covers topics ranging from the changes resulting from the Healthy, Hunger-Free Kids Act to the additional points of sale for school meals, and can be broken into shorter segments to appeal to teachers' busy school schedules. SFUSD and UCB collaborated with UCANR to develop educational flyers that correspond to the video segments for distribution to teachers. Ten flyers were produced and disseminated between August 2016 and May 2017. UCANR hired a Teacher Outreach Coordintor whose job is to visit the 12 intervention schools to support teachers in their efforts to promote the school meal program to students. Finally, each of the 12 intervention schools hosted an event for teachers in winter 2017 to provide updates about school meals and offer meal taste tests for teachers with the goal of reshaping their perceptions about the school food program. Each of these interventions was informed by teacher input gathered by SFUSD and UCB staff. How have the results been disseminated to communities of interest?As previously reported, the project partners hosted an inaugural Advisory Board (AB) meeting on November 13, 2015. AB members (see list below) received project updates via email in Spring and Fall 2016. The research team expects to convene the AB in Summer 2018 to share results and gather feedback about how to disseminate findings to influence the field. May C. Wang, UCLA Fielding School of Public Health Deanna M. Hoelscher, The University of Texas School of Public Health Carol Chase Huegli, The University of Texas School of Public Health Pat Crawford, Nutrition Policy Institute Wendi Gosliner, Nutrition Policy Institute David Just, Cornell Center for Behavioral Economics in Child Nutrition Programs Sondra Parmer, PhD, Alabama Cooperative Extension Virginia Bolshakova, PhD, UC Cooperative Extension Coe Stafford, IDEO Paula Jones, San Francisco Department of Public Health Fred Espinosa, San Diego Unified School District Finally, the Co-Project Director presented at a poster session at the AFRI conference in August 2017. What do you plan to do during the next reporting period to accomplish the goals?The Project Timeline below reflects the timeline for the next reporting period. Year 4 Project Timeline Date Activity Owner Deliverable/Milestones Apr/May In all schools: 1) administer student and staff surveys, 2) assess plate waste, 3) assess meal participation data UCB F/U Data on: 1) student F/V intake; staff perceptions, 2) plate waste, 3) meal participation, 4) BMI Objectives 4-6 met Apr/May In all schools: 1) administer student and surveys, 2) assess plate waste, 3) collect meal participation, BMI, and intervention cost data UCB 2-yr F/U data on: 1) student F/V intake and staff perceptions, 2) plate waste, 3) meal participation, and 4) BMI, 5) intervention costs. Objectives 4-8 met Jun Complete paper on Year 1 results UCB Year 1 paper submitted to journal Jul Annual Advisory Board call UCB Minutes from call that inform study

Impacts
What was accomplished under these goals? Impact Statement: This project seeks to support & evaluate efforts in the San Francisco Unified School District (SFUSD) to promote student participation in the school lunch program, improve students' dietary intake & reduce plate waste in schools. The results of this project are intended to impact knowledge among school food professionals, policymakers & researchers across the country about increasing participation in school lunch through distributed points of sale & engaging teachers & staff in promoting meals, as well as understanding barriers to adoption of school-meal apps. Goal 1 Obj 1: Finalize content for the SmartMeal e-app by June 2017 Major activities: After 3+ years working to develop an app, that component of the intervention was determined infeasible in December 2017 Data collected: Nothing to report Summary statistics: Nothing to report Key outcomes: Documenting key challenges to SmartMeal app implementation; intend to publish a "lessons learned" manuscript Obj 2: Finalize staff wellness curriculum by June 2017 Major activities: The funded partners have developed a professional development program to educate SFUSD teachers & staff about school meals Data collected: Attendance at meetings where Teacher Outreach Coordinator (TOC) presented & provided tastings at all 12 intervention schools. Views for password-protected school food video. Summary statistics: Presentations at all-staff meetings reached 750 teachers & staff & 206 unique video plays Key outcomes: The TOC started in October 2016 & visited each of the 12 intervention schools multiple times to provide information about school meals to teachers. SFUSD produced & distributed a video for teachers about SFUSD school food starting in April 2017. Monthly flyers developed & distributed starting in August 2016. Obj 3: By August 2015, formalize a partnership between San Francisco & CE Major activities: CE & SFUSD developed monthly flyers for teachers & hired a TOC who provides ongoing school-based nutrition education at the 12 intervention schools. TOC aligning efforts with SFUSD's Wellness Policy for sustainability. Data collected: Tracked communication with teachers, administrators, & support staff. Worked with SFUSD's School Health Programs to identify school Wellness Champions Summary statistics: Relationships have been established with 40 teachers, 25 administrators, & 20 support staff at the 12 intervention schools to provide ongoing nutrition ed support; 38 staff have been identified as Wellness Champions Key outcomes: Created website to provide ongoing for sharing nutrition education resources, including CE specific links & relevant content. Developed & shared activities for teachers to integrate school food topics into classroom instruction. Obj 4: Increase nutrition knowledge by MS & HS students Major activities: Due to suspension of the Smart Meals app, students will not receive nutrition ed via an app. Teacher outreach efforts include equipping them with strategies to deliver nutrition ed in the classroom. Data collected: In spring 2017, 8,720 7th-10th grade students in the 24 study schools completed a baseline survey, representing 65% of enrolled students Summary statistics: Compared to their peers at control schools, students at intervention schools had a significantly greater increase in their agreement that school meals are healthier than the food they bring from home Key outcomes: Nothing to report Obj 5: Increase proportion of school staff who encourage students to eat school lunch & model healthy eating Major activities: Teacher survey was administered to collect data regarding teacher perceptions of the school meal program. Annual student survey includes a question about teacher promotion of the school meals. Program to educate teachers & staff about the school meal program has been launched. Data collected: Teacher survey was conducted at all 24 schools to determine changes in attitudes about & perceptions of school meals Summary statistics: Compared to their peers at control schools, students at intervention schools were more likely to report that adults at their schools encourage them to eat lunch. Results from teacher survey are expected at the time of the next report. TOC has enrolled 47 new teachers for online school meal accounts, which allow them to participate in school meals. Key outcomes: Expanding staff access to affordable healthy meals to increase likelihood that they encourage students to try school meals Goals 1 & 2 Obj 6 a. Increase student participation in school lunch [Goals 1 & 2] Major activities: Data collected in 2016 & 2017 Data collected: Baseline school meal participation data was determined. Data on where students eat lunch most frequently was obtained via 8,720 student surveys in Spring 2017. Summary statistics: Average daily lunch participation in SY16-17 among the 24 schools, based on SFUSD records, was 5,878 students in MS & 9,228 students in HS; 31.23% Latino, 8.17% African American, 46.77% Asian, 13.83% other Key outcomes: No difference between groups in the proportion of students reporting eating the school lunch the day prior to the survey. However,students in intervention schools had significantly higher odds of getting their meal from a mobile cart compared to students in control schools (OR 3.1, p<0.01). b. Increase student consumption of fruits & vegetables at lunch & daily [Goal 1] Major activities: Baseline data collected in 2016; data collected again in spring 2017 Data collected: 8,720 student surveys establishing fruit & vegetable intake were completed in Spring 2017 Summary statistics: At baseline, 66% of students reported eating green salad, vegetables, &/or fruit (not including fruit juice) at lunch the day before the survey. Among those eating school meals, the proportion was significantly higher at 81%. Key outcomes: Students in intervention schools did not report increases in fruits or vegetable consumption relative to control students c. Increase the variety of vegetables students consume at lunch & daily [Goal 1] Major activities: The research team revised protocol & recruited/trained volunteers for plate waste data collection in 2016. This data will help determine changes in fruit & vegetable consumption during lunch. Data will be collected again in Spring 2018. Data collected: Baseline plate waste data was collected for each of the 24 schools in 2016 among 2,815 students. Summary statistics: Results forthcoming after plate waste data is collected in 2018 Key outcomes: Nothing to report d. Decrease plate waste at school lunch [Goal 2] Major activities: The research team revised protocol & recruited/trained volunteers for plate waste data collection, which was executed at the 24 study schools in 2016 & will be conducted again in 2018 Data collected: Baseline plate waste data was collected for each of the 24 schools in 2016 among 2,815 students Summary statistics: Results forthcoming after plate waste data is collected in 2018 Key outcomes: Nothing to report e. Reduce obesity [Goal 1] Nothing to report Obj 7: After two years, determine the intervention's financial sustainability [Goal 2] Major activities: Due to implementation delays, the financial sustainability of the intervention cannot yet be evaluated. However, SFUSD is actively tracking intervention costs, including labor & food waste. Data collected: Nothing to report Summary statistics: Nothing to report Key outcomes: Nothing to report Obj 8: Over one year, assess the replicability & sustainability of the intervention [Goal 2] Major activities: Due to implementation delays, the financial sustainability of the intervention will be evaluated in 2018-2019 Data collected: Nothing to report Summary statistics: Nothing to report Key outcomes: Nothing to report

Publications


    Progress 03/15/16 to 03/14/17

    Outputs
    Target Audience:In 2014, we selected 12 middle and 12 high schools to participate in the study and assigned half (6 middle and 6 high schools) to intervention and half to the control condition. The following target audience was involved in baseline data collection in school year 2015-16: School enrollment: 6,195 students in grades 7-8; 7,182 students in grades 9-10. Race/ethnicity: 25% Latino, 48% Asian, 7% African American, 10% non-Hispanic white, 9% other. Student survey participation: 4,464 students in grade 7 and 8 and 4,593 students in grades 9 and 10; Student lunchtime plate waste assessment participation: 2,815 students. Changes/Problems:The funded partners have worked diligently in the last year to overcome the several unanticipated challenges (summarized below) that delayed implementation of the intervention and its subsequent evaluation as described in the last progress report. These challenges are similar to those faced by many large, urban school districts seeking to implement a project of this scope, and thus provide important insight for districts seeking to implement similar interventions. As reported last year, the project team decided to postpone implementation of all intervention components to Fall 2016 due to significant delays with vendor contracts and technology. The mobile carts and vending machines are on-track for implementation. However, due to additional challenges with staffing and vendors, the SmartMeal app and teacher intervention have been delayed to January 2017. Importantly, we do not anticipate that the delays will limit our ability to achieve our stated goals. The following narrative summarizes the challenges to each intervention component encountered thus far, the programmatic and evaluation changes those challenges have precipitated, and the efforts of the funded partners to launch the program with fidelity to maximize its impact. In addition to the technical and contractual challenges described below, staff turnover in SFUSD School Nutrition Services at the school and leadership level that has posed real challenges to project continuity. In March 2016, the interim SNS director gave notice that she would be leaving the district. The district has conducted a national search for a replacement and the position has been revised and posted numerous times. However, as of October 2016, the position has not been filled. The Future Dining Experience Project Manager who was overseeing the implementation of the funded interventions subsequently left her position in August 2016 to participate in a Civic Innovation Fellowship at Stanford's d-school. The Project Coordinator, who has been responsible for the day-to-day logistics of project implementation, has been promoted to the Project Manager position. While SFUSD has taken great lengths to prioritize project implementation, the turnover in leadership has posed a significant challenge in capacity at the district. All vacant positions, including the SNS Director and the Project Coordinator, are posted and the district hopes to have them filled before the new calendar year. At the school level, staff retirements, attrition, and unanticipated medical leaves, have left several of the schools without sufficient capacity to fill and manage the vending machines. Moreover, dining staff has required more support and technical assistance from central office staff during the launch than originally anticipated. This has limited the district's ability to launch machines more quickly. Vending Machines All vending machines were installed in the 12 invention schools in August 2016. Wiring and electrical were all functional at the time of installation. After successfully installing all the machines, SFUSD experienced a significant glitch in their point-of-sale system that prevented the vending machines from connecting to the software. As a result, implementation was delayed by approximately a month while technicians worked around the clock to resolve the problem. Additionally, several of the machines continue to have technical problems that have hindered their launch. For example, the pin pad on four of the machines is stuck and thus unable to vend or account for reimbursable meals, rendering it usable until the vendor can repair it. Additional software issues with the vending machines reporting to the server, among others, have further added unforeseen delays. Despite these setbacks, SFUSD expects that all vending machines will be launched by December. Since launching the first set of vending machines, SFUSD has realized the importance of a promotional campaign to ensure that students are aware of the availability of the meals during extended times throughout the school day. To that end, the district is currently working to develop a marketing strategy for the vending machines. Mobile Carts While originally SFUSD intended to produce the mobile cart prototype designed by IDEO in the project conceptualization phase, concerns about food safety from the Department of Public Health forced the district to pivot and purchase commercial carts. After considerable communication challenges with the vendor throughout Spring 2016, all mobile carts were delivered to SFUSD in July 2016. Many of the logistical issues, including ensuring food temperature, were resolved through the pilot conducted at Marshall High School and Everett Middle School during the last grant period. However, because the machines are larger and more unwieldy than originally anticipated, SFUSD has struggled to identify locations within each building where the mobile carts can be used and safely stored overnight outside the cafeteria. These issues have largely been resolved and the district is now launching one to two mobile carts each week. The staffing challenges mentioned above - both in terms of capacity and the support needed to launch the carts -have also posed significant challenges for the mobile carts. SmartMeal App After the significant contractual setbacks with selecting a vendor to design the SmartMeal app described in the last progress report, SFUSD signed a contract with Xtopoly. Though the district has worked diligently with the vendor since the contract was signed, the timeline for initial testing has been delayed to Spring 2017 with full implementation at the 12 intervention schools in Fall 2017. In the meantime, SFUSD is working to develop a plan to promote the tool to maximize its use once it becomes available to students and teachers. What opportunities for training and professional development has the project provided?This study provides intensive training in nutrition-related research methodology for one doctoral student, one graduate student and several undergraduate students in public health. Specifically, this project provides training in: research design, survey design, plate waste assessment methodology, study management, and training in ethical research. Significant efforts have been undertaken to plan professional development for the teachers in SFUSD. Recognizing the limited time available for teacher professional development, which is often optional for teachers, SFUSD contracted with a cinematographer to develop an educational video about the school food program that teachers can access at their convenience. The video features SFUSD leadership, principals, teachers, food service workers and students and covers topics ranging from the changes resulting from the Healthy, Hunger-Free Kids Act to the SmartMeal app being implemented at SFUSD, and can be broken into shorter segments to appeal to teachers' busy school schedules. SFUSD and UCB are collaborating with UCANR to develop flyers that correspond to the video segments for distribution to teachers. The first flyers were disseminated in August 2016, and each month thereafter a new flyer is released. UCANR hired a Teacher Outreach Coordintaor whose job is to visit the 12 intervention schools to support teachers in promoting school meals to students. Finally, each of the 12 intervention schools will host an event for teachers in winter 2017 to premiere the video and offer taste tests for teachers with the goal of reshaping their perceptions about the school food program. Each of these interventions has been informed by teacher input gathered by SFUSD and UCB staff. How have the results been disseminated to communities of interest?As previously reported, the project partners hosted an inaugural Advisory Board (AB) meeting on November 13, 2015. AB members (see list below) received project updates via email in Spring and Fall 2016. May C. Wang, UCLA Fielding School of Public Health Deanna M. Hoelscher, The University of Texas School of Public Health Carol Chase Huegli, The University of Texas School of Public Health Pat Crawford, Nutrition Policy Institute Wendi Gosliner, Nutrition Policy Institute David Just, Cornell Center for Behavioral Economics in Child Nutrition Programs Sondra Parmer, PhD, Alabama Cooperative Extension Virginia Bolshakova, PhD, UC Cooperative Extension Coe Stafford, IDEO Paula Jones, San Francisco Department of Public Health Fred Espinosa, San Diego Unified School District Additionally, a progress report detailing the outcomes of the project thus far was submitted to SFUSD in Spring 2016 at their request. Finally, the Co-Project Director presented at a poster session at the AFRI conference in August 2016. What do you plan to do during the next reporting period to accomplish the goals?We are on track to meet the majority of deliverables outlined in our last progress report, with the exception of developing and piloting the SmartMeal e-application, which was delayed due to significant school site staffing challenges at SFUSD and additional and unforeseen challenges in contracting with a vendor. The Project Timeline below reflects the intervention launch dates for the next reporting period. Year 3 Project Timeline Date Activity Owner Deliverable/Milestones Jan Finalize/pilot front-end of SmartMeal app SFUSD SmartMeal learning layer. Objective 1 met Jan/Feb Implement SmartMeal app and teacher intervention in 6 middle and 6 high schools SFUSD Schools receive front-endSmartMealapp, staff wellness components Mar/May In all schools: 1) administer student and staff surveys, 2) assess plate waste, 3) assess meal participation data UCB F/U Data on: 1) student F/V intake; staff perceptions, 2) plate waste, 3) meal participation, 4) BMI Objectives 4-6 met Apr/May In all schools: 1) administer student and surveys, 2) assess plate waste, 3) collect meal participation, BMI, and intervention cost data UCB 2-yr F/U data on: 1) student F/V intake and staff perceptions, 2) plate waste, 3) meal participation, and 4) BMI, 5) intervention costs. Objectives 4-8 met Jun Complete paper on Year 1 results UCB Year 1 paper submitted to journal Oct Annual Advisory Board call UCB Minutes from call that inform study

    Impacts
    What was accomplished under these goals? This project seeks to support and evaluate efforts in the San Francisco Unified School District (SFUSD) to promote student participation in the school lunch program, improve students' dietary intake and reduce plate waste in schools. In this second year of the project, we leveraged the strong collaboration we built with both SFUSD and Cooperative Extension (CE) to successfully pilot and launch most of the major intervention components. Specifically, two of the four project components - including mobile carts and vending machines were launched in Fall 2016; the SmartMeal e-application and teacher and staff professional development (PD) will begin in Winter 2017. The results of this project are expected to impact knowledge among school food professionals, policymakers and researchers across the country about increasing participation in school lunch through using a mobile app to promote school meals, offering additional meals at distributed points of sale, and engaging teachers and staff in promoting meals. Additionally, the research component of this project will evaluate the impact of the combined intervention (distributed points of sale, SmartMeal app, and staff PD) on school meal participation, dietary intake, and plate waste among students. Goal 1 Objective 1: Finalize the student-centered content for the SmartMeal e-application by June 2017. 1) Activities: SFUSD developed and published a Request for Proposals for an app developer, which was selected in late Spring 2016. The app is currently in development with vendor Xtopoly. A pilot is planned for winter 2017. 2) Data collected: Nothing to report 3) Summary statistics: Nothing to report 4) Key outcomes: Nothing to report Objective 2: Finalize a staff wellness curriculum that educates teachers and staff about school meals by June 2017. 1) Activities: The funded partners have developed a 3-part PD program to educate SFUSD teachers and staff about school meals. 2) Data collected: Nothing to report 3) Summary statistics: Nothing to report 4) Key outcomes: SFUSD contracted with a cinematographer who has finalized a video about SFUSD school food to disseminate to teachers starting in January 2017. Starting August 2016, monthly flyers are being developed and distributed. The CE Teacher Outreach Coordinator started in October 2016 and will educate teachers about school meals starting in January 2017. Objective 3: By August 2015, formalize a partnership between San Francisco and CE that supports continued nutrition education for students, staff, and teachers in SFUSD. 1) Activities: CE and SFUSD have worked collaboratively to develop monthly flyers for distribution to teachers and to hire a Teacher Outreach Coordinator who will support school-based nutrition education in the 12 intervention schools. 2) Data collected: Nothing to report 3) Summary statistics: Nothing to report 4) Key outcomes: Nothing to report Objective 4: Increase nutrition knowledge, with an emphasis on school meals, by middle and high school students. 1) Activities: SFUSD is working with vendor Xtopoly to develop the SmartMeal e-application. 2) Data collected: In spring 2016, 9,157 7th-10th grade students in the 24 study schools completed a baseline survey, representing 69% of enrolled students. 3) Summary statistics: A baseline survey of over 9,000 students in 2016 showed that 50% of students felt the school meal was less healthy than foods brought from home or purchased off-campus. 4) Key outcomes: Nothing to report Objective 5: Increase the proportion of school staff who encourage students to eat the school lunch and who model healthful eating behaviors. 1) Activities: Ateacher survey was administered in spring 2016 to collect baseline data regarding teacher perceptions of the school meal program. A student survey was distributed in spring 2016 that included a question regarding teacher promotion of the school meals. The student survey will be administered again in spring 2017 and 2018. A program to educate teachers and staff about school meals, described in detail in the "Training/Professional Development" section, is being launched. 2) Data collected: Surveys were distributed to teachers and staff at the 24 schools winter 2016 to establish baseline data about attitudes and perceptions of school meals. 3) Summary statistics: In 2016, 17% of students reported that school staff encouraged them to eat the school lunch. Among the 557 teachers/staff responding to the survey, 34% reported encouraging students to eat healthy foods at least weekly, and 20% reported encouraging students, at least weekly, to eat the school lunch (43% reported never encouraging their students to eat the school lunch). 4) Key outcomes: Nothing to report Goals 1 & 2: Objective 6 a. Increase student participation in the school lunch program [Goals 1 and 2]. 1) Activities: Baseline data collected in 2016. 2) Data collected: Baseline school meal participation data was determined. Data on where students eat lunch most frequently was obtained in spring 2016 via 9,157 student surveys. 3) Summary statistics: Average daily lunch participation in school year 2015-2016 among the 24 schools, based on Student Nutrition records, was 6,685 students in middle schools and 10,693 students in high schools; 29% Latino, 9% African American, 49% Asian, 13% other. 4) Key outcomes: Nothing to report b. Increase student consumption of fruits and vegetables at lunch and daily [Goal 1]. 1) Major activities: Baseline data were collected in 2016. 2) Data collected: Over 9,000 student surveys establishing fruit and vegetable intake were completed and returned in spring 2016. 3) Summary statistics: 66% of students reported eating green salad, vegetables, and/or fruit (not including juice) at lunch the day before the survey. Among those eating school meals, the proportion was higher at 81%. 4) Key outcomes: Nothing to report c. Increase the variety of vegetables that students consume at lunch and daily [Goal 1]. 1) Major activities: The research team revised protocol, recruited volunteers, and conducted volunteer training for plate waste data collection conducted at the study schools in March and April 2016. This data will help determine changes in fruit and vegetable consumption during lunch. 2) Data collected: Baseline plate waste data was collected among 2,815 students at the 24 schools in March and April 2016. 3) Summary statistics: The research team expects to have results available for the next progress report. 4) Key outcomes: Nothing to report d. Decrease plate waste at school lunch [Goal 2]. 1) Activities: The research team revised protocol, recruited volunteers, and conducted volunteer training for plate waste data collection, which was executed at the 24 study schools in March and April 2016. 2) Data collected: Baseline plate waste data was collected among2,815 students at the 24 schools in March and April 2016. 3) Summary statistics: The research team expects to have results available for the next progress report. 4) Key outcomes: Nothing to report e. Reduce obesity [Goal 1]. Nothing to report Objective 7: After two years, determine the intervention's financial sustainability [Goal 2]. 1) Activities: Due to the recent launch of the intervention, its financial sustainability cannot yet be evaluated. However, SFUSD is actively tracking associated costs, including labor and food waste. 2) Data collected: Nothing to report 3) Summary statistics: Nothing to report 4) Key outcomes: Nothing to report Objective 8: Over one year, assess the replicability and sustainability of the intervention [Goal 2]. 1) Activities: Due to the recent launch of the intervention, its financial sustainability cannot yet be evaluated. However, SFUSD is actively tracking associated costs, including labor and food waste. 2) Data collected: Nothing to report 3) Summary statistics: Nothing to report 4) Key outcomes: Nothing to report

    Publications

    • Type: Journal Articles Status: Under Review Year Published: 2017 Citation: Assessing plate waste in schools: aggregate weighing versus quartering methods. Martinez S, Ritchie L, Thompson HR, Madsen KA.


    Progress 03/15/15 to 03/14/16

    Outputs
    Target Audience:We recruited 12 middle and 12 high schools to participate in the study and assigned half (6 middle and 6 high schools) to intervention and half to the control condition. The following target audience was involved in baseline data collection: School meal participation data: 7,442 students in middle schools and 9,860 students in high schools; 23% Latino, 7% African American, 50% Asian, 20% other; Student survey participation: 3,844 students in grade 7 and 2,470 students in grades 9 and 10; and Student lunchtime plate waste assessment participation: 3,327students. Changes/Problems:Several unanticipated challenges (summarized below) have delayed implementation of the intervention and its evaluation. These challenges are similar to those faced by many large, urban school districts seeking to implement a project of this scope, and thus provide important insight for districts seeking to implement similar interventions. Due to concerns about maximizing the impact of the full intervention, rather than launch the components of the intervention individually as they are ready, the project team made the decision to delay implementation of all components, including the SmartMeal app, mobile carts, vending machines, and teacher professional development, to fall 2016. Accordingly, the research team will re-collect plate waste data at the 24 schools - 12 intervention and 12 control schools - in spring 2016 to ensure that baseline data corresponds as closely as possible to the intervention launch. This delay will give the team the opportunity to re-administer the student surveys as well as collect feedback from teachers via surveys, which was not originally planned but will further strengthen the evaluation. Importantly, we do not anticipate that the delays will limit our ability to achieve our stated goals. The following narrative summarizes the challenges to each intervention component encountered thus far, the programmatic and evaluation changes those challenges have precipitated, and the efforts of the funded partners to launch the program with fidelity to maximize its impact. SmartMeal App In March 2015, SFUSD received funding for the development of a SmartMeal app to provide nutrition education and promote school meals to students, with back-end analytics to enable SFUSD to get 'smarter' about meal purchasing. The original search for a developer for the app started in April 2015 with a Request for Proposals (RFP) that was released by the Future Dining Experience (FDE), a separate nonprofit. The process lasted 7.5 months and five firms submitted proposals. A new SFUSD Chief Technology Officer was hired during the summer of 2015 and questions regarding what demands the app would put on the SFUSD IT department delayed vendor selection. In addition, the Student Nutrition Services IT specialist left at the start of the 2015-2016 school year, which limited the amount of technical information the department had to develop the app. In November 2015, the SFUSD and FDE teams interviewed two firms to develop the app and selected YMedia Labs. After selecting YMedia, the team consulted the SFUSD legal department to draft a contract between SFUSD, FDE and YMedia Labs to enter into an agreement that would allow for SFUSD to be a third party beneficiary of the app. The contract was needed because the funds to develop the app were coming from the separate FDE project budget but the intention was to ensure the school district, not FDE, owned the app in perpetuity. Upon reviewing the agreement, SFUSD's legal department determined that the services for app development were subject to a bidding process that includes Board approval (both of the RFP and the selected vendor). The district worked with purchasing to streamline the bidding process in order to expedite the timeline. SFUSD went out to bid for the project on January 19th, a vendor will be selected by February 22nd, and the contract will go to the School Board for approval on March 8th, at which point SFUSD will begin development of the app. SFUSD will have a Beta version for testing in June 2016, a requirement that is built into the scope of services for the vendor contract. The district will engage in a publicity campaign in summer 2016 to ensure that students, teachers and parents in the 12 intervention schools are aware of the app in order to maximize its use when it launches in September 2016. Vending Machines SFSUD currently owns four vending machines that are each housed at intervention schools. The machines were in use during the 2014-2015 school year until three of the four began to have technical problems caused by the software and physical infrastructure. Problems with the original machines, which are property of Horizon (SFUSD's current software company), were noted by SFSUD starting in March 2015. SFUSD worked with Horizon to fix the machines with the expectation that repairs would be complete by October 2015. This would have enabled SFUSD to purchase machines for the remaining intevention schools from Horizon. Because of technical issues that could not be resolved by the vendor, SFUSD sought other vendors for non-proprietary machines, but the cost of the machines was greater than the bid threshold that is in place by the district, and SFUSD had to go out to bid for the machines. The bid process began in December 2015 and was completed in January 2016. A new vendor has been selected. The new vendor requires that the point-of-sale (POS) software that will be used is identified before it begins to build the machines. Unfortunately, the district's current POS provider recently determined that their software is not compatible with the district's new vending machines. As a result, SFUSD must identify a different POS software for the machines. Once identified, the new machines will take three weeks to build and ship. The team anticipates that the new machines will be delivered to the schools in March 2016. Delays were also experienced with the installation of wiring and electrical at the school sites required for the vending machines. Once approval was granted by the Fire Marshall, the SFUSD team with the SFUSD Director of Facilities to review the project. A number of machines and proposed locations needed to be reviewed and an outside contractor had to be hired in order to perform Hazmat reviews. The reviews are now complete and the project is now going out to an informal bid for the installation of the wiring and data. These upgrades are expected to be complete in Spring 2016. Mobile Carts Through an engagement with IDEO, a novel mobile meal cart design was created, fabricated, and piloted starting in October 2014. The pilot produced a great deal of information concerning the design of the cart, and the team began to work with a designer to develop the next iteration of the cart. During this time period, the SF Health Department stopped the pilot pending approval of paperwork, which was submitted in April 2015. The cart was rejected due to not having a mechanical cold holding unit. The previous method of using time and temperature as a control was not allowed for SFUSD in this instance. The Health Department approved SFUSD for a one-semester pilot (Fall 2015) of a cart that had a mechanical cold holding unit. Because the IDEO designed cart was not capable of being outfitted with a cold holding unit at a reasonable cost, the SFUSD team identified a less innovative commercial unit that is more conducive to food safety requirements. In July of 2015, three commercial carts were purchased to pilot. The carts were designed by the FDE team to be aligned with the style guidelines for the redesigned cafeterias. The carts arrived in September 2015 and one cart was tested at a high school during two meal periods in October 2015 to gauge the number of meals it could hold, temperature at which meals could be held and overall viability. After successful tests, nine additional carts were ordered on November 10, 2015 and, once food service staff is trained to use the POS software, will be ready for launch in Fall 2016. What opportunities for training and professional development has the project provided?This study provides intensive training in nutrition-related research methodology for one doctoral student and several undergraduate students in public health.Specifically, this project provides training in: research design, survey design, plate waste assessment methodology, study management, and training in ethical research. Significant efforts have been undertaken to plan robust professional development for the teachers in SFUSD. Recognizing the limited time available for teacher professional development, which is most often optional for teachers, SFUSD has contracted with a cinematographer to develop an educational video about the school food program that teachers can access at their convenience. The video will cover topics ranging from the changes resulting from the Healthy, Hunger-Free Kids Act to the SmartMeal app being implemented at SFUSD, and can be broken into shorter segments to appeal to teachers' busy school schedules. SFUSD and UCB are collaborating with UCANR to develop flyers that correspond to the video segments for distribution to teachers. UCANR is also in the process of hiring a nutrition educator who will visit the 12 intervention schools to support teachers in their efforts to promote the school meal program to students. Finally, each of the 12 intervention schools will host a school food party for teachers in Fall 2016 to premiere the video, distribute the flyers, and provide taste tests for teachers with the goal of reshaping their perceptions about the school food program. Each of these interventions will be informed by teacher input, which will be gathered by SFUSD staff in early 2016. How have the results been disseminated to communities of interest?The project partners hosted an inaugural Advisory Board (AB) meeting on November 13, 2015. AB members (see list below) were updated on the components of the intervention, the updated project timeline, and research methods and outcomes of interest. Additionally, members provided valuable advice on three key intervention questions: 1) best methods for promoting student use of the SmartMeal app, once it is developed; 2) techniques to ensure student use of the mobile carts and vending machines ; and 3) best practices regarding the teacher intervention. SFUSD and the research team have already begun using this feedback to strengthen multiple intervention components. May C. Wang, MA, MPH, DrPH: Associate Professor, Department of Community Health Sciences, UCLA Fielding School of Public Health Deanna M. Hoelscher, PhD, RN, LD: John P. McGovern Professor in Health Promotion, and Director, Michael & Susan Dell Center for Healthy Living, The University of Texas Health Science Center School of Public Health Carol Chase Huegli: Associate Director, Nutrition Services at the California Department of Education Pat Crawford: Senior Director of Research and CE Specialist, Nutrition Policy Institute Wendi Gosliner: Nutrition Education and Obesity Prevention Research and Evaluation Unit, Nutrition Policy Institute David Just: Professor, Co-Director, Cornell Center for Behavioral Economics in Child Nutrition Programs Sondra Parmer, PhD: Nutrition Education Programs Coordinator Alabama Cooperative Extension Virginia Bolshakova, PhD: County Director and Elkus Ranch Director, UC Cooperative Extension San Mateo/San Francisco Counties Coe Stafford: Design Director, IDEO Paula Jones: Director of Food Systems, San Francisco Department of Public Health Fred Espinosa: Food Service Manager, San Diego Unified School District Kim Coates: Director of Office of School Health Programs, San Francisco Unified School District What do you plan to do during the next reporting period to accomplish the goals?We are on track to meet deliverables outlined in our original proposal with the exception of the implementation of the intervention components caused by unforeseen delays in executing contracts with SFUSD vendors (see "Changes/Problems" section). To that end, the Project Timeline below reflects the adjusted intervention launch dates for the next reporting period. Date Activity Owner Deliverable/Milestones March Build / order equipment for RCT SFUSD Mobile carts and vending machines Mar/May In all schools: 1) administer student and staff surveys, 2) assess plate waste, 3) assess meal participation data UCB F/U Data on: 1) student F/V intake; staff perceptions, 2) plate waste, 3) meal participation, 4) BMI Objectives 4-6 met Mar/Jun Develop/pilot front-end of SmartMeal app CE & SFUSD SmartMeal learning layer. Objective 1 met Mar/Apr Refine intervention based on pilot All Optimal intervention configuration Jun Complete paper on baseline results UCB Baseline paper submitted to journal Aug Implement intervention in 6 middle and 6 high schools (including pilot schools) SFUSD Schools receive equipmt., front-end SmartMeal app, and staff wellness professional development

    Impacts
    What was accomplished under these goals? Impact Statement This project seeks to support and evaluate efforts in the San Francisco Unified School District (SFUSD) to promote student participation in the school lunch program, improve students' dietary intake and reduce plate waste in schools. In this first year of the project, we built a strong collaboration with both SFUSD and Cooperative Extension (CE). Due to unforeseen delays in project implementation (detailed in the "Changes/Problems" section), which are not uncommon when working in a complex, urban school district, the three project components - including a SmartMeal e-application, mobile carts and vending machines, and teacher and staff professional development - will be launched in Fall 2016. The results of this project are expected to impact knowledge among school food professionals, policymakers and researchers across the country about increasing participation in school lunch through using a mobile app to promote school meals, offering meals at distributed points of sale, and engaging teachers and staff in promoting meals. Additionally, the research component of this project will evaluate the impact of the combined intervention (distributed points of sale, SmartMeal app, and staff professional development) on school meal participation, dietary intake, and plate waste among students. Goal 1 Objective 1 1) Major activities:A focus group was conducted with twenty SFUSD students during summer 2015 about the school meal program. Students spoke about different reasons why they did not eat or did not like the school meals provided to them, including food packaging, flavor, and freshness. 2) Data collected: Focus groups were conducted at two high schools during the Spring 2015 semester to discuss a student-centered an e-application to promote healthier food choices. Students wanted an app that included short and simple messages, pictures, caloric information, food ingredients, and focused on taste. Students provided feedback on content for a SmartMeal app (for smartphones and tablets), including promotional nutritional messages for each meal item served at SFUSD. 3) Summary statistics/discussion: Nothing to report. 4) Key outcomes/accomplishments: Nothing to report. Objective 2 1) Major activities: As described in the "Significant Products/Outputs" section, the funded partners have developed a three-part professional development program to educate SFUSD teachers and staff about school meals. 2) Data collected: Nothing to report. 3) Summary statistics/discussion: Nothing to report. 4) Key outcomes/accomplishments Nothing to report. Objective 3 1) Major activities: CE met with SFUSD in November 2015 to discuss plans for collaboration, including the availability of CE nutrition educators to visit SFUSD schools to provide lessons to students, teachers and staff. The CE visits will launch in September 2016 at the 12 intervention schools. 2) Data collected: Nothing to report. 3) Summary statistics/discussion: Nothing to report. 4) Key outcomes/accomplishments: Nothing to report. Objective 4 1) Major activities: The SmartMeal app will help students understand the benefits of various foods and practical ways to make healthy eating decisions. SFUSD went out to bid on for a vendor to develop the app on January 19, 2016. The process will be completed on February 22nd, and will go to the School Board for approval on March 8th after which the funded partners will begin development of the app. 2) Data collected: In spring 2015, 10,000 paper surveys were distributed to 7th-10th grade students in the 24 study schools to collect baseline data establishing fruit and vegetable intake and school meal participation. Of the surveys distributed, 65% were completed and returned. Due to the implementation delays (see "Changes/Problems" section), baseline data will be collected again in Spring 2016 for a more robust baseline assessment that is closer in time to the intervention launch in Fall 2016. 3) Summary statistics/discussion: Baseline surveys will be readministered this spring given the implementation delays. Therefore, we have nothing to report yet. 4) Key outcomes/accomplishments: Nothing to report. Objective 5 1) Major activities: Plans are underway to develop a program to educate teachers and staff about the school meal program, described in detail in the "Training/Professional Development" section. Ateacher survey will be administered in spring 2016 to collect baseline data regarding teacher perceptions of the school meal program. Additionally, a baseline student survey was distributed spring 2015 that includes a question regarding teacher promotion of the school meals. The student survey will be administered again in spring 2016. 2) Data collected: Surveys were distributed to teachers and staff at the 24 schools in February 2016 to collect baseline data about their attitudes about and perceptions of school meals. Surveys will be collected in March 2016 and analyzed thereafter. The student surveys, which are being repeated in spring 2016, also include a question about teacher encouragement to eat school lunch. 3) Summary statistics/discussion: Nothing to report yet. 4) Key outcomes/accomplishments: Nothing to report yet. Goals 1 & 2: Objective 6 a. Increase student participation in the school lunch program [Goals 1 and 2]. 1) Major activities: Baseline data collected in 2015 and will be collected again in 2016. 2) Data collected: Over 6,000 student surveys establishing fruit and vegetable intake and school meal participation were completed and returned in Spring 2015. Due to the implementation delays, baseline data will be collected again in Spring 2016. 3) Summary statistics/discussion: As described above, baseline surveys will be repeated this spring given implementation delays. 4) Key outcomes/accomplishments: Nothing to report. b. Increase student consumption of fruits and vegetables at lunch and daily [Goal 1]. 1) Major activities: Baseline data collected in 2015 and will be collected again in 2016. 2) Data collected: Over 6,000 student surveys establishing fruit and vegetable intake and school meal participation were completed and returned in Spring 2015. Due to the implementation delays, baseline data will be collected again in Spring 2016. 3) Summary statistics/discussion: As described above, we will be repeating baseline surveys this spring given implementation delays. 4) Key outcomes/accomplishments: Nothing to report. c. Increase the variety of vegetables that students consume at lunch and daily [Goal 1]. 1) Major activities: Baseline data collected in 2015 and will be collected again in 2016. 2) Data collected: Over 6,000 student surveys establishing fruit and vegetable intake and school meal participation were completed and returned in Spring 2015. Due to the implementation delays baseline data will be collected again in Spring 2016. 3) Summary statistics/discussion: As described above, we will be repeating baseline surveys this spring given implementation delays. 4) Key outcomes/accomplishments: Nothing to report. d. Decrease plate waste at school lunch [Goal 2]. 1) Major activities: The research team developed protocol, recruited volunteers, and conducted volunteer training for plate waste data collection, which was executed at the 24 study schools in March and April 2015. Baseline surveys will be readministered this spring given implementation delays. 2) Data collected: Plate waste data was collected for each of the 24 schools in April and May 2015. 3) Summary statistics/discussion: Baseline data for student participation was collected in Spring 2015. Due to intervention implementation delays, meal participation baseline data will be collected again in Spring 2016. The research team expects to have results available for the next NIFA progress report. 4) Key outcomes/accomplishments: Nothing to report. e. Reduce obesity [Goal 1]. Nothing to report. Objective 7: Nothing to report. Objective 8: Nothing to report.

    Publications