Advertisement

Two-Year Healthy Eating Outcomes: An RCT in Afterschool Programs

      Introduction

      Across the U.S., afterschool programs (ASPs, 3:00pm–6:00pm) are trying to achieve nationally endorsed nutrition standards (Healthy Eating Standards) calling for fruits/vegetables and water to be served every day, while eliminating sugar-sweetened beverages and foods. The purpose of this study was to evaluate the 2-year changes in the types of foods and beverages served during a community-based intervention designed to achieve the Healthy Eating Standards.

      Study design

      Randomized delayed treatment trial with an immediate (1-year baseline and 2-year intervention) or delayed (2-year baseline and 1-year intervention) group.

      Setting/participants

      Twenty ASPs serving 1,700 children (aged 5–12 years) were recruited, with baseline occurring spring 2013, and outcome assessment occurring spring 2014 and 2015.

      Intervention

      The multistep intervention, Strategies To Enhance Practice for Healthy Eating, assisted ASP leaders/staff to serve foods/beverages that meet the nutrition standards.

      Main outcome measures

      The foods and beverages served for snack were observed directly.

      Results

      Compared with non-intervention years, both the immediate and delayed groups increased the number of days/week that fruits/vegetables (0.6 vs 1.7 days/week and 0.6 vs 4.4 days/week, OR=3.80, 95% CI=1.45, 9.95) and water (2.3 vs 3.7 days/week and 2.7 vs 4.8 days/week, OR=4.65, 95% CI=1.69, 12.79) were served. Sugar-sweetened beverages were almost eliminated by post-assessment (1.2 vs 0.2 days/week and 3.2 vs 0.0 days/week, OR=0.05, 95% CI=0.02, 0.13). Only the immediate group decreased the number of days/week desserts were served (2.9 vs 0.6 days/week, OR=0.10, 95% CI=0.03, 0.33). Implementation barriers for the delayed group included once/month delivery schedules for fruits/vegetables and limited storage space for foods meeting the Healthy Eating Standards.

      Conclusions

      Improvements in the foods/beverages served in ASPs can be made, yet were hindered by structural barriers related to procurement and storage of perishable foods. Additional efforts are needed to support ASPs as they work toward fully achieving the Healthy Eating Standards.

      Introduction

      Across the nation, afterschool programs (ASPs, 3:00pm–6:00pm) serve more than 10 million children annually every day of the school year.
      Afterschool Alliance
      America After 3 PM: A Household Survey on Afterschool in America.
      As part of their daily offering, most ASPs provide children with a food and beverage for snack, which represents an important opportunity to promote and nurture healthy eating habits of children.
      • Beets M.W.
      • Tilley F.
      • Kim Y.
      • et al.
      Nutritional policies and standards for snacks served in after-school programmes: a review.
      The consistent and repeated nature of these snacking occasions also provides an important opportunity for preference development and shaping.
      • Skinner J.D.
      • Carruth B.R.
      • Wendy B.
      • et al.
      Children’s food preferences: a longitudinal analysis.
      • Bertino M.
      • Beauchamp G.K.
      • Engelman K.
      Increasing dietary salt alters salt taste preference.
      • Ifland J.R.
      • Preuss H.G.
      • Marcus M.T.
      • et al.
      Refined food addiction: a classic substance use disorder.
      • Cooke L.
      The importance of exposure for healthy eating in childhood: a review.
      Unfortunately, the foods and beverages served rarely include fruits/vegetables, and often consist of processed foods high in added sugars and artificial flavors and colors.
      • Beets M.W.
      • Weaver R.G.
      • Tilley F.
      • et al.
      Salty or sweet? Nutritional quality, consumption, and cost of snacks served in afterschool programs.
      • Cassady D.
      • Vogt R.
      • Oto-Kent D.
      • et al.
      The power of policy: a case study of healthy eating among children.
      • Coleman K.J.
      • Geller K.S.
      • Rosenkranz R.R.
      • et al.
      Physical activity and healthy eating in the after-school environment.
      • Mozaffarian R.S.
      • Wiecha J.L.
      • Roth B.A.
      • et al.
      Impact of an organizational intervention designed to improve snack and beverage quality in YMCA after-school programs.
      National- and state-level organizations have developed and voluntarily adopted food and beverage standards for the types of snacks ASPs serve.
      • Beets M.W.
      • Tilley F.
      • Kim Y.
      • et al.
      Nutritional policies and standards for snacks served in after-school programmes: a review.
      One set of standards that can have a considerable public health impact on the dietary intake of children are the Healthy Eating (HE) Standards.

      National Afterschool Association. Standards for healthy eating and physical activity in out-of-school time programs. http://naaweb.org/images/NAA_HEPA_Standards_new_look_2015.pdf. Published 2011. Accessed December 2, 2011.

      The Standards, which can be viewed as organizational policies, call for programs to serve a fruit or vegetable every day and to eliminate foods and beverages with added sugars.
      Although improvements in the types of foods and beverages served for snack have been reported in previous studies,
      • Cassady D.
      • Vogt R.
      • Oto-Kent D.
      • et al.
      The power of policy: a case study of healthy eating among children.
      • Mozaffarian R.S.
      • Wiecha J.L.
      • Roth B.A.
      • et al.
      Impact of an organizational intervention designed to improve snack and beverage quality in YMCA after-school programs.
      • Giles C.M.
      • Kenney E.L.
      • Gotrtmaker S.L.
      • et al.
      Increasing water availability during afterschool snack: evidence, strategies, and partnerships from a group randomized trial.
      the majority of intervention studies have focused on changes in snacks over a single school year (i.e., fall to spring) or were conducted in a limited number of homogenous ASPs. Given the diversity of operations among ASPs and the amount of staff turnover (both at the site leader and frontline staff positions),
      • Beets M.W.
      • Weaver R.G.
      • Moore J.B.
      • et al.
      From policy to practice: strategies to meet physical activity standards in YMCA afterschool programs.
      • Hastmann T.J.
      • Bopp M.
      • Fallon E.A.
      • et al.
      Factors influencing the implementation of organized physical activity and fruit and vegetable snacks in the HOP’N after-school obesity prevention program.
      • Kelder S.
      • Hoelscher D.M.
      • Barroso C.S.
      • et al.
      The CATCH Kids Club: a pilot after-school study for improving elementary students’ nutrition and physical activity.
      which occurs across and within years, investigating changes in snacks over multiple years is important, as opposed to a single year. Investigations over multiple years can help identify if intervention effects are maintained across years within a setting that is likely to have different individuals in the positions that make decisions about the types of snacks served.
      This study reports the 2-year snack outcomes from the Making Healthy Eating and Physical Activity Policy Practice group randomized delayed treatment trial in 20 ASPs.
      • Beets M.W.
      • Glenn Weaver R.
      • Turner-McGrievy G.
      • et al.
      Making Healthy Eating and Physical Activity Policy Practice: the design and overview of a group randomized controlled trial in afterschool programs.
      • Beets M.W.
      • Weaver R.G.
      • Turner-McGrievy G.
      • et al.
      Making Healthy Eating Policy Practice: a group randomized controlled trial on changes in snack quality, costs, and consumption in after-school programs.
      The first-year snack outcomes (reported elsewhere)
      • Beets M.W.
      • Weaver R.G.
      • Turner-McGrievy G.
      • et al.
      Making Healthy Eating Policy Practice: a group randomized controlled trial on changes in snack quality, costs, and consumption in after-school programs.
      showed statistically significant improvements in the foods (e.g., fruits/vegetables) and beverages (e.g., sugar-sweetened beverages) served for snack in those ASPs receiving a multicomponent, capacity-building intervention. This study extends these findings by following the same ASPs an additional year and uses an immediate and delayed treatment group design that allowed for the investigation of the effect of the intervention on the types of foods and beverages served for snack over multiple years in the immediate group and the replication of the intervention within the delayed group. It was hypothesized that 2 years of receiving the intervention would result in greater gains than a single year of intervention and that the intervention effect would be replicated in the delayed treatment group.

      Methods

      A detailed description of the study design, intervention, measures, and first-year outcomes have been published previously.
      • Beets M.W.
      • Glenn Weaver R.
      • Turner-McGrievy G.
      • et al.
      Making Healthy Eating and Physical Activity Policy Practice: the design and overview of a group randomized controlled trial in afterschool programs.
      • Beets M.W.
      • Weaver R.G.
      • Turner-McGrievy G.
      • et al.
      Making Healthy Eating Policy Practice: a group randomized controlled trial on changes in snack quality, costs, and consumption in after-school programs.
      • Beets M.W.
      • Turner-McGrievy B.
      • Weaver R.G.
      • et al.
      Intervention leads to improvements in the nutrient profile of snacks served in afterschool programs: a group randomized controlled trial.
      ASPs, defined as child care programs operating immediately after the school day; every day of the school year for a minimum of 2 hours; serving a minimum of 30 children of elementary age (6–12 years); operated in a school, community, or faith setting; served a snack; and were located within a 1.5-hour drive from the university of the first author, were eligible to participate. Programs were identified from a registry of ASPs operating in the state and randomly selected for invitation to participate in the study. Programs were randomly selected from the list and called, with the first 20 that agreed to participate selected. Of the 535 programs, 376 met the eligibility criteria—76 did not operate Monday–Friday, 53 enrolled <30 children, and 28 did not have sufficient information to evaluate eligibility. No programs refused to participate, but were excluded based on the eligibility criteria stated above. The final sample of 20 afterschool programs represented 12 different organizations in South Carolina (Table 1). Across the 20 ASPs, mean enrollment during spring 2013 was 88 children (range, 30–162). Five programs received federal or state reimbursement for snacks. All study procedures were approved by the IRB at the University of South Carolina. Informed consent for the data collection procedures was obtained at the program level from each organization and program site leader.
      Table 1Spring 2013, 2014, and 2015 Demographic Characteristics of Programs and Children Enrolled by Treatment Condition
      CharacteristicsDelayed treatment group (n=10)Immediate treatment group (n=10)
      Spring 2013 (non-intervention year 1)Spring 2014 (non-intervention year 2)Spring 2015 (intervention year 1)Spring 2013 (non-intervention year 1)Spring 2014 (intervention year 1)Spring 2015 (intervention year 2)
      Afterschool program characteristics
       Total children enrolled870864760895763894
       Average enrollment (M±SD)87.0 ± 47.886.4 ± 39.576.0 ± 44.289.5 ± 52.076.3 ± 45.989.4 ± 59.3
       Daily program length (minutes per day, M±SD)205.5 ± 134.3NC190.5 ± 128.7NCNC
       Reimbursement for snack (n)1NCNC4NCNC
       Percentage of population in poverty, Census 2010, M±SD17.5% ± 10.2%NCNC13.3% ± 15.6%NCNC
       Location (n)
        School6NCNC3NCNC
        Faith/church1NCNC3NCNC
        Community (e.g., recreation center)3NCNC4NCNC
       Site leader and staff turnover
        Number of ASPs with site leader turnover3 Sp2014 versus Sp20135 Sp2015 versus Sp20141 Sp2014 versus Sp20132 Sp2015 versus Sp2014
        Number of ASPs with frontline staff turnover8988
        Median percent frontline staff turnover50%33%49%48%
      Child characteristics
       Boys (%)52.451.350.453.352.454.2
       Age (years, M±SD)8.1 ± 1.87.9 ± 1.97.8 ± 1.77.9 ± 1.87.9 ± 1.87.6 ± 1.7
       Race/ethnicity (%)
        White non-Hispanic48.446.855.264.659.066.9
        African American44.745.538.729.734.326.1
        Other6.97.76.15.76.77.0
      ASP, afterschool programs; NC, no change; Sp, Spring.

      Study Sample

      The design was a delayed treatment group RCT, which has been published previously, and is depicted in Figure 1.
      • Beets M.W.
      • Glenn Weaver R.
      • Turner-McGrievy G.
      • et al.
      Making Healthy Eating and Physical Activity Policy Practice: the design and overview of a group randomized controlled trial in afterschool programs.
      The 20 ASPs were randomized into one of two conditions: immediate or delayed group. To minimize contamination, ASPs within the same organization were matched with ASPs from other organizations and were all randomized to the same condition. Once an ASP from within an organization was randomized to the intervention or control group, all other ASPs from this organization were also designated to this group. The immediate treatment group received 2 years of the intervention during 2013–2014 and 2014–2015 school years. The delayed group was asked to continue with current ASP practices and received no technical assistance or support from the intervention staff during 2012–2013 and 2013–2014 school years until the final year of the study in 2014–2015 where they received the intervention.
      Figure 1
      Figure 1CONSORT diagram.
      ASP, afterschool programs; M-F, Monday–Friday.
      A detailed description of the Strategies To Enhance Practice for HE (STEPs-HE) has been previously published.
      • Beets M.W.
      • Glenn Weaver R.
      • Turner-McGrievy G.
      • et al.
      Making Healthy Eating and Physical Activity Policy Practice: the design and overview of a group randomized controlled trial in afterschool programs.
      In brief, the STEPs-HE conceptual framework involves a multistep, adaptive approach to incorporating healthy eating strategies into daily routine practice. The approach consists of identifying essential ASP characteristics that represent fundamental building blocks that function as necessary programmatic components to achieving full integration of healthy eating strategies and eventual achievement of the HE Standards. The Standards indicate that, on a daily basis, the ASP (1) serves a fruit or vegetable; (2) offers water at the table and has water accessible at all times; (3) serves no candy or other foods that are primarily sugar-based; (4) serves no beverages that are primarily sugar-based; and (5) avoids foods and beverages made with artificial ingredients (sweeteners, flavors, or colors).
      The STEPs-HE recognizes that each ASP is unique and, therefore, will require some similar and some different resources/strategies to achieve the HE Standards. The approach taken in STEPs-HE is one where some degree of local site–level tailoring will occur that is both responsive and adaptive to the characteristics of each ASP.
      • Hawe P.
      • Shiell A.
      • Riley T.
      Complex interventions: how “out of control” can a randomised controlled trial be?.
      STEPs-HE
      • Beets M.W.
      • Glenn Weaver R.
      • Turner-McGrievy G.
      • et al.
      Making Healthy Eating and Physical Activity Policy Practice: the design and overview of a group randomized controlled trial in afterschool programs.
      begins by focusing on the program leader as the primary target of the process of integrating the HE Standards into routine practice. The first steps of the healthy eating strategies begin with:
      • 1.
        the identification/development of a schedule/menu of daily/weekly snack offerings and following the menu; and
      • 2.
        identification of the amount a program has budgeted for snacks (either daily cost per snack/child/day or annual budget) and the location(s) where snacks are purchased.
      These steps are consistent with requirements outlined by the U.S. Department of Agriculture to receive federal reimbursement for snacks.
      U.S. Department of Agriculture Food and Nutrition Service.
      Monitoring Handbook for State Agencies: A Child and Adult Care Food Program Handbook.
      For ASPs without these items, technical assistance provided by research staff focused on developing a 2- or 4-week rotating snack menu that clearly defines the snacks to be served and their respective serving size and number of components, as well as working with ASP leaders to determine challenges associated with serving the menu-specified snacks. For ASPs without knowledge of snack expenditures, monthly snack purchase receipts were collected and subsequent computation of monthly and per snack/child/day costs were performed. Where ASPs were identified as having a limited budget or were purchasing snacks from retailers where the retail cost of snacks that meet the HE Standards was prohibitive, support was provided to identify grocery stores located conveniently to the ASPs where snack purchases could be made that fit within existing budgets. Alternative locations were first identified via Google Maps based on the ASP’s address or based on what the ASP leader indicated was conveniently located near the program. Secondly, the locations were visited, either by the investigative team or ASP leader, to determine if the prices met the ASP’s designated snack budget. Based on this information, an ASP could have selected one or more alternative outlets, at their discretion, to begin purchasing snacks.
      Trainings, conducted by research staff with the site leaders, occurred at the beginning of the school year (August 2013 and August 2014 for the immediate group and August 2014 for the immediate and delayed group) and lasted approximately 3 hours with the site leaders at their location of operation. Where possible, multiple ASPs attended the same training session. Additionally, booster sessions including site leaders and staff were provided. Booster sessions included a walkthrough with the program site leader to review opportunities to meet the HE Standards. For the boosters at each intervention year, research personnel, site leaders, and staff convened a 20 to 30–minute meeting immediately after the end of the ASP to discuss areas that were consistent and inconsistent with meeting the HE Standards. Strategies to address challenges were agreed upon and implemented in subsequent days. The immediate group received four boosters the first intervention year (August 2013 to May 2014) and two boosters during the second intervention year (August 2014 to May 2015), whereas the delayed group received four boosters during the first intervention year (August 2014 to May 2015).

      Measures

      Measurements occurred during spring 2013, 2014, and 2015 (March–April). Consistent with established protocols,
      • Beets M.W.
      • Glenn Weaver R.
      • Turner-McGrievy G.
      • et al.
      Making Healthy Eating and Physical Activity Policy Practice: the design and overview of a group randomized controlled trial in afterschool programs.
      each ASP was visited for data collection on 4 non-consecutive, unannounced days Monday–Thursday at each measurement occasion.
      • Beets M.W.
      • Rooney L.
      • Tilley F.
      • et al.
      Evaluation of policies to promote physical activity in afterschool programs: are we meeting current benchmarks?.
      • Beets M.W.
      • Huberty J.
      • Beighle A.
      Physical activity of children attending afterschool programs: research- and practice-based implications.
      At the beginning and end of each measurement wave, data on the food preparation and storage amenities were collected via self-report from the ASP site leaders. Each site leader was asked to indicate what access the program had to locations where they could store foods and beverages (e.g., refrigerator, pantry) and wash and prepare foods (e.g., sinks not located in bathrooms, kitchens). Additionally, site leaders were asked about snack selection and procurement. Questions covered whether they selected their own snacks; if snacks were provided by non-ASP personnel (e.g., food service personnel provided snack); and if snacks were delivered or picked up by the site leader.
      The primary outcome for this study was the types of foods and beverages ASPs served for snack. The types of foods and beverages served as snacks were recorded via direct observation by trained research personnel in each ASP. Observers were trained via a classroom-based training using photographs of snacks from ASPs, along with in-the-field trainings of snacks served in ASPs not taking part in the study. Immediately at the start of snack, the trained observers recorded the brand name(s), size, and packaging, where appropriate, of the foods and beverages served as the snack for that day. Foods and beverage items served as snacks were classified according to existing categories for snacks and beverages
      • Beets M.W.
      • Tilley F.
      • Kim Y.
      • et al.
      Nutritional policies and standards for snacks served in after-school programmes: a review.
      • Mozaffarian R.S.
      • Wiecha J.L.
      • Roth B.A.
      • et al.
      Impact of an organizational intervention designed to improve snack and beverage quality in YMCA after-school programs.
      : sugar-sweetened beverages (soda, powdered drink mixes); dairy food (string cheese); milk unsweetened (non-fat, 1%); milk sweetened (chocolate, strawberry); 100% fruit juice; salty flavored snacks (Doritos); salty unflavored snacks (pretzels); desserts (cookies); candy (chocolate, frozen treats); non-fruit fruit (fruit roll-ups); prepackaged fruit (applesauce); cereal sugar-sweetened (Fruit Loops); cereal unsweetened (Cheerios); and fruits and vegetables (fresh, frozen, dried) recorded separately. Water was recorded if programs provided water in cups or bottles during snack time. Inter-rater agreement among five observers on the snacks served was 98.4% (κ=0.98) across 133 reliability observations of snacks, which represented approximately 55% of all snacking occasions. The average number of days each food and beverage category served was calculated and standardized to a 5-day school week across sites.

      Statistical Analysis

      For each treatment group, mixed effects logistic regression models, with observation days nested within ASPs, were estimated to evaluate the odds of observing a food or beverage category served (i.e., “1” for observed, “0” for not observed) on one of the 4 observation days during the spring 2013, 2014, and 2015 measurements. Initially, comparisons between the immediate and delayed treatment groups for intervention and non-intervention years were examined. This model included a main effect for treatment group, a time variable dummy coded as non-intervention or intervention time points, and the group X time interaction. Secondary models were run separately for the immediate and delayed groups to examine within-group changes across the three time points. A significant main effect of time would support the hypothesis that the intervention effect would be replicated in the delayed treatment group, as well as a significant within-group contrast between spring 2014 (last year of non-intervention) and spring 2015 (intervention received) in the delayed treatment group. The hypothesis that receiving 2 years of the intervention would result in greater gains than a single would be supported by a significant within-group effect between spring 2014 and spring 2015 in the immediate treatment group, as well as a significant group X time interaction. Each model included the following covariates: Census 2010 ZIP code poverty level, enrollment size, and operating location of the program (community and faith [entered separately] versus school as the reference group). Additionally, the number of ASPs meeting the HE Standards was evaluated by classifying whether ASPs met the standards every day for each study year by treatment group. The criteria for a statistically significant difference between groups was set a priori at p<0.05. All analyses were performed using Stata, version 13.

      Results

      The statistical comparisons between treatment groups for non-intervention and intervention time points are presented in Table 2. The number of days/week a food and beverage category was observed at each study year is presented in Appendix Table 1 (available online) and Appendix Figure 1 (available online). Overall, there were statistically significant main effects of time for fruits (OR=2.84); total fruits/vegetables (OR=3.8); and water (OR=4.65). There were also statistically significant group X time interactions for fruits (OR=13.81); total fruits/vegetables (OR=20.03); and desserts (OR=0.10), which indicated that the immediate group served more fruits, total fruits/vegetables, and served fewer desserts during intervention time points compared with the delayed treatment group. These effects mirror the changes observed in the foods and beverage categories presented in Appendix Table 1 (available online). Based on these estimates, the hypothesis that the intervention effect would be replicated in the delayed treatment group was supported, yet not at the magnitude observed for the immediate treatment group.
      Table 2Between-Group and Within-Group Comparisons for Changes in Snacks Served
      VariablesPrimary treatment group comparison analysesWithin-group comparison between measurement years
      Delayed treatment group (2 years non-intervention, 1 year intervention)Immediate treatment group (1 year non-intervention, 2 years intervention)
      Main effect treatment group,
      Test of baseline equivalency between the immediate and delayed treatment groups. Significant main effect indicates difference in baseline values between the two treatment groups.
      OR (95% CI)
      Main effect time,
      Test of difference between non-intervention and intervention years. Significant time effect indicates a difference between non-intervention and intervention years for the delayed treatment group.
      OR (95% CI)
      Group X time interaction,
      Test of difference on pre- vs post-intervention change between delayed and immediate treatment groups. Significant interaction indicates difference in treatment effect between the two groups. NA, no analysis due to “0” days food or beverage served in one of the groups for one or more time points.
      OR (95% CI)
      Spring 2013 first year non-intervention versus Spring 2014 second year non-intervention, OR (95% CI)Spring 2013 first year non-intervention versus Spring 2015 intervention, OR (95% CI)Spring 2014 second year non-intervention versus Spring 2015 intervention, OR (95% CI)Spring 2013 non-intervention versus Spring 2014 first year intervention, OR (95% CI)Spring 2013 non-intervention versus Spring 2015 second year intervention, OR (95% CI)Spring 2014 first year intervention versus Spring 2015 second year intervention, OR (95% CI)
      Foods
       Fruits0.80 (0.24, 2.73)2.84 (1.13, 7.17)13.81 (3.27, 58.37)0.84 (0.21, 3.35)2.82 (0.86, 9.30)3.36 (1.00, 11.32)32.92 (9.92, 109.19)44.02 (12.60, 153.78)1.39 (0.51, 3.81)
       VegetablesNANANA1.83 (0.13, 24.95)8.61 (0.80, 91.99)4.40 (0.69, 28.03)NANA0.56 (0.21, 1.50)
       Total fruits/ vegetables0.66 (0.15, 2.92)3.80 (1.45, 9.95)20.03 (4.18, 96.03)0.98 (0.24, 3.95)4.26 (1.21, 15.09)5.96 (1.39, 25.66)44.36 (12.95, 151.90)65.33 (17.54, 243.36)1.56 (0.51, 4.76)
       DipsNANANANA1.62 (0.25, 10.56)5.97 (2.04, 17.47)2.06 (0.67, 6.32)0.35 (0.14, 0.85)
       Dairy0.93 (0.23, 3.86)0.68 (0.19, 2.43)3.13 (0.60, 16.22)0.74 (0.18, 3.06)0.60 (0.15, 2.45)0.78 (0.17, 3.53)1.02 (0.30, 3.49)4.12 (1.28, 13.22)4.33 (1.42, 13.23)
       Salty snacks—flavored2.87 (1.10, 7.54)0.56 (0.24, 1.30)0.78 (0.25, 2.44)1.04 (0.41, 2.64)0.57 (0.22, 1.47)0.53 (0.20, 1.38)0.22 (0.09, 0.58)0.67 (0.28, 1.64)3.54 (1.31, 9.55)
       Salty snacks—unflavored6.31 (1.80, 22.08)0.87 (0.26, 2.85)0.71 (0.17, 3.01)0.68 (0.16, 2.94)0.71 (0.18, 2.80)1.02 (0.24, 4.33)0.61 (0.25, 1.48)0.73 (0.30, 1.76)1.14 (0.48, 2.70)
       Desserts1.43 (0.62, 3.31)0.64 (0.31, 1.31)0.10 (0.03, 0.33)1.25 (0.51, 3.07)0.69 (0.29, 1.63)0.54 (0.23, 1.29)0.03 (0.01, 0.13)0.07 (0.02, 0.22)2.21 (0.45, 10.86)
       Cereal—sugar-sweetenedNANANA0.85 (0.22, 3.22)0.58 (0.15, 2.25)0.68 (0.19, 2.43)NANANA
       Cereal—unsweetened0.44 (0.01, 14.65)1.61 (0.32, 8.11)3.71 (0.24, 58.50)12.58 (0.80, 197.86)7.72 (0.55, 107.81)0.62 (0.09, 4.46)12.45 (1.19, 129.94)1.72 (0.13, 23.23)0.11 (0.01, 0.79)
      Beverages
       Sugar-sweetened non-dairy beveragesNANANA1.01 (0.27, 3.70)0.09 (0.1, 0.54)0.09 (0.01, 0.57)NANANA
       Milk—sugar-sweetenedNANANA1.55 (0.33, 7.33)0.61 (0.11, 3.25)0.41 (0.08, 2.02)NANANA
       Non-100% fruit juiceNANANANANANANANANA
        Total sugar-sweetened beveragesNANANA1.09 (0.37, 3.21)0.21 (0.06, 0.74)0.19 (0.05, 0.68)NANANA
       100% fruit juiceNANANA1.95 (0.54, 7.01)0.35 (0.10, 1.25)0.16 (0.04, 0.72)NANANA
       Water1.12 (0.16, 7.60)4.65 (1.69, 12.79)4.17 (0.91, 19.17)1.53 (0.51, 4.55)6.00 (1.84, 19.54)6.13 (1.49, 25.19)10.61 (3.00, 37.52)48.64 (8.00, 295.71)12.53 (1.13, 138.37)
       Milk—unflavored2.60 (0.33, 20.41)2.46 (0.73, 8.36)0.75 (0.12, 4.52)0.32 (0.05, 2.00)1.52 (0.39, 5.91)5.70 (0.86, 37.53)2.57 (0.59, 11.08)1.58 (0.27, 9.10)1.10 (0.12, 10.06)
      Note: Boldface indicates a statistically significant OR at p<0.05.
      a Test of baseline equivalency between the immediate and delayed treatment groups. Significant main effect indicates difference in baseline values between the two treatment groups.
      b Test of difference between non-intervention and intervention years. Significant time effect indicates a difference between non-intervention and intervention years for the delayed treatment group.
      c Test of difference on pre- vs post-intervention change between delayed and immediate treatment groups. Significant interaction indicates difference in treatment effect between the two groups.NA, no analysis due to “0” days food or beverage served in one of the groups for one or more time points.
      The within-group model estimates are presented in Table 2. For the immediate group, improvements in the number of days where a fruit/vegetable was served were observed at both intervention time points (i.e., spring 2014 and 2015) from 0.6 days/week at baseline to 4.1 and 4.3 days/week (OR=44.4 and 65.3). The number of days that desserts were served was substantially reduced from 2.9 days/week at baseline to 0.3 and 0.6 days/week (OR=0.03 and 0.07). Sugar-sweetened beverages were entirely eliminated with programs serving them 2.4 days/week at baseline to 0.1 and 0.0 days/week by Intervention Years 1 and 2. Water served as a beverage increased across intervention years from 2.7 days/week at baseline to 4.4 and 4.8 days/week (OR=10.6 and 48.6). The only improvement in the immediate group from receiving 2 years of the intervention (improvement from spring 2014 to spring 2015) was for the number of days that water was served (OR=12.5). All other improvements made during the first year of intervention were maintained in the immediate treatment group. Based on these estimates, the hypothesis that 2 years of receiving the intervention would result in greater gains than receiving 1 year of the intervention was not supported.
      For the delayed group, improvements in foods and beverages were observed between the 2 non-intervention years (spring 2013 and 2014) and the intervention year (spring 2015), but not at the magnitude of the immediate group, partially supporting the original hypothesis of a replication of the immediate group’s first-year effects. Specifically, the number of days that fruits/vegetables were served were 0.7 and 0.8 days/week during spring 2013 and 2014 (both non-intervention years), with this increasing to 1.7 days/week during the intervention (spring 2015, OR=5.96). The number of days that sugar-sweetened beverages were served reduced from 1.5 days/week during both non-intervention years to 0.5 days/week during the intervention (OR=0.19). Water served as a beverage went from 2.3 and 3.0 days/week during non-intervention years to 3.7 days/week during the intervention (OR=6.13).
      The number of programs meeting some of the components of the HE Standards improved from non-intervention to intervention years in both groups. The immediate intervention went from zero ASPs at baseline (spring 2013) to six of ten ASPs by the last year of intervention (spring 2015) serving a fruit/vegetable every day. Additionally, the immediate treatment group went from only three ASPs serving water every day to nine of ten ASPs by spring 2015. The delayed group went from zero ASPs serving a fruit/vegetable every day and one to three ASPs serving water every day during both non-intervention years (spring 2013 and 2014) to two of ten serving a fruit/vegetable every day and six of ten ASPs serving water every day by spring 2015. Importantly, across treatment groups, there were only eight of 20 ASPs during non-intervention years that served a fruit/vegetable on ≥1 days, with this increasing to 17 of 20 ASPs (seven delayed and ten immediate) serving a fruit/vegetable at least once a week by the end of the intervention. Further, the number of ASPs serving a dessert or sugar-sweetened beverage went from 18 and 12 of 20, respectively, at baseline, to 11 and one of 20, respectively, by the end of the intervention. Table 3 includes descriptive information on the snack purchasing structure for each program by treatment group. Overall, there were nine different snack purchasing scenarios across the 20 programs in the study. The delayed treatment group included more ASPs (seven delayed versus three immediate) operating as third parties within a building owned by another entity (e.g., school) and indicated fewer ASPs with adequate storage space for snacks that meet the HE Standards (three delayed versus ten immediate), compared with the immediate group.
      Table 3Snack Purchasing Structure for Each Afterschool Program (ASP) (N=20)
      Snack purchasing characteristicImmediate treatment groupDelayed treatment group
      1234567891012345678910
      Operating in own physical structure” refers to programs operated within building owned by program (e.g., program operated by school and conducted within school buidling)
      Operating as third party” refers to programs operating within building owned by another entitiy (e.g., program operated by non-school entirty and conducted within school building)
      Adequate storage space for enrollment size” refers to the availability and use of snack storage space that meets the needs of the number of children attending each day
      Site leader selects own snack” refers to programs where the site leader has direct control over the selection of snacks to serve
      Snack provided by non-ASP affiliated personnel” refers to programs that rely on individuals outside the program to provide snack (e.g., food service personnel provide snack)
      Site leader picks up snack” refers to programs where the site leader needs to go to a location, outside of where the program operates, to pick up snacks (e.g., site leader goes to grocery store to pick up snacks)
      Snack delivered” refers to programs where snack is delivered to the site leader at the location of operation
      Food preparation and storage
       Refrigerator and freezer combination unit
       Standalone freezer
       Pantry or storage closet for dry goods
       Portable cooler/ice chest
       Kitchen
       Food preparation sink that is not in a bathroom
       Stove/oven
       Microwave
       Serving plates or bowls
      Note: Numbers 1–10 represent individual afterschool programs in each treatment group.

      Discussion

      This study presents the 2-year outcomes from a group RCT evaluating changes in foods and beverages served as snacks in ASPs.
      • Beets M.W.
      • Glenn Weaver R.
      • Turner-McGrievy G.
      • et al.
      Making Healthy Eating and Physical Activity Policy Practice: the design and overview of a group randomized controlled trial in afterschool programs.
      Overall, improvements in the types of food and beverages served for snack were observed in both treatment groups, yet the delayed group did not improve the foods and beverages served as much as the immediate group. These findings support that using the STEPs framework with ASP providers to serve snacks that comply with the HE Standards can lead to meaningful changes in the types of foods and beverages served for snack. However, full compliance to the Standards was not achieved in any ASP and only a single component, sugar-sweetened beverages, was the closest across all 20 ASPs to full compliance (i.e., only one of 20 APS serving sugar-sweetened beverages following the intervention). This suggests other strategies or longer exposure to the intervention (>2 years) is required to help programs fully comply with the HE Standards.
      Important findings in this study were the differences in the magnitude of the changes between the immediate and delayed groups, which can be attributed to differences in the snack purchasing structure across groups. For the immediate group, substantial changes were observed by the end of the first year of the intervention (spring 2014) for most of the primary HE Standards (Appendix Table 1 and Appendix Figure 1, available online), thus not supporting the hypothesis that receiving 2 years of the intervention would result in greater improvements versus only 1 year. Conversely, the changes for the delayed group, although statistically significant, were not as large, thus only partially supporting the hypothesis of replicating the immediate group’s first-year improvements in the delayed treatment group. This pattern was observed across the other HE Standards. A possible reason for this was five of the ten ASPs in the delayed group had a standing contract with a food service vendor that delivered orders twice a month (Table 3). During the intervention year, these sites added a fruit or vegetable in the deliveries, but chose to order fruits or vegetables for only the 2 days immediately following the delivery owing to limited storage space that would accommodate storage for fruits or vegetables for an average of 75 children/ASP each day for 10 week days. Attempts to increase storage space to store boxes of fruits or vegetables, even at room temperature (e.g., apples, bananas) for these programs, beyond the closets each program already used, were unsuccessful.
      It is important to note that these programs were operated by third parties within elementary schools and had been operating in these schools for a number of years prior to this intervention. Storage space, however, was not an issue for programs that either operated within their own physical building (e.g., school operating the ASP within the school or a church operating an ASP on church property) or that were provided a classroom or office space within the physical building. Lack of adequate storage was reported by school food personnel in a previous study
      • Nanney M.S.
      • Olaleye T.M.
      • Wang Q.
      Incorporating a healthy reimbursable snack in an afterschool homework program for middle school students: a case study.
      and highlights one of the challenges of achieving the HE Standards for programs that are unable to procure sufficient space to store fruits or vegetables that, because of their size, require more storage space than the snacks they are replacing (e.g., apples versus individual serving–size bags of chips). Further, although the availability/access to storage space for foods and beverages were collected from all programs, it appears that, although most of the ASPs reported space available, the space’s physical size in relation to what would be required to store fruits/vegetables was not considered. Future studies need to clarify available space with ASPs to ensure that what is reported can ultimately accommodate the healthier snacks.
      Previous studies
      • Cassady D.
      • Vogt R.
      • Oto-Kent D.
      • et al.
      The power of policy: a case study of healthy eating among children.
      • Mozaffarian R.S.
      • Wiecha J.L.
      • Roth B.A.
      • et al.
      Impact of an organizational intervention designed to improve snack and beverage quality in YMCA after-school programs.
      • Giles C.M.
      • Kenney E.L.
      • Gotrtmaker S.L.
      • et al.
      Increasing water availability during afterschool snack: evidence, strategies, and partnerships from a group randomized trial.
      • Beets M.W.
      • Tilley F.
      • Turner-McGrievy G.
      • et al.
      Community partnership to address snack quality and cost in after-school programs.
      • Beets M.W.
      • Tilley F.
      • Weaver R.G.
      • et al.
      From policy to practice: addressing snack quality, consumption, and price in after-school programs.
      have demonstrated improvements in the types of foods and beverages served in ASPs over a single year, similar to those reported herein. These studies utilized the adoption of policy (i.e., the HE Standards or a variant) and assistance with developing healthier menus to help guide the selection of snacks that met HE Standards. This approach, therefore, has considerable merit for programs that, as identified above, have appropriate space to store the healthier snack options. A previous study
      • Dzewaltowski D.A.
      • Rosenkranz R.R.
      • Geller K.S.
      • et al.
      HOP’N after-school project: an obesity prevention randomized controlled trial.
      did not observe any changes to the types of foods and beverages served for snack in ASPs operating as a third party within elementary schools. Primary challenges identified were the difficulties working with school food personnel who were providing the snack and costs associated with purchasing fruits/vegetables.
      • Hastmann T.J.
      • Bopp M.
      • Fallon E.A.
      • et al.
      Factors influencing the implementation of organized physical activity and fruit and vegetable snacks in the HOP’N after-school obesity prevention program.
      In this study, there were no reports of programs having difficulties working with non-ASP affiliated personnel who provided the snack (e.g., school food personnel providing snack to site leader), and previous studies have shown that purchasing fruits/vegetables does not increase overall snack cost
      • Beets M.W.
      • Weaver R.G.
      • Turner-McGrievy G.
      • et al.
      Making Healthy Eating Policy Practice: a group randomized controlled trial on changes in snack quality, costs, and consumption in after-school programs.
      • Beets M.W.
      • Tilley F.
      • Weaver R.G.
      • et al.
      From policy to practice: addressing snack quality, consumption, and price in after-school programs.
      and improves the nutrition profile of the provided snacks.
      • Beets M.W.
      • Turner-McGrievy B.
      • Weaver R.G.
      • et al.
      Intervention leads to improvements in the nutrient profile of snacks served in afterschool programs: a group randomized controlled trial.
      Based on the collective findings across these studies, additional work is necessary to understand what works best given the wide range of likely snack purchasing scenarios that will successfully assist ASPs in achieving these important nutrition policy goals. As shown in Table 3, there were nine different snack procurement scenarios, each of which required a different set of strategies of whom to involve in the final snack purchasing decisions. This information should serve to inform future efforts at improving the quality of snacks served in ASPs.

      Limitations

      There are numerous strengths of this study, including a large number of diverse ASPs, direct observation of served snacks, and the delayed treatment group design. However, there are several limitations. All the ASPs operated within a single southeastern state, and generalizing the findings to programs operating in other states may not be appropriate. Second, although the number of programs included in this study represents one of the largest conducted to date, the sample size still limited the ability to adequately model program-level characteristics, such as snack procurement scenarios, to determine if these influenced the snacks that were served. Third, the randomization to condition did not account for the snack procurement scenarios. Despite these limitations, this study represents one of the largest conducted to date with a diverse range of programs, and therefore provides valuable information for future studies attempting to assist ASPs in meeting the HE Standards.

      Conclusions

      Future studies need to focus on assisting providers in identifying and securing adequate storage space for fruits/vegetables, especially those programs operating as a third party in a physical structure. Studies need to include a larger number of programs in order to better understand program-level characteristics and their association with improving the foods and beverages served.

      Acknowledgments

      Research reported in this publication was supported by the National Heart, Lung, and Blood Institute of NIH under Award Number R01HL112787. The content is solely the responsibility of the authors and does not necessarily represent the official views of NIH. ClinicalTrials.gov: NCT02144519.
      No financial disclosures were reported by the authors of this paper.

      Supplementary material

      References

        • Afterschool Alliance
        America After 3 PM: A Household Survey on Afterschool in America.
        Afterschool Alliance, Washington, DC2014
        • Beets M.W.
        • Tilley F.
        • Kim Y.
        • et al.
        Nutritional policies and standards for snacks served in after-school programmes: a review.
        Public Health Nutr. 2011; 14: 1882-1890https://doi.org/10.1017/S1368980011001145
        • Skinner J.D.
        • Carruth B.R.
        • Wendy B.
        • et al.
        Children’s food preferences: a longitudinal analysis.
        J Am Diet Assoc. 2002; 102: 1638-1647https://doi.org/10.1016/S0002-8223(02)90349-4
        • Bertino M.
        • Beauchamp G.K.
        • Engelman K.
        Increasing dietary salt alters salt taste preference.
        Physiol Behav. 1986; 38: 203-213https://doi.org/10.1016/0031-9384(86)90155-1
        • Ifland J.R.
        • Preuss H.G.
        • Marcus M.T.
        • et al.
        Refined food addiction: a classic substance use disorder.
        Med Hypotheses. 2009; 72: 518-526https://doi.org/10.1016/j.mehy.2008.11.035
        • Cooke L.
        The importance of exposure for healthy eating in childhood: a review.
        J Hum Nutr Diet. 2007; 20: 294-301https://doi.org/10.1111/j.1365-277X.2007.00804.x
        • Beets M.W.
        • Weaver R.G.
        • Tilley F.
        • et al.
        Salty or sweet? Nutritional quality, consumption, and cost of snacks served in afterschool programs.
        J Sch Health. 2015; 85: 118-124https://doi.org/10.1111/josh.12224
        • Cassady D.
        • Vogt R.
        • Oto-Kent D.
        • et al.
        The power of policy: a case study of healthy eating among children.
        Am J Public Health. 2006; 96: 1570-1571https://doi.org/10.2105/AJPH.2005.072124
        • Coleman K.J.
        • Geller K.S.
        • Rosenkranz R.R.
        • et al.
        Physical activity and healthy eating in the after-school environment.
        J Sch Health. 2008; 78: 633-640https://doi.org/10.1111/j.1746-1561.2008.00359.x
        • Mozaffarian R.S.
        • Wiecha J.L.
        • Roth B.A.
        • et al.
        Impact of an organizational intervention designed to improve snack and beverage quality in YMCA after-school programs.
        Am J Public Health. 2010; 100: 925-932https://doi.org/10.2105/AJPH.2008.158907
      1. National Afterschool Association. Standards for healthy eating and physical activity in out-of-school time programs. http://naaweb.org/images/NAA_HEPA_Standards_new_look_2015.pdf. Published 2011. Accessed December 2, 2011.

        • Giles C.M.
        • Kenney E.L.
        • Gotrtmaker S.L.
        • et al.
        Increasing water availability during afterschool snack: evidence, strategies, and partnerships from a group randomized trial.
        Am J Prev Med. 2012; 43: S136-S142https://doi.org/10.1016/j.amepre.2012.05.013
        • Beets M.W.
        • Weaver R.G.
        • Moore J.B.
        • et al.
        From policy to practice: strategies to meet physical activity standards in YMCA afterschool programs.
        Am J Prev Med. 2014; 46: 281-288https://doi.org/10.1016/j.amepre.2013.10.012
        • Hastmann T.J.
        • Bopp M.
        • Fallon E.A.
        • et al.
        Factors influencing the implementation of organized physical activity and fruit and vegetable snacks in the HOP’N after-school obesity prevention program.
        J Nutr Educ Behav. 2013; 45: 60-68https://doi.org/10.1016/j.jneb.2012.06.005
        • Kelder S.
        • Hoelscher D.M.
        • Barroso C.S.
        • et al.
        The CATCH Kids Club: a pilot after-school study for improving elementary students’ nutrition and physical activity.
        Public Health Nutr. 2005; 8: 133-140https://doi.org/10.1079/PHN2004678
        • Beets M.W.
        • Glenn Weaver R.
        • Turner-McGrievy G.
        • et al.
        Making Healthy Eating and Physical Activity Policy Practice: the design and overview of a group randomized controlled trial in afterschool programs.
        Contemp Clin Trials. 2014; 38: 291-303https://doi.org/10.1016/j.cct.2014.05.013
        • Beets M.W.
        • Weaver R.G.
        • Turner-McGrievy G.
        • et al.
        Making Healthy Eating Policy Practice: a group randomized controlled trial on changes in snack quality, costs, and consumption in after-school programs.
        Am J Health Promot. 2016; 30: 521-531https://doi.org/10.4278/ajhp.141001-QUAN-486
        • Beets M.W.
        • Turner-McGrievy B.
        • Weaver R.G.
        • et al.
        Intervention leads to improvements in the nutrient profile of snacks served in afterschool programs: a group randomized controlled trial.
        Transl Behav Med. 2016; 6: 329-338https://doi.org/10.1007/s13142-015-0342-z
        • Hawe P.
        • Shiell A.
        • Riley T.
        Complex interventions: how “out of control” can a randomised controlled trial be?.
        BMJ. 2004; 328: 1561-1563https://doi.org/10.1136/bmj.328.7455.1561
        • U.S. Department of Agriculture Food and Nutrition Service.
        Monitoring Handbook for State Agencies: A Child and Adult Care Food Program Handbook.
        U.S. Department of Agriculture, Washington, DC2013
        • Beets M.W.
        • Rooney L.
        • Tilley F.
        • et al.
        Evaluation of policies to promote physical activity in afterschool programs: are we meeting current benchmarks?.
        Prev Med. 2010; 51: 299-301https://doi.org/10.1016/j.ypmed.2010.07.006
        • Beets M.W.
        • Huberty J.
        • Beighle A.
        Physical activity of children attending afterschool programs: research- and practice-based implications.
        Am J Prev Med. 2012; 42: 180-184https://doi.org/10.1016/j.amepre.2011.10.007
        • Nanney M.S.
        • Olaleye T.M.
        • Wang Q.
        Incorporating a healthy reimbursable snack in an afterschool homework program for middle school students: a case study.
        Health Educ Behav. 2012; 39: 127-130https://doi.org/10.1177/1090198111404066
        • Beets M.W.
        • Tilley F.
        • Turner-McGrievy G.
        • et al.
        Community partnership to address snack quality and cost in after-school programs.
        J Sch Health. 2014; 84: 543-548https://doi.org/10.1111/josh.12175
        • Beets M.W.
        • Tilley F.
        • Weaver R.G.
        • et al.
        From policy to practice: addressing snack quality, consumption, and price in after-school programs.
        J Nutr Educ Behav. 2014; 46: 384-389https://doi.org/10.1016/j.jneb.2013.10.005
        • Dzewaltowski D.A.
        • Rosenkranz R.R.
        • Geller K.S.
        • et al.
        HOP’N after-school project: an obesity prevention randomized controlled trial.
        Int J Behav Nutr Phys Act. 2010; 7: 90https://doi.org/10.1186/1479-5868-7-90