Introduction

While the prevalence of overweight and obesity have risen globally, these challenges are particularly acute in the Caribbean.[1–3] Low levels of physical activity, a lack of health-promoting nutrition policies, and a food environment oversaturated with nutrient-poor products all contribute to an elevated prevalence of obesity among Caribbean populations.[1,4] Commercial determinants, such as the excessive import of ultra-processed foods and sugar-sweetened beverages, have also contributed to a regional rise in nutritional non-communicable disease, including cardiovascular disease and diabetes.[1,5,6] Furthermore, poverty and inequalities in education have been shown to drive higher rates of obesity across the region.[7,8]

In Jamaica, specifically, investigations have been conducted to describe population level obesity prevalence and its determinants. From 2000 to 2020, the country experienced more than a 10% increase in adult obesity prevalence.[6,9] Among children and adolescents, rates of both overweight and obesity have more than doubled in the same time period.[9] The most recent data suggests a 13.9% prevalence and 25.3% among children and adults, respectively.[10]

Researchers have tried to understand key influencers of childhood obesity that are embedded into the spaces in which children live, learn and play. At the child and household level, there is evidence that parental income, child food and beverage knowledge, and parental guidance all influence children’s dietary habits. At the school-level, where children spend much of their time, the large presence of ultra processed food and drinks, limited outdoor recreational space, and influence of external vendors have been shown to play a significant role.[11]

As a subset of ultra processed foods, sugar-sweetened beverages (SSB) are a major contributor to obesity and cardiometabolic conditions, and the consumption of such products by adolescents are higher in Latin American and Caribbean region than any other area of the world.[12–16] Since 1990, Jamaican youth have consistently consumed, on average, more than 10 SSB servings per week, an intake greater than the regional mean.[15,17] Previous work has shown that children and adolescents who consume SSB are more likely to be obese and consume other types of ultra-processed products.[18,19]

Data have shown that SSB excise taxes, food labeling policies, and other policies to decrease the presence of SSB in food environments can reduce children’s consumption, and consequently reduce excess weight.[20–23]

In June 2025, Jamaica’s Parliament approved the new national school nutrition policy, jointly developed by the Ministry of Education, Skills, Youth and Information and the Ministry of Health and Wellness.[24]. One focus of this policy will be regulating nutritional standards within school and managing the internal and external food environments.[25] Defined as all of the spaces, infrastructure, and conditions inside and around school premises where food is available, obtained, purchased, and/or consumed, school food environments (SFE) are instrumental in determining what and how children consume when at school.[26]

In this secondary analysis, we will be assessing the associations between our primary exposures, child demographics and school environment factors, and our primary outcomes, sweetened beverage, unsweetened beverage, and water intake. This investigation will allow for an analysis across multiple levels of determinants, as well as identification of where school nutrition interventions may be most effective. This secondary analysis will further assess which components of Jamaican SFE are most influential on primary children’s beverage consumption patterns, helping to inform and enrich school nutrition policies and guidelines.

Methods

Analytic sample

A total of 1745 primary school children were invited to participate in anticipation of a < 50% response rate. From these responses, 907 children returned surveys, reflecting a 52.0% response rate. The analytical data for this paper was restricted to those with complete data for all included covariates (dropped 66) as well as children’s dietary intake surveys which reflected a school day consumption (dropped 148). This resulted in an analytic sample of data from 693 children (Fig 1), representing an estimated 40.1% of the total student population of the study schools (2214) and 3.66% of children in grades 3–5 across the two parishes (24785). Children’s consumption volumes were capped at 2500 mL, as volumes greater than this could be considered over-reporting by the study population.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Eligibility Exclusion Criteria.

https://doi.org/10.1371/journal.pone.0354352.g001

Results

Descriptive statistics were done on both exposures and outcomes of this study as shown in Table 1. Child demographics and school environment covariates are described by percents and frequencies, while school level indices are described using means and standard deviation.

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Descriptive statistics of Analytical Sample (N=693).

https://doi.org/10.1371/journal.pone.0354352.t001

The sample reflects near equal distribution of children’s sex, and children with parents above and below the 2018 national poverty line. Slightly less than half of children in the study (44.6%) have parents whose highest level of education is secondary school, and at least 10% of the study population has parents whose highest-level education are one of the following: No Formal School/Primary School (12.8%), Technical/Vocational School (15.3%), and Tertiary, College/University (21.8%).

In the internal school environment, schools had an average of 10.8 food marketing elements which promoted 12.7 sweetened beverage brands and 3.5 plain water brands. Children more frequently attended schools with outdoor drinking apparatuses (77.5%) than they attended schools with indoor drinking water options (15.3%). Regarding school meals, children, on average, attended schools with more than 8 school-meal related equipment, spaces, and indoor school-sponsored vending. External to school campuses, school vendors were the food source which were cited most frequently in the audit (73.4%) compared to food stalls (33.0%), cook shops (34.2%), supermarkets (25.5%), and restaurants/fast food (8.7%).

The beverage intake outcomes of the study are depicted in Fig 2, with the proportion of non-consumers shown in black, as well as quartiles among consumers of each beverage type noted. Across all three beverage types, there is a positive skew, indicating a trend of relatively low amounts of consumption with a small number of children consuming large amounts.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Distributions of children’s consumption of sweetened beverages (2A), unsweetened beverages (2B), and plain water (2C).

N = 693. Non-consumers and quartiles are depicted, and the percentage of non-consumers are included above the black bar in each plot.

https://doi.org/10.1371/journal.pone.0354352.g002

Among consumers of sweetened beverages, 25% consumed less than 400 mL and 75% consumed less than 1000 mL. Relative to a standard can of sweetened beverage (330–355 mL), this is more than one and three cans of drinks, respectively. A relatively low number of children (50, 7.2%) reported consuming either no sweetened beverage.

With regard to the distribution of unsweetened beverages and water consumption, there is a much greater proportion of non-consumers, 19.2% and 20.9, respectively. Given the inclusion of water as an unsweetened beverage and the low consumption of other non-water unsweetened beverages, these two distributions are similar. Among consumers of both types of beverages, 50% consumed less than 500 mL.

As seen in Table 2, as parental education level increases through vocational school training, children’s consumption of sweetened beverages increased. Though children having parents with a college/university education level (the highest measured level) had a lower consumption of sweetened beverages compared to children of parents without formal schooling. With respect to school environment measures, the presence of indoor drinking water apparatuses and the presence of supermarkets near schools both are associated with a significant increase in intake (p < 0.05). The presence of urban residences, while insignificant, was also associated with an increase in sweetened beverage intake.

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Margins Post-Test Estimation and Wald Test Results.

https://doi.org/10.1371/journal.pone.0354352.t002

Our model showed no single environmental covariate being significantly associated with unsweetened beverage intake. Among child-level demographics, we found that female students in our sample consumed more unsweetened beverages than their male counterparts.

The water intake model closely resembled the unsweetened beverage model, indicating a low consumption of non-water unsweetened beverages. Similarly, being female was significantly associated with an increase in intake. Having at least one parent or guardian of the highest education level was shown to have a weak correlation with consumption of more water (p < 0.1).

Wald Tests showed that, child demographic factors were jointly significant in being associated with all three beverage types, while grouped external school environment measures were proved only significant with students’ sweetened drink intake.

The table in S1 Table provide the results for each part of the two-part models, revealing some potential factors that may drive the probability of consuming select beverages. We found that a classroom’s policy of preventing drinking water in the classroom perfectly predicted a student’s probability of consuming sweetened beverage.

Discussion

This cross-sectional study sought to measure the relationships between child and school level influencers on Jamaican children’s beverage intake, providing evidence that child demographic and external school environment measures have significant associated with volumes of beverages consumed.

Evaluating sweetened and unsweetened beverage intake, the results showed varying associations of modeling covariates on consumption. At the child level, gender seemed to play a role, with female school children drinking more water and other unsweetened beverages than male school children. Although, the two groups drank indistinguishable amounts of sweetened beverages, aligning with other regional and global trends.[15] Parental education appeared to be loosely correlated to consumption with children of parents at the highest level of education drinking more water than children of parents with no formal schooling, though this correlation was not significant at conventional levels. Across all three beverage types, child-level measures were shown to jointly contribute to beverage intake model fit.

Internal and external school metrics showed varied associations on beverage intake. The presence of indoor drinking water and supermarkets near schools proved to be the only significant covariates associated with sweetened beverage intake, both correlated with increases. Wald Tests provided sufficient evidence to reject the possibility that external school environment covariates are jointly insignificant in contributing to the sweetened beverage outcome, meaning that their interaction may have a role in children’s choice to drink these sweetened drinks.

Distinguishing between consumers and non-consumers helps to interpret the differences between sweetened and unsweetened beverage consumption patterns. More than a fifth of the children included in the sample did not consume any water during the school day, a stark comparison to the 7.2% of sweetened beverages non-consumers. This difference in overall drinking habits may be related to students’ access to water during the school day. The initial probit model defines, unsurprisingly, that students who were permitting drinking water in the classroom did drink water more often during school days. Yet, even more striking, was that every student who was in a classroom in which drinking water was prohibited consumed sweetened beverages. Generally, children are encouraged to bring water containers into the classroom, though the results of school surveys show that they may be unable to refill these containers throughout the day.

Interactions between classroom rules, school wide policies, and the internal school environment could also contribute to overall patterns in beverage choice. Nearly all students in the study sample were able to consume water in the classroom, yet only 15.3% of them attended schools with indoor water access points (see Table 1). Previous work has shown associations between district-wide policies on beverage consumption, particularly that of sweetened drinks. Policies restricting both the presence of these drinks and the advertisement of them within schools has been shown to be associated with lower odds of regular sweetened drink consumptions.[33] Furthermore, our study found similar results to others in that an increase in the presence of healthier beverage alternatives was not associated with a decrease in sweetened beverages consumption.[33,34]

Previous work among global populations have shown that a range of interventions can impact children’s beverage consumption in school.[35–38] Interventions, including both environmental and educational types, have been shown to decrease SSB intake, with combined approaches being particularly effective.[38] While children in this study seem to have a high level of beverage knowledge, an intervention that engages parents and teacher may help strengthen the influence of beverage knowledge on actual beverage choice. Previous work has shown that engaging parents can significantly decrease the consumption of sodas and other SSBs.[39,40]

Beyond the physical environment, social environments at schools have also been shown to influence young children’s dietary patterns. A peer mediated behavioral study in the Caribbean showed that peer influencers can help reduce other school children’s consumption of SSB, while also potentially increasing their water intake.[41] Children and adolescents’ behavior can be influenced more by their friends or peers, especially in school settings, than their parents, and thus is important to account for in any future school-based interventions in this setting.[42]

A recently approved national school nutrition policy in Jamaica seeks to regulate nutritional standards within school and manage the internal and external food environments. Though the presence of external food sources was low in this survey, the quality of their provisions is largely unregulated. With a large focus oversight of school food vendors, this policy may help improve the quality of beverages students consume. The country’s school beverage guidelines, which increasingly limits the amount of allowed sugar since 2018, are not necessarily supported by adequate policy structures and has allowed for the provision of unhealthy beverages both on and near school campuses. A regional public health collaborative has provided a framework for these limitations, encouraging not only the prohibition on unhealthy drinks, but also the availability and promotion of healthier options.[43]

There is precedence for formal policies that limit the sales of SSBs in other Caribbean countries, namely in Bermuda, Barbados, and Trinidad and Tobago. In 2006, Bermuda banned the sale of SSBs in schools and in 2018, implemented a tax on sugary drinks and food in 2018.[44,45] This resulted in a nearly half of the population consuming fewer taxes goods, including drinks. Barbados implemented a similar SSB tax in 2015 which led to both a decrease in SSB sales, as well as an increase in unsweetened beverage purchases.[22] In 2017, St. Lucia’s Ministry of Health and Ministry of Education moved forward with a ban on the sale of SSBs in schools, though there is limited evidence on the effectiveness of this ban.[44]

Given the low presence of policies regulating nutrition and dietary components in Jamaican schools and the challenges of enforcing these types of policy interventions, a synergistic legislative framework is needed to drive effective change.[4,46,47] A suite of such policies must be multisectoral, improving the quality of food and beverage marketing in schools, offering healthy nutrition education opportunities, and modeling desired consumption behavior.[46,48,49] Programs improving children’s diet in school will have the greatest benefit if closely tied to other related policies and legislation, such as those touching upon social protections, health, agriculture, and urban food planning.[50] While individual policies may falter or be difficult to enforce, a coherent set of legislative implementation can lead to better student health outcomes. [51,52]

Conclusion

Through a cross-sectional analysis of Jamaican school children’s sweetened and unsweetened beverage consumption, we described the differences and potential child-level and school-level influencers of in-school drinking habits. Gender was shown to be significantly associated was differences in unsweetened beverage intake, while the presence of indoor drinking water apparatuses and supermarkets near schools were associated with variations in sweetened beverage consumption. Collectively, external school environment factors and child-level demographics improved the fit of sweetened and unsweetened models, respectively. The ability of students to drink water in the classroom and schools likely influenced sweetened beverage intake. Schools can improve their student overall beverage intakes by implementing additional policies to decrease the presence and marketing of SSBs in and around schools. These types of policies and other interventions to improve children’s beverage intake should be considered and incorporated in the implementation of Jamaica’s nation school nutrition policy and other intervention to better school children’s nutritional health.

Supporting information

Acknowledgments

We thank Ms. Annalee Gray-Brown and Dr. Alexandra Ross for their assistance in this effort.