Individual- and School-Level Factors Related to School-Based Salad Bar Use Among Children and Adolescents

Abstract

Background. Consumption levels of fruits and vegetables (F/V) among children/adolescents are low. Programs like school-based salad bars (SB) provide children/adolescents increased F/V access. Aims. The purpose of this study was to examine the relationship between SB use and individual and school-level factors among elementary and secondary school students in New Orleans public schools. Method. Twelve schools receiving SB units from the Let’s Move Salad Bars to Schools Campaign participated in this study. Self-reported data were collected from students (n = 1,012), administrators (n = 12), and food service staff (n = 37). School environmental data were obtained through direct observation. Generalized estimating equation regression methods were used to develop a multilevel model including both school-level (e.g., length of lunchtime, SB marketing, vending machines) and individual-level (e.g., sex, food preferences, nutrition knowledge) effects. Results. Female students had higher odds of using the SB compared to males. Students with healthier food preferences had higher odds of using the SB than those who reported less healthy food preferences. Within the multilevel model for all students, only sex and healthy food preferences remained significant. In a multilevel model assessing secondary students only, student encouragement toward others for healthy eating and school-based SB marketing were significantly related to SB use. Conclusions. Little research has examined factors related to school-based SB use. These findings suggest recommendations that may help improve student use of SBs. For example, increasing the promotion of SB, particularly in secondary schools, might encourage their use among students.

Keywords

child health fruit and vegetable consumption nutrition salad bars school health services

Children and adolescents in the United States do not consume the nationally recommended levels of fruits and vegetables (F/V). Low consumption of F/V can have multiple negative health consequences, such as increased risk for overall mortality, certain cancers, and cardiovascular disease (Bazzano et al., 2002; Blomhoff, 1998; Boffetta et al., 2010; George et al., 2009; X. Wang et al., 2014). Included in the negative effects for low consumption of F/V are overweight and obesity. Increased F/V consumption can displace the consumption of high-fat, energy-dense foods (Epstein et al., 2001; Hill & Peters, 1998) and thus aid in the prevention of child and adolescent obesity; yet many children and adolescents are not currently meeting dietary guideline recommendations (Kim et al., 2014).

More than 55 million students are enrolled in primary and secondary schools in the United States, and more than one third of the recommended total daily caloric intake is consumed within the school environment (Briefel, Wilson, & Gleason, 2009; Synder & Dillow, 2011). Programs and interventions aimed at increasing F/V consumption in schools have proliferated and provide strategies toward healthier eating for U.S. youth. These strategies, then, have the potential to reach large numbers of students across the entire country.

Improving access to F/V by children and adolescents is associated with increased F/V consumption, particularly in the school environment (Blanchette & Brug, 2005; Ganann et al., 2014; Terry-McElrath, O’Malley, & Johnston, 2014). School-based salad bars (SBs) have increased in popularity and are one suggested method for improving F/V access in school-aged youth (French & Wechsler, 2004). Between the years 2008 and 2012, 39% of middle school students and 49% of high school students had access to SBs (Terry-McElrath et al., 2014), and approximately 31% of elementary students had access to SBs (Turner & Chaloupka, 2015). The Let’s Move Salad Bars to Schools (LMSB2S) initiative offered schools with limited resources an opportunity to implement an SB (Harris et al., 2012). Through grassroots fundraising, the LMSB2S initiative donates the SB unit to schools who have requested the donation (Harris et al., 2012). Empirical evidence suggests that F/V consumption can be increased with school-based SBs. Elementary school students with access to an SB report a higher intake of F/V compared to students without access (Slusser, Cumberland, Browdy, Lange, & Neumann, 2007), and SBs with a greater variety are associated with increase F/V consumption in elementary school students (Adams, Pelletier, Zive, & Sallis, 2005). It is important to note that Adams et al. (2005) used objective measures to quantify F/V consumption. Additionally, Gosliner (2014) and Terry-McElrath et al. (2014) found self-reported vegetable intake was greater among adolescent students attending schools with SBs than students without access to SBs; yet more evidence linking F/V consumption to SBs is needed (Adams, Bruening, & Ohri-Vachaspati, 2015).

The value of an SB program depends on whether students actually use the SB, but few studies have examined individual- and school-level factors related to their use. Adams et al. (2016) investigated the impact of the location of the SBs in school cafeterias and found that students who had access to the SB within their serving lines served themselves more F/V and consumed more F/V compared to students who accessed the SB outside the main food line (Adams, Bruening, Ohri-Vachaspati, & Hurley, 2016). Other research identified predictors of SBs’ presence in the cafeteria; schools that participated in the Fresh Fruit and Vegetable, Team Nutrition, and Farm-to-School programs and schools with lunch provided by a food service management company were more likely to have an SB in the cafeteria (Ohri-Vachaspati, Turner, Adams, Bruening, & Chaloupka, 2016). In a study investigating individual characteristics and their association on SB use in adolescents, Andersen et al. (2015) identified social support and food preferences as predictors of SB use.

The social ecological model (SEM) suggests that individuals are influenced by multiple interacting and overlapping systems ( McLeroy, Bibeau, Steckler, & Glanz, 1988). This means attention needs to be placed not only on individual behavior but also on peer group and school environmental factors that may influence the use of school-based SBs. The SEM guided the development and implementation of this research to identify variables associated with SB use. The research question guiding this study was, “What individual- and school-level factors are related to student use of the SB?” The purpose of this study was to examine the relationship between SB use and individual- and school-level factors among children and adolescents in New Orleans public schools.

Method

Forty-three New Orleans, Louisiana, schools received SB units through the LMSB2S initiative and were contacted for participation in this cross-sectional study. The donation was facilitated for the 2011-2012 school year. Of the 43 schools who received SBs, 7 schools were not operating SBs and were eliminated from participation; of the remaining schools, 12 schools agreed to participate. Schools that did and did not participate were similar in percent African American and free and reduced lunch (FRL) percent (participating schools: FRL range: 55% to 100%, African American range: 60% to 99%; nonparticipating schools: FRL range: 53% to 100%, African American range: 43% to 99%). Because of the unique charter school structure in New Orleans, these 12 schools represent 8 different charter associations; no charter association had more than three schools represented. Self-reported data were collected from students, administrators, food service directors, and food service staff via pencil-and-paper surveys, and school environmental data were documented through direct observation. All data collection instruments, procedures, and protocols were approved by the Tulane Social-Behavioral Institutional Review Board.

Student Survey

The student survey consisted of questions extracted from previously published studies (Edmundson et al., 1996; Hoelscher, Day, Kelder, & Ward, 2003; Kelder et al., 2005) and was organized and reviewed by an expert panel at the Tulane University School of Public Health and Tropical Medicine Prevention Research Center. The student survey was pilot-tested for readability and understanding at a similar SB school that did not participate in the main study. After pilot testing, revisions to the questionnaire were made as needed. Survey data included the following: demographic characteristics, typical SB use, food preferences, knowledge about healthy food, and social support for eating healthy foods for secondary students. Some data, including social support and additional knowledge variables, were collected only for secondary students. Parents were notified of the study through a flyer distributed at school. Passive consent was obtained for the study. At school, students had the opportunity to assent into the study. The consent and assent documents described to parents and students that the study would ask about knowledge, attitudes, and behaviors about healthy foods.

Two questions that differentiated SB users from nonusers were combined to determine SB student use: (1) “How many days during a usual week do you eat from the salad bar during lunch served at school?” (Responses: Never, 1-2 days, 3-4 days, and every day) and (2) “When you eat from the salad bar, do you get salad more than once during the lunch period?” (Responses: yes or no). Responses from the first question were collapsed into two categories: students who responded that they never eat from the salad bar were coded “0,” and students who responded that they eat from the SB 1 to 2 days, 3 to 4 days, and every day were coded “1.” Responses from the second question were coded “0” for “no” and “1” for “yes.” Students who had conflicting results (e.g., answered no to one questions and yes to the other; n = 135) were excluded from the analysis.

A number of demographic variables were considered for analysis. The variable grade had three levels in the analysis of the total sample, elementary (3rd-5th), middle (6th-8th), and high school (9th-12th). In the analysis examining secondary students only, this was refined to 7th to 8th grades, 9th to 10th grades, and 11th to 12th grades. Race was collapsed from six categories (White, Black/African American, Hispanic, Asian/Pacific Islander, American Indian/Alaskan Native, and Other) into two categories (Black/African American and Other) since the majority of the sample was African American. All other variables used in the analysis can be seen in Table 1.

Table 1.

Variable Names and Items Included in Each Variable.

Student survey
Variable name	Items included in variable	Creation of variable
Healthy food preference	Q1. If you were at the movies, which one would you pick? A1. Popcorn with salt or butter or popcorn without salt or butter* Q2. Which one would you pick to fix with dinner? A2. Fresh or frozen vegetables or canned vegetables Q3. If you were going to eat your lunch, which would you do? A3. Eat the food without adding salt or shake salt on the food before eating Q4. Which would you put on your hamburger? A4. Ketchup or tomato Q5. Which would you pick to drink? A5. Regular milk or low-fat or skim milk Q6. Which food would you eat for a snack? A6. Candy bar or fresh fruit Q7. Which would you do if you were going to eat a piece of chicken? A7. Leave the skin on or take the skin off Q8. Which food would you ask for? A8. Frozen yogurt or ice cream Q9. Which would you choose to cook if you were going to help make dinner at home? A9. French fries or baked potato Q10. Which would you do if you were going to eat cooked vegetables? A10. Eat without butter or add butter Q11. Which would you order if you were going to eat at a fast food restaurant? A11. Regular hamburger or salad*Italicized text indicates correct response option	One point was assigned per correct response. Maximum number of points was 11. All questions summed and dichotomized at midpoint.
Nutrition knowledge	Q1. Which food is healthier? A1. Whole wheat bread* or white bread A2. Broiled beef or broiled fish A3. Oatmeal or eggs and bacon A4. Beef or beans A5. Chicken or regular hamburger A6. Regular milk or low-fat/skim milk A7. Peanut butter or bologna A8. Frozen yogurt or ice cream A9. Green salad or French fries A10. Raisins or candy bar A11. Frozen corn or canned corn A12. French fries or baked potato A13. Regular peanut butter or freshly ground peanut butter*Italicized text indicates correct response option	One point was assigned per correct response. Maximum number of points was 13. All questions summed and dichotomized at midpoint.
Nutrition knowledge (secondary students)	Q1. The dietary guidelines state that half your plate should be A1. Whole grains, low-fat dairy products, fruits and vegetables, meats, beans, fish, poultry, I don’t know Q2. Which of the following food or food components should be decreased in our diet? A2. Low-fat dairy products, fruits and vegetables, saturated and trans-fats, refined grains, whole grains, sodium, I don’t know Q3. Which of the following food or food components should be increased in our diet? A3. Low-fat dairy products, fruits and vegetables, saturated and trans-fats, refined grains, whole grains*, sodium, I don’t know	One point was assigned per correct response. Maximum number of points was 17. All questions summed and dichotomized at midpoint.
	Q4. Which of the following has the most calories?
	A4. One gram of protein, one gram of fat, one gram of carbohydrate, I don’t know
	*Italicized text indicates correct response option
Family/peer modeling	Q1. My parents eat lots of fruits and vegetablesA1. Never, Some of the time, Most of the time, All of the timeQ2. My friends eat lots of fruits and vegetablesA2. Never, Some of the time, Most of the time, All of the time	Responses were dichotomized to Never/Some of the time and Most of the time/All of the time. One point was assigned for a response of Most of the time/All of the time and responses to both questions were summed.
Respondent encouragement	Q1. I encourage my family and friends to eat fruits and vegetables.A1. Never, Some of the time, Most of the time, All of the time	Responses were dichotomized to Never/Some of the time and Most of the time/All of the time
Administrator survey
Variable name	Items included in variable	Creation of variable
Length of lunch	A. Lunch start timeB. Lunch end timeC. Number of lunch periods	Subtracted lunch end time by lunch start time; divided by number of lunch periods; dichotomized at midpoint.
Wellness policy	Q1. Are there any school policies about wellness, nutrition, or healthy food for students?Q2. Are there policies governing the sale of food and beverages on campus other than the USDA reimbursable meal requirements for vending machines?Q3. Are there policies governing the sale of food and beverages on campus other than the USDA reimbursable meal requirements for snack stores?Q4. Are there policies governing the sale of food and beverages on campus other than the USDA reimbursable meal requirements for a la carte items?Q5. Are there policies governing the sale of food and beverages on campus other than the USDA reimbursable meal requirements for school sponsored fundraising activities?	One point was assigned per each positive response. All questions summed and dichotomized at midpoint.
Salad bar (SB) marketing	Q1. Please indicate the types of promotional activities that are being done to encourage students to eat from the salad bar . . .A1. Posters in cafeteriaA2. Displays at the lunch line or registerA3. Table tentsA4. Posters outside the cafeteriaA5. Creative labelingA6. Special title or name for the salad barA7. Opportunities for students or families to visit the cafeteria to learn about items in the salad barA8. Taste testing of salad bar itemsA9. Promotional days for the salad barA10. information materials for families about the salad bar	One point assigned for each promotional strategy selected. All responses were summed and dichotomized at midpoint.
Nutrition education	Q1. Do students receive nutrition education?	Yes/no
Other nutrition education	Q1. Are there any other nutrition programs in the school?A1. Cooking coursesA2. Farm-to-SchoolA3. School gardenA4. Team NutritionA5. Healthier US ChallengeA6. Alliance for Healthier GenerationA7. Other	One point was assigned per each positive response. Summed and dichotomized (some programming versus none).
School food environmental scan
Variable name	Items included in variable	Creation of variable
SB offerings	Q1. Are fresh vegetables offered in the salad bar?Q2. Are canned vegetables offered in the salad bar?Q3. Is fresh fruit offered in the salad bar?Q4. Is canned fruit offered in the salad bar?Q5. Are vegetable/fruit salads offered in the salad bar?	One point assigned for each positive response. Maximum number of reported offerings was 10. Summed and dichotomized at midpoint.
Vending machines	Q1. Number of vending machines in the school outside the cafeteria	Summed and dichotomized (some versus none)
	Q2. Number of vending machines in the cafeteria
Nutrition posters	Q1. Are there any posters on nutrition in the cafeteria?	Yes/no
Food advertisements	Q1. Are there any advertisements for food or beverages within the cafeteria?	Yes/no
Food service staff survey
Variable name	Items included in variable	Creation of variable
Food service attitudes	A. The salad bar has changed the lunchroom atmosphere for the better for the studentsB. The salad bar has changed the lunchroom atmosphere for the better for the food service staff.C. I have not received adequate training to maintain the salad bar.D. The food service staff has the proper equipment to maintain the salad barE. The amount of time and work spent on preparing and maintaining the salad bar is not worthwhile.F. The salad bar has changed the way I feel about my job for the better.G. Since the salad bar was put in the school, I have been eating more fresh fruits and vegetables.H. Since the salad bar was put in the school, I think the students are eating less fresh fruits and vegetables.I. I believe that the students have positive attitudes toward the salad bar.J. Overall, the reception of the food service staff to the salad bar has been positive.Response options: Strongly Agree, Agree, Neutral, Disagree, Strongly DisagreeItalicized text was reverse coded	Responses were dichotomized into positive responses (Strongly agree/Agree) and neutral/negative responses (Neutral/Disagree/Strongly Disagree), summed and then dichotomized at the midpoint.

School personnel provided the number of enrolled students for each grade level and classroom and the number of classrooms per grade for their respective schools. A sampling scheme was created based on this information and classrooms were assigned to participate by a liaison within each school. Liaisons varied between schools but were frequently administrative secretaries and administrators; none of the liaisons were food service directors or staff. Surveys were administered in classrooms to students in Grades 3 through 12 in SB schools by research assistants. All research assistants completed 3 hours of human subject research compliance training and completed a half day training specific to the tools used in this study. Staff accompanied research assistants to participating sites to observe data collection methods for quality assurance.

Administrator Survey and Variables

The self-report survey that was administered to the school principal or the principal’s designee was developed from previously published studies (Dority, McGarvey, & Kennedy, 2010; Gordon, Crepinsek, Nogales, & Condon, 2007; National Farm to School Network, 2011; Rockman et al., 2010; San Francisco Department of Public Health, 2009). The survey included questions pertaining to the use of the SB, financial support, staff support, barriers and facilitators to implementation, marketing and promotions for the SB, as well as school wellness and/or nutrition policies and education, and lunch period structure. All administrator variables used in the analysis can be found in Table 1.

School Food Environmental Scan

Two trained observers conducted a food environmental scan together in SB schools during lunch period on two different days (12 SB schools; 24 observations). Observers recorded information about the presence of vending machines, nutrition policies, competitive foods (e.g., school stores and a la carte lines), the overall cafeteria environment, SB offerings, and beverages sold in the cafeteria (see Table 1). Observers completed the scan together and came to consensus during the observation, therefore inter-rater reliability was not tested.

Food Service Staff Survey

A short self-administered survey was developed to assess food service staff attitudes and acceptability toward the implementation of the SB. Surveys were completed by 37 food service staff. Table 1 shows the variable created from the food service staff survey.

Data Analyses

Data were analyzed using SAS version 9.3. Descriptive statistics, including frequencies and percentages, were calculated for all variables. Chi-square analysis was used to compare SB users and nonusers for categorical variables. Generalized estimating equation regression methods (proc glimmix) were used to develop a multilevel model including both school-level (e.g., length of lunchtime, SB marketing, vending machines) and individual-level (e.g., sex, food preferences, nutrition knowledge) effects. An independent correlation structure was used to model data. Unadjusted variables with a p value <.20 were considered for inclusion in the adjusted model. A forward stepwise modeling strategy was used. Variables that were p < .05 were retained in the model. Grade was used in final models because grade variable had a lower quasi-likelihood information criterion (QIC) compared to age. Aikake’s information criterion (AIC) and Schwarz Bayesian criteria (SBC) values were examined for goodness-of-fit in selection of a final model. The baseline AIC and SBC values were compared with each predictor model; the final model AIC and SBC values were compared to the previous model using the forward stepwise modeling strategy.

Results

Overall Student Sample

The total response rate for the student survey was 92%. Of the total sample of students (n = 1,334), 142 observations (11.54%) were eliminated from the analyses because the students reported not eating lunch served at school. Another 180 students were excluded because the dependent variable was either missing or could not be determined due to conflicting responses as described above. The total sample size for the subsequent analysis was n = 1,012.

Demographic characteristics of the sample are presented in Table 2. Just more than half of the sample was female (52.04%), and the majority identified as African American (82.65%). Almost 45% of the students were in elementary school, and 41% were in middle school. More than half the sample had low nutrition knowledge (53.4%), and almost 73% of students had low preference for healthy food; the majority of the sample self-identified as an SB user (68.68%).

Table 2.

Descriptive Characteristics of School-Based Salad Bar (SB) Users and Nonusers Among the Overall Sample of Elementary and Secondary Grade Students (N = 1,012).

	SB user, n (%)	SB nonuser, n (%)	Total, n (%)
Variable	N = 695 (68.68)	N = 317 (31.32)	N = 1,012
Sex**
Male	306 (63.35)	177 (36.65)	483 (47.96)
Female	384 (73.28)	140 (26.72)	524 (52.04)
Race
African American	568 (68.52)	261 (31.48)	829 (82.65)
Other	120 (68.97)	54 (31.03)	174 (17.35)
Grade***
Elementary	337 (75.39)	110 (24.61)	447 (44.70)
Middle school	279 (68.22)	130 (31.78)	409 (40.90)
High school	67 (46.53)	77 (53.47)	144 (14.40)
Age***
≤11 years old	360 (75.79)	115 (24.21)	475 (47.64)
12-14 years old	254 (65.13)	136 (34.87)	390 (39.12)
≥15 years old	66 (50.00)	66 (50.00)	132 (13.24)
Healthy food preference**
Low	472 (65.37)	250 (34.63)	722 (72.78)
High	206 (76.30)	64 (23.70)	270 (27.22)
Nutrition knowledge
Low	357 (68.92)	161 (31.08)	518 (53.40)
High	303 (67.04)	149 (32.96)	452 (46.60)

p < .001. ***p < .0001.

Students reported they would be more likely to eat from the SB if there were more food options (31.14%) or if the items tasted better (25.26%). Only 5% of students reported they would eat from the SB if their friends did, and few students reported cost or meal plan as barriers to using the SB (data not shown).

In a multilevel model for the total sample of students, only sex and healthy food preference were significant. No other individual-level or school-level variables were significantly related to SB use. Females were 1.61 times as often to be SB users than males and respondents with high preference for healthy foods were SB users 1.46 times as often as those who preferred less healthy options (Table 3).

Table 3.

Correlates of Salad Bar Use Among the Overall Sample of Elementary and Secondary Grade Students (N = 1,012).

	Unadjusted effects				Adjusted effects
Variable	OR	CI	p value	Variable	OR	CI	p value
Student variables
Sex				Sex
Male	Ref.			Male	Ref.
Female	1.64	1.33-2.02	<.001	Female	1.61	1.17-2.22	.0073
Race
African American	Ref.			—	—	—	—
Other	1.00	0.48-2.06	.991
Grade
Elementary	Ref.			—	—	—	—
Middle school	0.66	0.32-1.34	.251
High school	0.26	0.13-0.55	<.001
Healthy food preference				Healthy food preference
Low	Ref.			Low	Ref.
High	1.75	1.23-2.49	.002	High	1.46	1.00-2.13	.0479
Nutrition knowledge
Low	Ref.			—	—	—	—
High	0.92	0.69-1.23	.573
Administrator variables
Lunchtime
<26 minutes	Ref.			—	—	—	—
≥26 minutes	0.49	0.18-1.32	.1401
Wellness policy
Weak	Ref.			—	—	—	—
Strong	0.98	0.28-3.52	.9971
SB marketing
Weak	Ref.			—	—	—	—
Strong	1.79	0.65-4.90	.2251
Nutrition education
No	Ref.			—	—	—	—
Yes	2.40	0.69-8.33	.1458
Other nutrition programs
No	Ref.			—	—	—	—
Yes	0.84	0.25-2.77	.7515
Environmental variables
SB offerings
<5 items	Ref.			—	—	—	—
≥5 items	1.21	0.39-3.70	.7180
Vending machines
No	Ref.			—	—	—	—
Yes	0.33	0.10-1.09	.0650
Nutrition posters
No	Ref.			—	—	—	—
Yes	1.68	0.54-5.24	.3342
Food advertisements
No	Ref.			—	—	—	—
Yes	0.83	0.19-3.60	.7800
Food service variables
Attitudes of food service
Neutral/negative	Ref.			—	—	—	—
Positive	1.92	0.47-7.75	.3232

Note. OR = odds ratio; CI = confidence interval; SB = salad bar. Unadjusted variables with a p value <.20 were considered for inclusion in the adjusted model. A forward stepwise modeling strategy was used. Variables that were p < .05 were retained in the model. No interactions were significant and therefore were not included in the final model.

Secondary Students Additional Analysis (Grades 7-12)

Because additional questions were asked in the secondary student survey (7th-12th grades), separate analyses were conducted exploring additional variables related to SB use. The total sample size for the subsequent analysis was n = 382.

For these secondary students, 54% were female and 87% were African American. Most of the students were in 7th and 8th grades (62.3%), had low preference for healthy foods (79.4%), and had high nutrition knowledge (86.65%). Few students reported high perceptions of parental/peer modeling of F/V intake (18.43%) and few students reported encouraging their parents/peers to consume F/V (24.32%; Table 4).

Table 4.

Descriptive Characteristics of School-Based Salad Bar (SB) Users and Nonusers Among 7th- to 12th-Grade Students (N = 382).

	SB user, n (%)	SB nonuser, n (%)	Total, n (%)
Variable	N = 231 (60.47)	N = 151 (39.53)	N = 382
Sex*
Male	94 (54.02)	80 (45.98)	174 (45.55)
Female	137 (65.87)	71 (34.13)	208 (54.45)
Race
African American	203 (61.14)	129 (38.86)	332 (87.14)
Other	28 (57.14)	21 (42.86)	49 (12.86)
Grade***
7th-8th	164 (68.91)	74 (31.09)	238 (62.30)
9th-10th	36 (37.89)	59 (62.11)	95 (24.87)
11th-12th	31 (63.27)	18 (36.73)	49 (12.83)
Age**
≤13 years old	107 (70.86)	44 (29.14)	151 (39.63)
14-15 years old	77 (51.33)	73 (48.67)	150 (39.37)
≥16 years old	46 (57.50)	34 (42.50)	80 (21.00)
Healthy food preference*
Low	165 (56.31)	128 (43.69)	293 (79.40)
High	55 (72.37)	21 (27.63)	76 (20.60)
Nutrition knowledge
Low	35 (68.63)	16 (31.37)	51 (13.35)
High	196 (59.21)	135 (40.79)	331 (86.65)
Perceived family/peer modeling
Low	76 (56.30)	59 (43.70)	135 (36.59)
Medium	99 (59.64)	67 (40.36)	166 (44.99)
High	44 (64.71)	24 (35.29)	68 (18.43)
Respondent encouragement**
Low	153 (54.64)	127 (45.36)	280 (75.68)
High	66 (73.33)	24 (26.67)	90 (24.32)

p < .01. **p < .001. ***p < .0001.

In a multilevel model, student encouragement and SB marketing were significantly related to SB use. Students who reported a high frequency of encouraging their friends or family to consume F/V were SB users 2.3 times as often as those who reported low levels of encouragement. Students who attended schools with high levels of SB marketing were SB users almost three times as often as those students who attended schools with low SB marketing (Table 5).

Table 5.

Correlates of Salad Bar Use Among 7th- to 12th-Grade Students (N = 382).

	Unadjusted effects				Adjusted effects
Variable	OR	CI	p value	Variable	OR	CI	p value
Student variables
Sex
Male	Ref.			—	—	—	—
Female	1.69	1.16-2.46	.006
Race
African American	Ref.			—	—	—	—
Other	1.31	0.62-2.73	.478
Grade
7th-8th	Ref.			—	—	—	—
9th-10th	0.28	0.14-0.56	<.001
11th-12th	0.38	0.32-2.05	.665
Healthy food preference
Low	Ref.			—	—	—	—
High	2.17	1.36-3.47	.001
Nutrition knowledge
Low	Ref.			—	—	—	—
High	0.92	0.69-1.23	.573
Respondent encouragement				Respondent encouragement
Low	Ref.			Low	Ref.
High	2.38	1.45-3.90	<.001	High	2.30	1.08-4.90	0.0353
Perceived family/peer modeling
Low	Ref.				—	—	—
Medium	1.12	0.69-1.81	.657
High	1.43	0.81-2.51	.218
Administrator variables
Lunchtime
<26 minutes	Ref.			—	—	—	—
≥26 minutes	0.41	0.16-1.70	.2535
Wellness policy
Weak	Ref.			—	—	—	—
Strong	0.99	0.23-4.31	.9867
SB marketing				SB marketing
Weak	Ref.			Weak	Ref.
Strong	2.41	0.88-6.58	.0778	Strong	2.86	1.09-7.52	0.0371
Nutrition education
No	Ref.			—	—	—	—
Yes	1.03	0.22-4.61	.9688
Other nutrition programs
No	Ref.			—	—	–	—
Yes	1.92	0.49-7.63	.3044
Environmental variables
SB offerings
<5 items	Ref.			—	—	—	—
≥5 items	0.84	0.24-2.92	.7553
Vending machines
No	Ref.			—	—	—	—
Yes	0.39	0.12-4.90	.0865
Nutrition posters
No	Ref.			—	—	—	—
Yes	0.68	0.21-2.16	.4590
Food advertisements
No	Ref.			—	—	—	—
Yes	0.77	0.12-4.90	.7545
Food service variables
Attitudes of food service
Neutral/negative	Ref.			—	—	—	—
Positive	1.72	0.44-6.76	.3902

Note. OR = odds ratio; CI = confidence interval; SB = salad bar. Unadjusted variables with a p value <.20 were considered for inclusion in the adjusted model. A forward stepwise modeling strategy was used. Variables that were p < .05 were retained in the model. No interactions were significant and therefore were not included in the final model.

Discussion

In a model exploring individual- and school-level characteristics related to SB use, two significant variables were identified; female sex and preference for healthy foods were both significantly and positively associated with SB use. For secondary students only, both respondent encouragement and SB marketing were correlates of SB use. The results generated from this study can inform school food service providers, administrators, and other proponents of public health interested in increasing participation in SB programs.

The finding that female students were more likely to be SB users than males is consistent with work previously conducted regarding sex differences in SB use where males were more likely to report that they ate from the SB not very often or never (Andersen et al., 2015; Slusser et al., 2007). Other research has identified that among African American adolescents, females consumed more green salad compared with males (Y. Wang et al., 2007). Research specifically examining consumption of all F/V has also identified that fruit and vegetable intake among youth is greater for females (Caine-Bish & Scheule, 2009; Di Noia & Byrd-Bredbenner, 2013). While research exploring the rationale behind child/adolescent sex differences in F/V intake is sparse, experts hypothesize that, in adults, knowledge for F/V intake is greater in females (Baker & Wardle, 2003). Additionally, attitudes and self-efficacy for F/V consumption are greater among women than men (Emanuel, McCully, Gallagher, & Updegraff, 2012).

One of the strongest predictors of food choice for children and adolescents is preference (Baxter & Thompson, 2002; Blanchette & Brug; 2005; French & Wechsler, 2004; Molaison, Connell, Stuff, Yadrick, & Bogle, 2005). A significant relationship between SB use and preference for healthy food was identified where students who had high preference for healthy foods were more likely to be SB users than those with low preference for healthy food. The majority of the students in the study had a low preference for healthy food, indicating there may be potential for increasing preference for healthy food through taste testing, marketing, and other strategies. Finding opportunities to develop healthy food preferences among children and adolescents may be an important factor in promoting the use of school-based SBs. While this study did not find a significant relationship between SB use and the number of F/V offered, other research suggests the number of SB items offered can have an influence on the amount of F/V selected (Huynh, Pirie, Klein, Kaye, Moore, 2015).

At the secondary student level, respondent encouragement for consumption of F/V was significantly related to SB use. These results suggest that respondents who encourage others to consume F/V may be more inclined to engage in behaviors that increase their own F/V consumption. Research has explored the influence of family and peers on F/V consumption (Darmon & Drewnowski, 2008; Gross, Pollock & Braun, 2010; Pearson, Biddle, & Gorely, 2009), yet little research has examined the association of respondent encouragement of peer/family consumption of F/V.

Secondary students who attended schools with high SB marketing were almost three times more likely to be an SB user than students who attended schools with low SB marketing. An entire body of literature studying the relationships of marketing on children and adolescents has been established (Story, Kaphingst, Robinson-O’Brien, & Glanz, 2008). While most of the literature in this field has observed a relationship between marketing and the consumption of unhealthy food, similar strategies can be used to encourage consumption of healthy foods. The most common types of promotions done for SBs in the present study were taste testing of SB items, posters in the cafeteria, and encouragement from food service staff to use the SB. Research has identified that multicomponent interventions, including promotional signage and taste tests, can result in increased consumption of F/V (French & Stables, 2003; Heim, Stang, & Ireland, 2009). School nutrition service personnel have often used marketing strategies to encourage students to participate in school meals (Briggs, Fleischhacker, & Mueller, 2010). A recent study identified that elementary school students selected a greater number of vegetables from a school-based SB when a banner with cartoon vegetable characters was placed around the SB and when the banner was used in combination with health education segments delivered via television (Hanks, Just, & Brumberg, 2016), demonstrating that traditional health education strategies and environmental marketing strategies can influence behavior. The promotional elements of the SB for secondary students may be an important component of encouraging student use of the SB and food service directors and administrators should consider incorporating such interventions that may increase student use of the SB. Additionally, food service directors and administrator should consider the placement of the SB structure (Adams et al., 2016; Huynh et al., 2015).

Some limitations of this study follow. The design of the study was convenience-based and cross-sectional. Survey data were self-reported. Consequently, the results are subject to the biases inherent with these types of data. Classrooms were selected by school liaisons and may have potential for bias if liaisons selected specific classrooms, but because none of the liaisons were involved in food service, it is unlikely that liaisons selected classrooms based on knowledge of high or low SB use. This study also focused on the use of SB by students, rather than their total consumption of F/V, which ultimately would be more useful in assessing dietary outcomes. The study operates on the assumption that SBs can increase F/V consumption, but further evidence is needed to solidly state if SBs do, in fact, improve F/V consumption. This study was also conducted in an area with a high concentration of African American children and adolescents. These results may not be generalizable to all areas of the country but provide insight within a population at-risk for childhood obesity.

Some strengths of this study lend confidence to the results. Study instruments were pilot tested in a similar student population. Additionally, very few studies have examined simultaneously individual- and school-level factors related to student use in an urban, low-income, predominantly African American population.

This study tested a number of individual- and school-level factors that could be associated with the use of SB in New Orleans. Most variables were not significantly associated with their use, but sex, healthy food preference, respondent encouragement, and school-based marketing of SB were all positively associated with SB use at the secondary level. If confirmed by subsequent studies, such findings suggest the possibility of interesting interventions to promote SB use. Several of the factors related to SB use focus on the individual level of the SEM, for example, sex, respondent encouragement, and healthy food preference. Yet the opportunity for improving SB use through environmental change also exists. The SB itself is an environmental factor that exists within the food service environment. Marketing is also part of the school environment that administrators or food service directors can change by altering promotion strategies; this environmental change strategy was successful in altering SB use in another study (Hanks et al., 2016). While this study examined the effect of sum marketing, several environmental strategies identified by administrators included posters in and outside the cafeteria and displays within the cafeteria. Other studies have identified the placement of the SB may influence the number of F/V children or adolescents take from the SB (Adams et al., 2016; Huynh et al., 2015). In conclusion, both environmental and individual strategies must be considered in order to improve SB use.

Footnotes

Authors’ Note

Work described in this study was conducted at the Prevention Research Center, Department of Global Community Health and Behavioral Sciences, School of Public Health and Tropical Medicine, Tulane University, New Orleans, Louisiana.

The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention, the funding agency.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by Cooperative Agreement #U48DP001948, under the Prevention Research Centers Program, funded by the Centers for Disease Control and Prevention.

References

Adams

M. A.

Bruening

Ohri-Vachaspati

(2015). Use of salad bars in schools to increase fruit and vegetable consumption: Where’s the evidence? Journal of the Academy of Nutrition and Dietetics, 115, 1233-1236. doi:10.1016/j.jand.2015.02.010

Adams

M. A.

Bruening

Ohri-Vachaspati

Hurley

J. C.

(2016). Location of school lunch salad bars and fruit and vegetable consumption in middle schools: A cross-sectional plate waste study. Journal of the Academy of Nutrition and Dietetics, 116, 407-416. doi:10.1016/j.jand.2015.10.011

Adams

M. A.

Pelletier

R. L.

Zive

M. M.

Sallis

J. F.

(2005). Salad bars and fruit and vegetable consumption in elementary schools: A plate waste study. Journal of the American Dietetic Association, 105, 1789-1792.

Andersen

Myers

O’Malley

Mundorf

A. R.

Harris

D. M

Johnson

C. C.

(2015). Adolescent student use of school-based salad bars. Journal of School Health, 87, 722-727.

Baker

A. H.

Wardle

(2003). Sex differences in fruit and vegetable intake in older adults. Appetite, 40, 269-275.

Baxter

S. D.

Thompson

W. O.

(2002). Fourth-grade children’s consumption of fruit and vegetable items available as part of school lunches is closely related to preferences. Journal of Nutrition Education and Behavior, 34, 166-171.

Bazzano

L. A.

Ogden

L. G.

Loria

C. M.

Vupputuri

Myers

Whleton

P. K.

(2002). Fruit and vegetable intake and risk of cardiovascular disease in US adults: The first National Health and Nutrition Examination Survey Epidemiologic Follow-up Study. American Journal of Clinical Nutrition, 76, 93-99.

Blanchette

Brug

(2005). Determinants of fruit and vegetable consumption among 6-12-year-old children and effective interventions to increase consumption. Journal of Human Nutrition and Dietetics, 18, 431-443.

Blomhoff

(1998). Fruit and vegetables in the prevention of cancer. Tidsskr Nor Laegeforen, 118, 3462-3465.

10.

Boffetta

Couto

Wichmann

Ferrari

Trichopoulos

Bueno-de-Mesquita

H. B.

. . . Trichopoulou

(2010). Fruit and vegetable intake and overall cancer risk in the European Prospective Investigation into Cancer and Nutrition (EPIC). Journal of the National Cancer Institute, 102, 529-537.

11.

Briefel

R. R.

Wilson

Gleason

P. M.

(2009). Consumption of low-nutrient, energy-dense foods and beverages at school, home, and other locations among school lunch participants and nonparticipants. Journal of the American Dietetic Association, 109, S79-S90.

12.

Briggs

Fleischhacker

Mueller

C. G.

(2010). Position of the American Dietetic Association, Society for Nutrition Education, and American School Food Service Association: Comprehensive school nutrition servies. Journal of Nutrition Education and Behavior, 42, 360-371.

13.

Caine-Bish

N. L.

Scheule

(2009). Gender differences in food preferences of school-aged children and adolescents. Journal of School Health, 79, 532-540.

14.

Darmon

Drewnowski

(2008). Does social class predict diet quality? American Journal of Clinical Nutrition, 87, 1107-1017.

15.

Di Noia

Byrd-Bredbenner

(2013). Adolescent fruit and vegetable intake: Influence of family support and moderation by home availability of relationships with afrocentric values and taste preferences. Journal of the Academy of Nutrition and Dietetics, 113, 803-808.

16.

Dority

B. L.

McGarvey

M. G.

Kennedy

P. F.

(2010). Marketing foods and beverages in schools: The effect of school food policy on students’ overweight measures. Journal of Public Policy & Marketing, 29, 204-218.

17.

Edmundson

Parcel

G. S.

Feldman

H. A.

Elder

Perry

C. L.

Johnson

C. C.

. . . Webber

(1996). The effects of the child and adolescent trial for cardiovascular health upon psychosocial determinants of diet and physical activity behavior. American Journal of Preventative Medicine, 25, 442-454.

18.

Emanuel

A. S.

McCully

S. N.

Gallagher

K. M.

Updegraff

J. A.

(2012). Theory of planned behavior explains gender difference in fruit and vegetable consumption. Appetite, 59, 693-697.

19.

Epstein

L. D.

Gordy

C. C.

Raynor

H. A.

Beddome

Kilanowski

C. K.

Paluch

(2001). Increasing fruit and vegetable intake and decreasing fat and sugar intake in families at risk for childhood obesity. Obesity Research, 9, 171-178.

20.

French

S. A.

Stables

(2003). Environmental interventions to promote vegetable and fruit consumption among youth in school settings. American Journal of Preventative Medicine, 37, 593-610.

21.

French

S. A.

Wechsler

(2004). School-based research and initiatives: Fruit and vegetable environment, policy, and pricing workshop. American Journal of Preventative Medicine, 39, S101-S107.

22.

Ganann

Fitzpatrick-Lewis

Ciliska

Peirson

L. J.

Warren

R. L.

Fieldhouse

. . . Wolfenden

(2014). Enhancing nutritional environments through access to fruit and vegetables in schools and homes among children and youth: A systematic review. BioMed Central Research Notes, 7, 422. doi:10.1186/1756-0500-7-422

23.

George

S. M.

Park

Leitzmann

M. F.

Freedman

N. D.

Dowling

E. C.

Reedy

. . . Subar

A.F.

(2009). Fruit and vegetable intake and risk of cancer: A prospective cohort study. American Journal of Clinical Nutrition, 89, 347-353.

24.

Gordon

Crepinsek

M. K.

Nogales

Condon

(2007). School Nutrition Dietary Assessment Study-III:Volume I: School foodservice, school food environment, and meals offered and served. Retrieved from http://www.fns.usda.gov/sites/default/files/SNDAIII-Vol1.pdf

25.

Gosliner

(2014). School-level factors associated with increased fruit and vegetable consumption among students in California middle and high schools. Journal of School Health, 84, 559-568. doi:10.1111/josh.12188

26.

Gross

S. M.

Pollock

E. D.

Braun

(2010). Family influence: Key to fruit and vegetable consumption among fourth- and fifth-grade students. Jounral of Nutrition Education and Behavior, 42, 325-341.

27.

Hanks

A. S.

Just

D. R.

Brumberg

(2016). Marketing vegetables in elementary school cafeterias to increase uptake. Pediatrics, 138, e20151420. doi:10.1542/peds.2015-1720

28.

Harris

D. M.

Seymour

Grummer-Strawn

Cooper

Collins

DiSogra

. . . Evans

(2012). Let’s move salad bars to schools: A public-private partnership to increase student fruit and vegetable consumption. Childhood Obesity, 8, 294-297.

29.

Heim

Stang

Ireland

(2009). A garden pilot project enhances fruit and vegetable consumption among children. Journal of the American Dietetic Association, 109, 1220-1226.

30.

Hill

J. O.

Peters

J. C.

(1998). Environmental contributions to the obesity epidemic. Science, 280, 1371-1374.

31.

Hoelscher

D. M.

Day

R. S.

Kelder

S. H.

Ward

J. L.

(2003). Reproducibility and validity of the secondary level school-based nutrition monitoring student questionnaire. Journal of the American Dietetic Association, 103, 186-194.

32.

Huynh

L. M.

Pirie

Klein

E. G.

Kaye

Moore

(2015). Identifying associations between format and placement of school salad bars and fruit and vegetable selection. Journal of Child Nutrition and Management, 39, 1-11.

33.

Kelder

Hoelscher

D. M.

Barroso

C. S.

Walker

J. L.

Cribb

(2005). The CATCH Kids Club: A pilot after-school study for improving elementary students’ nutrition and physical activity. Public Health Nutrition, 8, 133-140.

34.

Kim

S. A.

Moore

L. V.

Galuska

Wright

A. P.

Harris

Grummer-Strawn

L. M.

. . . Rhodes

D. G.

(2014). Vital signs: Fruit and vegetable intake among children—United States, 2003-2010. Morbibity and Mortablity Weekly Report, 63, 671-676.

35.

McLeroy

K. R.

Bibeau

Steckler

Glanz

(1988). An ecological perspective on health promotion programs. Health Education Quartly, 15, 351-377.

36.

Molaison

E. F.

Connell

C. L.

Stuff

J. E.

Yadrick

M. K.

Bogle

(2005). Influences on fruit and vegetable consumption by low-income black American adolescents. Journal of Nutrition Educucation and Behavior, 37, 246-251.

37.

National Farm to School Network, UNC Center for Health Promotion and Disease Prevention. (2011). Farm to school evaluation toolkit. Retrieved from http://www.farmtoschool.org/Resources/UNC%20Eval%20Toolkit.pdf

38.

Ohri-Vachaspati

Turner

Adams

M. A.

Bruening

Chaloupka

F. J.

(2016). School resources and engagement in technical assistance programs is associated with higher prevalence of salad bars in elementary school lunches in the United States. Journal of the Academy of Nutrition and Dietetics, 116(3), 417-426. doi:10.1016/j.jand.2015.10.023

39.

Pearson

Biddle

S. J.

Gorely

(2009). Family correlates of fruit and vegetable consumption in children and adolescents: A systematic review. Public Health Nutrition, 12, 267-283.

40.

Rockman , et al. (2010). Oakland Unified School District Salad Bar Program research study. San Francisco, CA: Center for Ecoliteracy.

41.

San Francisco Department of Public Health, Department of Children Youth and Their Families. (2009). Evaluation report on the Salad Bar Program for San Francisco Unified School District. Retrieved from https://www.sfdph.org/dph/files/hc/HCAgen/HCAgen2009/files407072009/FoodProg07072009.pdf

42.

Slusser

W. M.

Cumberland

W. G.

Browdy

B. L.

Lange

Neumann

(2007). A school salad bar increases frequency of fruit and vegetable consumption among children living in low-income households. Public Health Nutrition, 10, 1490-1496.

43.

Terry-McElrath

Y. M.

O’Malley

P. M.

Johnston

L. D.

(2014). Accessibility over availability: Associations between the school food environment and student fruit and green vegetable consumption. Childhood Obesity, 10, 241-250. doi:10.1089/chi.2014.0011

44.

Turner

Chaloupka

F. J.

(2015). Improvements in school lunches result in healthier options for millions of US children: Results from public elementary schools between 2006-07 and 2013-14 (A BTG Research Brief). Chicago: University of Illinois at Chicago.

45.

Story

Kaphingst

K. M.

Robinson-O’Brien

Glanz

(2008). Creating healthy food and eating environments: Policy and environmental approaches. Annual Review of Public Health, 29, 253-272.

46.

Synder

T. D.

Dillow

S. A.

(2011). Digest of education statistics, 2010. Retrieved from http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=2011015

47.

Wang

Ouyang

Liu

Zhu

Zhao

Bao

F. B.

(2014). Fruit and vegetable consumption and mortality from all causes, cardiovascular disease, and cancer: Systematic review and dose-response meta-analysis of prospective cohort studies. British Medical Journal, 349, g4490. doi:10.1136/bmj.g4490

48.

Wang

Liang

Tussing

Braunschweig

Caballero

Flay

(2007). Obesity and related risk factors among low socio-economic status minority students in Chicago. Public Health Nutrition, 10, 927-938.