Effect of Removing Chocolate Milk on Milk and Nutrient Intake Among Urban Secondary School Students

Hannah R. Thompson, PhD, MPH¹; Lorrene Ritchie, PhD, RD²; Esther Park, MPH, RD¹; Kristine A. Madsen, MD, MPH¹; Wendi Gosliner, DrPH, RD² (View author affiliations)

Suggested citation for this article: Thompson HR, Ritchie L, Park E, Madsen KA, Gosliner W. Effect of Removing Chocolate Milk on Milk and Nutrient Intake Among Urban Secondary School Students. Prev Chronic Dis 2020;17:200033. DOI: http://dx.doi.org/10.5888/pcd17.200033.

PEER REVIEWED

Abstract

Introduction

Schools across the United States have removed sweetened, flavored milk from cafeterias to reduce students’ sugar consumption and improve their health. However, evidence on the impact of the removal is limited. We examined the effect of a policy that removed chocolate milk from secondary schools on students’ milk consumption and estimated milk-related nutrient intake.

Methods

We collected data on milk selection and consumption during 1 lunch period in 24 California public secondary schools pre-policy (N = 3,158 students in 2016) and post-policy (N = 2,966 students in 2018). Schools had a student population that was 38% Asian and 29% Latino, with 63% qualifying for free or reduced-price meals. We used linear mixed effects models to assess changes in milk selection and waste, and we estimated related changes in added sugars, calcium, protein, and vitamin D consumed from milk.

Results

The proportion of students selecting milk declined 13.6%, from 89.5% pre-policy to 75.9% post-policy (95% CI for difference, 10.8% tο 16.4%), but the proportion of milk wasted remained stable (37.1% vs 39.3%; 95% CI for difference, −0.2% to 4.6%). Although average per-student milk consumption declined by less than 1 ounce per student (from 4.8 oz to 3.8 oz; 95% CI for difference, −1.1 oz to −0.7 oz), we observed no significant reductions in average per-student intake of calcium, protein, or vitamin D from milk. Estimated added sugars from milk declined significantly, by 3.1 grams per student (95% CI, −3.2 g to −2.9 g).

Conclusion

Removing chocolate milk modestly reduced student milk consumption without compromising average intake of key milk-related nutrients, and consumption of added sugars from milk declined significantly. Secondary schools should consider removing chocolate milk to support healthy beverage consumption.

Top

Introduction

Sugar-sweetened beverages (SSBs) are one of the leading sources of added sugar in youth’s diets (1), and consumption is causally linked to the development of obesity, cardiovascular disease, type 2 diabetes, and dental caries (2–4). Sixty-five percent of adolescents aged 12 to 19 years reported consuming at least one SSB on a given day (5). Furthermore, racial/ethnic and income-related disparities in SSB intake and obesity persist (5–7). Leading public health and medical organizations, including the World Health Organization, American Heart Association (AHA), and American Academy of Pediatrics, recommend a drastic reduction in SSB consumption to improve health outcomes and reduce obesity (8,9).

Students in the United States consume nearly half of their calories from school meals; thus, schools offer an opportune setting for improving their diets (10). Interventions that target the school food environment, rather than focusing on individual dietary behaviors, have been found most effective at positively affecting dietary intake (11). Sweetened, flavored milks remain the only SSBs available in schools as part of federally reimbursable school meal programs (12). As such, policies to remove chocolate milk in schools are one environmental approach increasingly used to reduce youth SSB consumption. Although chocolate milk has the same key nutrients as plain, unflavored milk, it has up to twice as much sugar.

Some stakeholders worry that eliminating chocolate milk from schools will decrease students’ consumption of essential nutrients, such as calcium and vitamin D, and increase milk waste (13–16). Findings from a few studies validate these concerns (15,17). For example, one study in 11 Oregon elementary schools found that total milk sales decreased and milk waste increased after a chocolate milk removal policy was implemented (18), and research in 4 elementary and middle schools in an urban Massachusetts district found a significant increase in milk waste immediately after chocolate milk was removed from cafeterias (19).

Other studies showed positive effects of chocolate milk removal policies (20,21). Research in a single rural Oregon elementary school reported a significant decrease in consumption of added sugars and only a negligible decrease in calcium and protein intake following chocolate milk removal (21). However, findings from a single elementary school may not be generalizable to larger, urban school districts. Little is known about the effect of removing chocolate milk from middle and high schools, but understanding this effect is important, because teenagers consume more SSBs than younger children (22). Moreover, research is also lacking among low-income students of color, who typically consume more SSBs than their high-income white peers.

During the 2017–18 school year, a large, diverse, urban California school district introduced a policy to remove chocolate milk from all middle and high schools as part of a broad district-wide strategy to reduce students’ intake of added sugar. The purpose of our study was to assess the effect of the school district’s chocolate milk removal policy on student milk selection, waste, and overall consumption and to estimate attendant changes in calcium, protein, vitamin D, and added sugar intake among racially/ethnically and socioeconomically diverse secondary school students.

Top

Methods

Setting and participants

We collected data on milk selection and waste as a part of the Multi-Pronged Intervention to Increase Secondary Student Participation in School Lunch (MPI) study (23). All participating schools (n = 24) were traditional middle schools (grades 6–8, n = 12) and high schools (grades 9–12, n = 12) in the San Francisco Unified School District, an urban district in Northern California. The MPI study employed a quasi-randomized design in which 6 middle and 6 high schools received a 3-pronged intervention (lunchroom redesign, additional points of lunch sale [mobile cart and vending machine], and teacher outreach) over a 3-year study period (school years 2015–16 through 2017–18) and 12 schools served as comparisons and did not receive the intervention. We found no evidence of effect modification by MPI intervention school versus comparison school status (P > .20); therefore, data for all 24 schools are presented together. This research was approved by the University of California, Berkeley, Committee for the Protection of Human Subjects and the Research, Planning, and Accountability Department of the San Francisco Unified School District.

During the 2017–18 school year, independent of the MPI study, the school district implemented a policy that removed chocolate milk from schools. This policy, part of a broader district initiative to reduce students’ sugar intake (including removing fruit juice from schools and cutting down on the amount of added sugar in foods served as part of school meals), began in middle schools in August 2017 and in high schools in January 2018. High school implementation was delayed by one semester after surveys and interviews with high school students suggested that additional education and engagement would lead to a smoother transition. Pre-policy, schools in the district offered students the choice of either low-fat (1%) plain, unflavored milk or nonfat chocolate milk. Post-policy, students could select either low-fat (1%) or nonfat plain, unflavored milk. All milk came in 8-ounce cartons (weighing 257 grams when full and 11 grams when empty). Each 8-ounce serving of milk (regardless of flavor or fat content) contained 300 milligrams of calcium, 9 grams of protein and 3.1 micrograms of vitamin D; each serving of nonfat chocolate milk contained 11.3 grams of added sugar.

Data collection

We collected data during 1 lunch period per school in the spring (February–April) of 2016 and 2018. All students participating in the school lunch program were eligible to participate. At the beginning of each lunch period, a maximum of 220 numbered trays were distributed to students waiting in lines for school lunch (students could consent or decline participation at this time). A maximum of 220 trays per school was based on the prior school year’s average school lunch participation to ensure that most or all students could participate in the study if they chose to. Students were selected as they lined up at the beginning of the lunch period until no more students were available to participate. Once students selected their lunch, a researcher recorded the tray number, whether the student selected milk, and the type of milk selected. Students were instructed to return the tray with leftover food, milk, and trash to the research team. When trays were returned, another researcher recorded the corresponding tray number on milk cartons; cartons were weighed to the nearest gram on an OXO Good Grips 11-pound–capacity food scale, and carton numbers and weights were recorded. For each carton of milk returned, the grams of milk wasted were calculated as the full carton weight minus weight when returned, and grams of milk consumed were estimated as 246 grams (calculated weight of milk) minus grams of milk wasted. The proportion of milk wasted was calculated as grams wasted divided by 246 grams.

Statistical analysis

We used linear mixed effects models to assess differences pre-policy (2016) and post-policy (2018) in 1) the proportion of students who selected milk (the number of students who selected milk divided by the total number of students participating in the study), 2) the proportion of milk wasted (the mean of proportion of milk wasted per carton), and 3) average ounces of milk consumed per student (sum of total grams of milk consumed divided by number of participating students). Additional models estimated the average amounts of calcium, protein, vitamin D, and added sugar consumed from milk per student across the study population. Estimates were based on estimates of the mean number of ounces of milk consumed per student. Additional models examined interaction by school type (middle vs high) and MPI intervention status (intervention vs control school) for each outcome. All models included a random effect for school and were controlled for school-level demographic characteristics (downloaded from the California Department of Education (24), including school type (middle vs high), total school enrollment, student enrollment by race/ethnicity, and proportion of students who qualified for free or reduced-price meals. We used Stata/SE 15.1 (StataCorp LLC) to perform analyses.

Top

Results

Students in participating schools were diverse (38% of students were Asian and 29% Latino) and predominantly low-income (63% of students qualified for free or reduced-price meals) (Table 1). Pre-policy, 3,158 students participated in lunch data collection across all 24 schools (mean, 132 per school). Post-policy, 2,966 students participated in lunch data collection (mean, 124 per school). Before policy implementation, 90% of participating students selected milk (low-fat, plain, unflavored milk, 48%; nonfat chocolate milk, 52%) (Table 2). After policy implementation, 76% selected milk (nonfat milk, 10%; low-fat, plain, unflavored milk, 83%; cartons missing milk type, 7%).

Results from adjusted models showed that significantly fewer students selected milk after the chocolate milk removal policy was implemented: the proportion of students selecting milk dropped 13.6% (from 89.5% pre-policy to 75.9% post-policy; 95% CI for difference, −16.4% to −10.8%) (Table 2). However, the proportion of milk wasted did not change significantly (37.1% vs 39.3%; 95% CI for difference, −0.2% to 4.6%).

Overall milk consumption declined by 0.9 ounces per student (4.8 oz vs 3.8 oz; 95% CI for difference, −1.1 oz to −0.7 oz). In addition, small but nonsignificant decreases in the average amount of nutrients consumed from milk per student participating in the study were observed: calcium (from 189 g to 182 g; 95% CI for change, −13.7 g to 0.7 g); protein (from 5.7 g to 5.5 g; 95% CI, for change −0.4 g to 0.02 g); and vitamin D (from 1.95 mcg to 1.88 mcg; 95% CI for change, −0.14 mcg to 0.01 mcg). Added sugar consumption from milk significantly declined; after policy implementation, students consumed an average of 3.1 fewer grams of added sugars from milk (95% CI, −3.2 g to −2.9 g).

We saw no evidence of effect modification by school type (middle vs high) for either the proportion of students who selected milk (P = .11) or the proportion of milk consumed (P = .76).

Top

Discussion

Ours was the first study to our knowledge to examine the effect of a district-level chocolate milk removal policy on milk selection and consumption among racially/ethnically and socioeconomically diverse middle and high school students. During the year when chocolate milk was removed from school meals, we observed a 14% decline in the proportion of students selecting milk with school lunch. However, among the nearly 75% of students who did select milk at follow-up, we saw no increase in milk waste post-policy. Despite a slight (<1 oz) decrease in student milk consumption across the population after the policy was passed, student intake of milk’s key nutrients (calcium, protein, and vitamin D) at the population level was not reduced. Furthermore, students’ consumption of added sugar from milk declined significantly. These findings suggest that policies removing chocolate milk from school meals could improve dietary intake and health among urban secondary school students.

Although the decline in overall milk consumption we saw in our study was small, it may, nonetheless, overstate any negative effects of removing chocolate milk. Follow-up data from our study were collected during the same year as implementation of the chocolate milk removal policy, which provided only short-term results of changes associated with the policy. Other studies have similarly demonstrated a decrease in milk purchasing during the same year chocolate milk was removed from school meals (18,25). However, in a study conducted in 2 kindergarten-to-8^th-grade New England schools educating primarily low-income students (20), researchers saw student milk selection increase from 52% immediately after a policy removing flavored milk was implemented to 72% 2 years post-policy (they did not have data on milk selection pre-policy). This suggests that over time, students’ acceptance of such policies may increase, and exclusive availability of plain, unflavored milk in schools may be normalized, leading to a rebound of milk selection and consumption. Repeating measures in our study schools after the policy becomes more established would provide a better understanding of the long-term effect of chocolate milk removal from these secondary school cafeterias.

Although chocolate milk constituted 52% of the milk selected pre-policy, in our study nearly 75% of students continued to select milk after chocolate milk was no longer available. Additionally, overall milk consumption declined by less than 1 ounce per student, and milk waste did not increase, suggesting that offering milk with added sugar is not necessary to promote healthy beverage intake in secondary schools. This is important both for students’ health and also for reducing school food waste and its related negative environmental impacts.

It is important to note that the school district made a substantial effort to ensure buy-in for the policy before implementation, including a phased roll-out that was sensitive to schools’ and students’ needs (ie, delaying implementation in high schools until the policy could be clearly explained and justified to staff and students). Although few studies have described the effect removing chocolate milk had on food waste, Hanks et al (18) reported an increase in milk waste after chocolate milk was removed from 11 Oregon elementary schools. Thus, anticipating opposition to such policies and providing clear and open communication may protect against milk waste and reduced milk consumption.

Furthermore, and not surprisingly, consumption of added sugar from milk significantly declined once chocolate milk was removed from school lunch. Davis et al (21) similarly saw average added sugar intake from milk decrease after chocolate milk was removed from school lunches, with negligible average reductions in calcium consumption. Hanks et al (18) also demonstrated a reduction in students’ consumption of added sugar after chocolate milk was removed from 11 Oregon elementary schools.

Avoiding SSBs is an important strategy to prevent childhood obesity (26). AHA recommends that youth consume only 6 teaspoons (25 grams) of added sugar a day (27). In our study, we estimated a 3.1 gram reduction in added sugar per student per day after the policy, which represents more than 12% of the AHA’s recommended 25-gram limit of added sugar. If students drink a full 8-ounce carton of chocolate milk, they will consume more than a third of the recommended daily limit of added sugar in 1 sitting. Furthermore, calories consumed in liquid form have been shown to have weak satiety properties and stimulate poor energy compensation compared with calories consumed from solid foods (28). Given that adolescents’ SSB consumption is typically higher than recommended, particularly among low-income youth and youth of color (22), removing chocolate milk from school meals could decrease added sugar and SSB consumption, with potentially important implications for public health equity.

Our study had several limitations. First, findings from our sample of 24 secondary schools in 1 urban school district may not be generalizable to rural or less socioeconomically or racially/ethnically diverse districts. Additionally, all of these schools were in a district working to reduce students’ sugar consumption, and half of them also were participating in an intervention that changed cafeteria environments and worked to increase students’ participation in school meals. However, no significant differences in outcomes were observed between schools originally assigned to intervention or control conditions. Second, although data collection days were carefully selected in collaboration with the school district (and menus at each school were aligned to be as similar at baseline and follow-up as possible), factors such as what foods were served for lunch on each data collection day could confound these findings. Furthermore, students did not choose milk in isolation, but rather as a component of a full meal, and we did not assess how this policy may have influenced students’ overall nutrient intake from school lunch. We were also unable to assess whether students compensated by consuming SSBs brought from home or purchased outside of school (42% of study schools — 10 high schools — had open campus policies), though our prior work showed that after SSBs were removed from low-income middle and high schools in California, students did not compensate by drinking SSBs outside of school (29). Our data collection was limited to lunch periods; analogous data were not collected at school breakfasts. Additionally, our study included only schools implementing the chocolate milk removal policy and lacked comparison schools. We do not have data on the number or characteristics of students who declined to participate in the study, which affects the generalizability of these findings. Moreover, with only 24 schools, our study was not adequately powered to examine differences by school-level characteristics. Finally, a lack of data on individual student characteristics precluded a closer examination of student-level factors that could affect milk selection and consumption.

Immediately after chocolate milk was removed from school cafeterias in our study, fewer urban middle and high school students selected milk during school lunch, though milk waste did not increase. Although average milk consumption declined slightly across the population, consumption of calcium, protein, and vitamin D from milk remained stable, while consumption of added sugar from milk decreased significantly. Our findings suggest that secondary schools should consider removing chocolate milk from school lunch to support students in consuming healthier beverages.

Top

Acknowledgments

This work was funded by the US Department of Agriculture, National Institute of Food and Agriculture, (grant no. 2015-68001-23236), Technology and Design Innovation to Support 21st Century School Nutrition. The authors thank the school district’s Student Nutrition Services department, district leadership, and the participating schools, without which this study would not have been possible. Authors declare no competing financial interests exist. No copyrighted materials were used in this article.

Top

Author Information

Corresponding Author: Hannah R. Thompson, PhD, MPH, University of California, Berkeley, School of Public Health, 2115 Milvia St, Third Floor, Berkeley, CA 94704. Telephone: 510-642-3589. Email: thompsonh@berkeley.edu.

Author Affiliations: ¹University of California, Berkeley, School of Public Health. ²Nutrition Policy Institute, Berkeley, California.

Top

References

1. Bailey RL, Fulgoni VL 3d, Cowan AE, Gaine PC. Sources of added sugars in young children, adolescents, and adults with low and high intakes of added sugars. Nutrients 2018;10(1):E102. CrossRef PubMed
2. Malik VS, Popkin BM, Bray GA, Després JP, Hu FB. Sugar-sweetened beverages, obesity, type 2 diabetes mellitus, and cardiovascular disease risk. Circulation 2010;121(11):1356–64. CrossRef PubMed
3. Hu FB. Resolved: there is sufficient scientific evidence that decreasing sugar-sweetened beverage consumption will reduce the prevalence of obesity and obesity-related diseases. Obes Rev 2013;14(8):606–19. CrossRef PubMed
4. Hu FB, Malik VS. Sugar-sweetened beverages and risk of obesity and type 2 diabetes: epidemiologic evidence. Physiol Behav 2010;100(1):47–54. CrossRef PubMed
5. Bleich SN, Vercammen KA, Koma JW, Li Z. Trends in beverage consumption among children and adults, 2003–2014. Obesity (Silver Spring) 2018;26(2):432–41. CrossRef PubMed
6. Ogden CL, Fryar CD, Hales CM, Carroll MD, Aoki Y, Freedman DS. Differences in obesity prevalence by demographics and urbanization in US children and adolescents, 2013–2016. JAMA 2018;319(23):2410–8. CrossRef PubMed
7. Madsen KA, Weedn AE, Crawford PB. Disparities in peaks, plateaus, and declines in prevalence of high BMI among adolescents. Pediatrics 2010;126(3):434–42. CrossRef PubMed
8. Levine JA, Eberhardt NL, Jensen MD. Role of nonexercise activity thermogenesis in resistance to fat gain in humans. Science 1999;283(5399):212–4. CrossRef PubMed
9. Rosenbaum M, Leibel RL. The physiology of body weight regulation: relevance to the etiology of obesity in children. Pediatrics 1998;101(3 Pt 2):525–39. PubMed
10. Woo Baidal JA, Taveras EM. Protecting progress against childhood obesity — the National School Lunch Program. N Engl J Med 2014;371(20):1862–5. CrossRef PubMed
11. Kubik MY, Lytle LA, Hannan PJ, Perry CL, Story M. The association of the school food environment with dietary behaviors of young adolescents. Am J Public Health 2003;93(7):1168–73. CrossRef PubMed
12. United States Department of Agriculture. Final rule: nutrition standards in the National School Lunch and School Breakfast programs. January 26, 2012. https://www.fns.usda.gov/school-meals/fr-012612. Accessed December 30, 2019.
13. Murphy MM, Douglass JS, Johnson RK, Spence LA. Drinking flavored or plain milk is positively associated with nutrient intake and is not associated with adverse effects on weight status in US children and adolescents J Am Diet Assoc 2008;108(4):631–9. CrossRef PubMed
14. Frary CD, Johnson RK, Wang MQ. Children and adolescents’ choices of foods and beverages high in added sugars are associated with intakes of key nutrients and food groups. J Adolesc Health 2004;34(1):56–63. CrossRef PubMed
15. Nicklas TA, O’Neil C, Fulgoni V 3d. Flavored milk consumers drank more milk and had a higher prevalence of meeting calcium recommendation than nonconsumers. J Sch Health 2017;87(9):650–7. CrossRef PubMed
16. Cruz G. US schools’ war against chocolate milk. Time magazine. December 18, 2009. http://content.time.com/time/nation/article/0,8599,1948865,00.html. Accessed November 18, 2019.
17. DeLany JP, Harsha DW, Kime JC, Kumler J, Melancon L, Bray GA. Energy expenditure in lean and obese prepubertal children. Obes Res 1995;3(Suppl 1):67–72. CrossRef PubMed
18. Hanks AS, Just DR, Wansink B. Chocolate milk consequences: a pilot study evaluating the consequences of banning chocolate milk in school cafeterias. PLoS One 2014;9(4):e91022. CrossRef PubMed
19. Cohen JF, Richardson S, Parker E, Catalano PJ, Rimm EB. Impact of the new U.S. Department of Agriculture school meal standards on food selection, consumption, and waste. Am J Prev Med 2014;46(4):388–94. CrossRef PubMed
20. Schwartz MB, Henderson KE, Read M, Cornelius T. Student acceptance of plain milk increases significantly 2 years after flavored milk is removed from school cafeterias: an observational study. J Acad Nutr Diet 2018;118(5):857–64. CrossRef PubMed
21. Davis MM, Spurlock M, Ramsey K, Smith J, Beamer BA, Aromaa S, et al. Milk options observation (MOO): a mixed-methods study of chocolate milk removal on beverage consumption and student/staff behaviors in a rural elementary school. J Sch Nurs 2017;33(4):285–98. CrossRef PubMed
22. Han E, Powell LM. Consumption patterns of sugar-sweetened beverages in the United States. J Acad Nutr Diet 2013;113(1):43–53. CrossRef PubMed
23. Machado S, Ritchie L, Thompson H, Reed A, Castro AI, Neelon M, et al. Multi-pronged intervention to increase secondary student participation in school lunch: design and rationale. Contemp Clin Trials 2019;78:133–9. CrossRef PubMed
24. California Department of Education. Dataquest State Education Data Reporting. http://data1.cde.ca.gov/dataquest/. Accessed November 19, 2019.
25. Henry C, Whiting SJ, Phillips T, Finch SL, Zello GA, Vatanparast H. Impact of the removal of chocolate milk from school milk programs for children in Saskatoon, Canada. Appl Physiol Nutr Metab 2015;40(3):245–50. CrossRef PubMed
26. Te Morenga L, Mallard S, Mann J. Dietary sugars and body weight: systematic review and meta-analyses of randomised controlled trials and cohort studies. BMJ 2012;346:e7492. CrossRef PubMed
27. Jenco M. AHA: Limit children’s sugar consumption to 6 teaspoons per day. AAP News August 23, 2016. https://www.aappublications.org/news/2016/08/23/Sugar082316. Accessed December 2, 2019.
28. Mourao DM, Bressan J, Campbell WW, Mattes RD. Effects of food form on appetite and energy intake in lean and obese young adults. Int J Obes 2007;31(11):1688–95. CrossRef PubMed
29. Woodward-Lopez G, Gosliner W, Samuels SE, Craypo L, Kao J, Crawford PB. Lessons learned from evaluations of California’s statewide school nutrition standards. Am J Public Health 2010;100(11):2137–45. CrossRef PubMed

Top

Tables

^a Values are percentage (standard deviation) unless otherwise indicated.
^b P value calculated from t tests.
^c The proportion of students who do not identify as African American, Asian, Latino, or white are not included; therefore, percentages do not total 100.

^a Data calculated from linear mixed effects models with a random effect for school, adjusted for secondary school type (middle or high, total school enrollment, student enrollment by race/ethnicity, and proportion of students eligible for free or reduced-price meals).
^b P < .05.

Top