It is well known that consumption of sugar-sweetened beverages (SSB) is associated with numerous health risks, including obesity, dental caries, diabetes and cardiovascular risk factors(Reference Malik, An and Willett1–Reference Te Morenga, Howatson and Jones4). SSB consumption among the general population is concerning; however, consumption among adolescents has been identified as particularly problematic globally(Reference Ooi, Wolfenden and Sutherland5). This is due to adolescents’ high consumption levels(6–Reference Bates, Lennox and Prentice8), and the fact that they are within a period where behaviours can become embedded and have long-term impacts on health status(6,9) . Soda is the most commonly consumed SSB among Australians and is the biggest contributor to sugar intake for adolescent males in Australia – the highest SSB consumers(6,10) .
Reducing consumption of SSB is a public health priority. While water is the ideal beverage to replace SSB, a variety of alternative beverages exist and there is risk that consumers may switch to artificially sweetened beverages (ASB) in reaction to health policy and interventions aimed at reducing SSB consumption(Reference Teng, Jones and Mizdrak11). This is concerning given evidence regarding the safety of artificial sweeteners is, at best, inconclusive, with conflicting results regarding health risks associated with consumption of artificial sweeteners(Reference Rios-Leyvraz and Montez12). A precautionary approach is warranted with respect to consumption of ASB so as not to inadvertently encourage consumption with SSB policies and interventions.
The determinants of health behaviour are complex, as evidenced by the multitude of health behaviour theories that offer frameworks to understand the social, environment and individual drivers of behaviour change(Reference Davis, Campbell and Hildon13). Many of these theories acknowledge adequate knowledge of health risks and consequences of behaviours as antecedents to behavioural intentions and behaviour change. Public health interventions often target knowledge and perceptions of a health behaviour and associated health risks, as part of a multi-faceted approach to encourage behaviour change(Reference Asakura, Todoriki and Sasaki14–Reference Avery, Bostock and McCullough16). There is increasing evidence of the effectiveness of campaigns in reducing SSB consumption that target knowledge and educate consumers on health effects(Reference Kite, Grunseit and Bohn-Goldbaum17). Policy makers require insight into the limitations and gaps in population knowledge, and hence the need for interventions. This would provide insight into the specific components of knowledge that may be best targeted to fulfil knowledge gaps, as well as specific groups who may benefit most from intervention.
While consumers, including adolescents, have a right to be informed of nutritional aspects of food and beverages they consume, many have low knowledge of the health effects of SSB consumption and find interpretation of nutrition information challenging(Reference Ni Mhurchu and Gorton18). This limited knowledge may contribute to unhealthy choices and increase health risks, e.g. weight gain or obesity. Adolescents themselves have been found to rate knowledge of health effects of SSB as an important resource for reducing consumption(Reference Kassem and Lee19,Reference Kassem, Lee and Modeste20) . A qualitative study found young school children perceived improving knowledge as important in changing levels of SSB consumption(Reference Battram, Piche and Beynon21). Moreover, nutritional/health knowledge has been associated with better health behaviours in other contexts, such as increased vegetable consumption among school aged children(Reference Asakura, Todoriki and Sasaki14,Reference Kristjansdottir, Thorsdottir and De Bourdeaudhuij22) , good eating habits among adolescents(Reference Pirouznia23), and reduced unhealthy food consumption among children and young adolescents(Reference Grosso, Mistretta and Turconi24). Despite this, the small number of studies that have specifically examined the association between knowledge and SSB consumption among adolescent samples have yielded inconsistent results. For example, two studies found higher nutritional knowledge was significantly associated with lower SSB consumption(Reference Grosso, Mistretta and Turconi24,Reference Irwin, Speechley and Gilliland25) . In contrast, knowledge was not found to be significantly associated with SSB consumption in two studies after adjustment for adolescent and parent demographics(Reference Lundeen, Park and Onufrak26,Reference Nelson, Lytle and Pasch27) . The inconsistency in results across these studies is similar to that found among adult samples; where few studies show evidence of association after adjusting for the effects of demographic characteristics(Reference Gase, Robles and Barragan15,Reference Rampersaud, Kim and Gao28–Reference Kim and House31) .
Adolescents’ knowledge and beliefs regarding SSB and ASB in terms of health and health risks have not been comprehensively assessed, with a notable absence of studies in the Australian context. Internationally, most studies regarding knowledge and understanding of the health risks associated with SSB consumption and nutritional knowledge have been conducted on adult populations(Reference Gase, Robles and Barragan15,Reference Irwin, Speechley and Gilliland25–Reference Rampersaud, Kim and Gao28,Reference Miller, Ettridge and Wakefield30–Reference Park, Onufrak and Sherry33) . Heterogeneity across the small number of adolescent studies in definitions, assessments and methods utilised has led to inconsistent results. Some studies indicate overall moderate-high levels of knowledge are found when general knowledge of health risk is assessed, for example, knowledge of health effects associated with consumption (e.g. diabetes) or overall nutritional knowledge(Reference Irwin, Speechley and Gilliland25,Reference Lundeen, Park and Onufrak26) . More modest levels are found when more specific aspects of knowledge are assessed, for example, high sugar content in SSB, and energy expenditure to eliminate sugar consumed from SSB(Reference Gase, Robles and Barragan15,Reference Rampersaud, Kim and Gao28,Reference Miller, Ettridge and Wakefield30–Reference Park, Onufrak and Sherry33) . Factors such as being older, coming from households with greater incomes, education and non-hispanic have also been found to be associated with knowledge of health effects (weight gain, dental caries) and/or overall nutritional knowledge among two adolescent samples(Reference Irwin, Speechley and Gilliland25,Reference Lundeen, Park and Onufrak26) . While, overall, consumers (adults and children) consider diet soda to be among the unhealthiest options when choosing from non-alcoholic beverages(Reference Battram, Piche and Beynon21,Reference Kim and House31,Reference Thomson, Worsley and Wang34–Reference Munsell, Harris and Sarda37) , studies on children’s perceptions and beliefs regarding ASB are sparse(Reference Battram, Piche and Beynon21,Reference Bucher and Siegrist36) , and have not specifically assessed adolescents’ beliefs regarding health risks of consumption. As noted, consumers may switch to ASB to avoid consuming SSB in reaction to health policy aimed at reducing consumption of SSB(Reference Teng, Jones and Mizdrak11).
There is an absence of studies in Australia that have comprehensively investigated knowledge and beliefs regarding the health risks associated with SSB and ASB consumption among adolescents, including exploring how knowledge and beliefs may vary with consumption and demographic characteristics. Identifying demographic subgroups with lower knowledge can assist in identifying those who will benefit the most from intervention. Notably, we were also unable to find any studies with adolescent samples that explored how beliefs regarding health risks of consuming diet sodas may vary by demographic characteristics, and/or how level of diet drink consumption may vary with such beliefs.
The aim of this study was to investigate knowledge and beliefs associated with SSB and ASB consumption in the Australian adolescent population using data from the 2018 National Secondary Students’ Diet and Activity (NaSSDA) survey, a nationally representative survey of Australian secondary students. The present study focused on knowledge of health effects of soda consumption, and beliefs regarding health risks of diet soda consumption. We refer to beliefs rather than knowledge in relation to diet soda due to the evolving nature of this evidence. Specifically, we sought to determine among Australian adolescents: (i) the level of knowledge of general health risks and nutritional knowledge of soda consumption, and beliefs regarding the health risks associated with diet soda; and (ii) the degree to which level of health knowledge of soda and beliefs regarding diet soda are associated with socio-demographic characteristics, specifically, sex, socio-economic status and age. These demographic characteristics were selected based on evidence of variation in knowledge in previous studies of adolescents and adults(Reference Irwin, Speechley and Gilliland25,Reference Lundeen, Park and Onufrak26,Reference Miller, Ettridge and Wakefield30) . We also explored associations between knowledge and beliefs and consumption of soda and diet drinks.
Methods
Design
The 2018 NaSSDA survey, a nationally representative survey of students in secondary school in Australia (year levels 8 to 11), was the sample source for this study. This survey employs a two-stage probability design with random selection of schools (first stage), and classes within schools (second stage), and stratified by the three education sectors (government, Catholic, independent) within each Australian state or territory. The NaSSDA survey collects data on Australian secondary students’ physical activity and eating behaviours. In 2018, we commissioned a series of additional questions in the NaSSDA survey to assess adolescents’ knowledge of the health risks of soda consumption and beliefs regarding diet soda consumption.
Procedure
Students were asked to complete a web-based survey during a class period. Data were available from 9102 adolescents, yielding a student response rate of 67 %.
Measures
Knowledge of general health risks from regular soda consumption
Based on questions used previously in similar studies(Reference Miller, Ettridge and Wakefield30,Reference Miller, Dono and Wakefield38) adolescents were asked to rate the likelihood of future health problems if they consumed a 600 ml (20 fl oz) soft drink (soda) every day, with available responses of: very likely, somewhat likely, neither likely nor unlikely, somewhat unlikely, very unlikely. This quantity (600 ml) was selected in this study and previous studies with Australian samples as it is a common size of ready-to-consume refrigerated beverages in Australia and adolescents would therefore be familiar with this size(Reference Miller, Ettridge and Wakefield30,Reference Miller, Dono and Wakefield38) .
Knowledge of health effects of soda and beliefs regarding health effects of diet soda
Knowledge of health effects of soda consumption, and beliefs regarding health effects of diet soda consumption, were assessed with questions used in a number of other studies(Reference Lundeen, Park and Onufrak26,Reference Miller, Ettridge and Wakefield30,Reference Miller, Dono and Wakefield38–Reference Boles, Adams and Gredler40) . Adolescents were asked ‘Which of the following diseases or health effects would you associate with drinking soft drinks (soda)/diet soft drinks (soda)?’. Adolescents were provided with a list of options which included: diabetes, weight gain/obesity, heart disease, tooth decay, cancer, other illnesses, no illness or ‘unsure’.
To assess beliefs regarding relative healthiness of diet sodas compared to regular soda, adolescents were asked to rate whether they perceived diet soft drinks [sodas] as more or less healthy than regular soft drinks (or the same). This question was based on similar questions used in studies of Australian adults(Reference Miller, Ettridge and Wakefield30,Reference Miller, Wakefield and Braunack-Mayer39) .
Nutritional knowledge of soda
Three questions assessed adolescents’ knowledge of the nutritional content of soda. Adolescents were asked to indicate, with a 600 ml (20 fl oz) soft drink (soda) in mind, how many teaspoons of sugar were in the beverage, the total calories or kJ (converted to calories for analysis), and the minutes of jogging required to work off the beverage. Responses were categorised into correct ranges based on the contents of popular sodas, allowing variation in estimates of approximately 4–5 teaspoons, ∼50 calories or 15 min (specific ranges are available from results tables). A 600 ml beverage was chosen as the reference point for this question for the same reasons it was a referred to in the assessment of ‘knowledge of general health risks’ in this study, i.e. previous use in other Australian studies, common size for ready consumption in Australia, and adolescents’ familiarity with this size(Reference Miller, Ettridge and Wakefield30,Reference Miller, Dono and Wakefield38) .
Consumption of soda and diet drinks
Consumption was assessed with questions of the main NaSSDA survey that are kept consistent across survey years to enable monitoring of health behaviour(Reference Scully, Morley and Niven41). Specifically, adolescents were asked to indicate how many cups (250 ml/∼9 fl oz) of soft drink (soda) they usually consumed either per day, week or month. Consumption of diet drinks (soft drinks (soda), sports drinks, and cordials) was assessed using the same format. Volumes were then categorised into low consumption (3 cups or less/week), and high consumption (4 cups or more/week) based on limits set in NaSSDA and other consumption studies(Reference Scully, Morley and Niven41).
Demographics
Demographic information was collected regarding adolescents’ sex, age, school year level and main language spoken at home. Given that most adolescents who attend secondary school grades 8 to 11 in Australia are aged between 13 and 16 years, adolescents aged 12 years (0·4 % of the sample) were included with the 13-year-old age group, and adolescents aged 17 years (6·9 %) were included with the 16-year-old age group for the purpose of reporting and analyses. Adolescents were also asked to indicate their postcode to enable the determination of level of disadvantage and remoteness of their residence, based on Socio-Economic Index for Areas (SEIFA)(42) and Australian Statistical Geography Standard Remoteness structure, respectively(43). The SEIFA categorises areas based on a range of socio-economic factors such as employment status, income and household status. Scores were also categorised into tertiles that ranked areas from the most disadvantaged (low score) to least disadvantaged (high score). The Australian Statistical Geography Standard Remoteness structure categorises areas into ‘metropolitan’ to ‘very remote’ based on relative access to services, with categories: ‘major cities,’ ‘inner regional’ and ‘outer regional, remote or very remote.’ Due to the low proportion of adolescents (3·6 %) who resided in remote locations in this survey year, we combined regional and remote categories for reporting, consistent with reporting from an earlier NaSSDA survey(Reference Scully, Morley and Niven41).
Statistical analyses
All analyses were conducted using Stata Statistical software release 15(44). Bivariate associations between knowledge or beliefs and demographics were tested using chi-square, and associations between knowledge or beliefs and consumption variables (regular soda and diet drink consumption; high v. low consumers for each beverage) were tested via a series of multilevel logistic regression models with random effect for school to adjust standard errors for correlations within and between schools. Models also adjusted for school type, state/territory, sex, age and level of disadvantage. OR were reported to describe the odds of being a high soda or diet drink consumer according to knowledge and beliefs regarding these beverages. The primary interest was the associations between knowledge (soda) or beliefs (diet soda) regarding, the health effects of a beverage and consumption of the same type of beverage. However, for completeness we also looked at associations between knowledge of the health effects of consuming soda with diet drink consumption, and vice-a-versa. The number of analyses undertaken and the large data set inflated the risk of Type I error, as such, we employed a conservative probability level (P < 0·001)(Reference Streiner and Norman45,Reference Tabachnick and Fidell46) .
Results
Adolescent characteristics are available from Table 1. There were slightly more females (52·1 %) than males (47·9 %) and adolescents from each of the older age groups (27·7 %–29·1 %) than those in the youngest age group (15·2 %) in the study sample. Most adolescents spoke predominantly English at home (86·1 %) and were from metropolitan areas (65·3 %). The sample consisted of more adolescents from areas of least disadvantage (40·5 %) than that of mid and most disadvantage (33·9 % and 25·6 %, respectively).
Table 1 Characteristics of adolescents
* Only 0.4 % of adolescents were aged 12 years; therefore, 12 and 13 year olds were reported in the same age category.
† Only 6.7 % of adolescents were aged 17 years; therefore, 16 and 17 year olds were reported in the same age category.
‡ Only 3.6 % were classified as remote; therefore, we combined remote and regional for simplicity in reporting.
§ Diet varieties of soft drink (soda), flavoured mineral waters, energy drinks, sports drinks and cordial.
Knowledge and beliefs regarding health risks
Knowledge (agreement) that consuming a 600 ml bottle of soda every day would likely lead to health problems in the future was high overall at 87·4 % (see Table 2). However, knowledge of specific health effects or diseases associated with soda consumption was varied. High proportions of adolescents (70 %+) indicated that tooth decay, weight gain/obesity and diabetes were associated with soda consumption, 56·0 % indicated heart disease, and 19·3 % indicated cancer. Beliefs regarding specific health effects or illnesses associated with diet soda consumption reflected similar patterns to those found with soda and knowledge of health effects, albeit proportions were slightly lower, and 30·6 % reported being unsure, whereas only 12·1 % were unsure of health effects of soda consumption.
Table 2 Variation in knowledge and beliefs according to demographic characteristics
Approx., approximately.
Note. P-values are the result of chi-square tests.
In general, knowledge of the health effects tooth decay, weight gain/obesity, diabetes and heart disease as a consequence of soda consumption was greater among those from areas of least disadvantage, older adolescent age groups, and females (see Table 2; P < 0·001). Knowledge of cancer as a health effect was greater among those from areas of least disadvantage (P < 0·001). The belief that tooth decay, weight gain/obesity and diabetes are associated with diet soda consumption was higher among females, older age groups and those from least disadvantaged areas (P < 0·001). The belief that heart disease and cancer are a health effect of diet soda consumption was higher among older adolescent age groups and those from least disadvantaged areas (P < 0·001).
Almost two-thirds of adolescents (64·3 %) indicated that diet sodas were similar in healthiness to sodas, with 20·3 % reporting they were more healthy and 15·5 % reporting they were less healthy. The proportion indicating diet sodas were healthier than sodas was slightly lower among females than males (18·4 % and 22·4 %, respectively; P < 0·001), and appeared to decline slightly with increasing age (P < 0·001). The proportion reporting diet sodas were less healthy than soda appeared to increase with greater disadvantage (P < 0·001).
Nutritional knowledge
Overall, 22·2 % of adolescents estimated the correct range for number of teaspoons of sugar in a 600-ml bottle of soda, with 36·5 % providing underestimates (see Table 2). Around one-third of adolescents (33·9 %) estimated in the correct range for minutes of jogging to offset the calories/kJs in a 600 ml bottle of soda, and very few adolescents (3·1 %) reported calories in the correct range, with most (82·2 %) indicating they did not know the amount of calories.
Variation in nutritional knowledge of soda was observed according to demographics (see Table 2). Correctly estimating the range for the number of teaspoons of sugar in a 600 ml bottle of soda was slightly higher among the 16–17-year-old age group (25·2 %) than younger age groups (19·5–22·3 %). Underestimating the number of teaspoons of sugar was slightly higher among females (38·1 %) than males (34·8 %; P < 0·001), and the proportion underestimating appeared to decrease as disadvantage increased (P < 0·001). Estimating in the correct range or overestimating minutes of jogging to eliminate sugar consumed from a 600 ml soda was slightly higher among females (35·6 % and 20·1 %, respectively) than males (31·9 % and 15·0 %, respectively; P < 0·001), and conversely, a slightly greater proportion of males underestimated (7·8 % and 11·6 %, respectively). The proportion reporting in the correct range for jogging was higher among the 16–17-year-old age group (36·0 %) compared to other age groups (32·3–33·3 %; P < 0·001), and estimates in the correct range appeared to increase as disadvantage decreased (see Table 2; P < 0·001). Proportions reporting ‘don’t know’ with respect to calories in a 600-ml soda were slightly higher among males (85·4 %) than females (79·5 %; P < 0·001), and showed a gradual decrease with increasing age (85·3 to 79·8 %; P < 0·001), and decreasing disadvantage (87·0 to 78·1 %; P < 0·001).
Results of multilevel logistic regression analyses
Results of multilevel logistic regression analyses are available from Table 3. General knowledge (agreement) that consuming a 600 ml bottle of soda every day would likely lead to health problems in the future was not significantly associated with high soda consumption or high diet drink consumption in these models. With the exception of reporting cancer as a health effect of soda consumption, reporting specific health effects (diabetes, weight gain/obesity, heart disease and tooth decay) from soda consumption was significantly associated with lower odds of higher soda and diet drink consumption (see Table 3; P < 0·001). Those who indicated no illnesses were associated with soda consumption had significantly greater odds of both high soda (OR = 2·23, 95 % CI (1·79, 2·78), P < 0·001) and diet drink consumption (OR = 2·48, 95 % CI (1·90, 3·34), P < 0·001). The same pattern of associations was observed between beliefs regarding health effects of diet soda consumption and consumption of soda and diet drinks (see Table 3).
Table 3 Degree of association between knowledge, beliefs and consumption (soda and diet drinks), adjusted for school type, state/territory, sex, age and level of disadvantage
ref, reference category; approx., approximately.
* P < 0·001.
† Likelihood would have health problems from daily 600 ml soda consumption.
‡ Diet varieties of soft drink (soda), flavoured mineral waters, energy drinks, sports drinks and cordial.
The belief that diet soda was more healthy than regular soda was significantly associated with great odds of both high soda consumption (OR = 1·43, 95 % CI (1·21, 1·68), P < 0·001) and diet drink consumption (OR = 2·31, 95 % CI (1·88, 2·83), P < 0·001). The belief that diet soda was less healthy than regular soda was associated only with high soda consumption (OR = 1·40, 95 % CI (1·17, 1·67)).
Regarding the number of teaspoons of sugar in a 600 ml bottle of soda, compared to those estimating within the correct range, those who responded they did not know were higher consumers of soda (OR = 2·06, 95 % CI (1·68, 2·52), P < 0·001) and diet drinks (OR = 1·72, 95 % CI (1·31, 2·27), P < 0·001). Those who overestimated were also higher consumers of diet drinks (OR = 1·79, 95 % CI (1·30, 2·46), P < 0·001). Compared to those providing approximately correct estimates of jogging to work off a 600 ml bottle of soda, those who underestimated had significantly greater odds of being high soda consumers (OR = 1·62, 95 % CI (1·27, 2·06), P < 0·001), and those indicating they did not know had significantly greater odds of being high soda (OR = 1·92, 95 % CI (1·62, 2·26), P < 0·001) and diet drink consumers (OR = 2·07, 95 % CI (1·64, 2·62), P < 0·001). Estimation of calories in a 600 ml bottle of soda was not associated with consumption of either beverage.
Discussion
This study aimed to explore knowledge and beliefs relating to the health effects of soda and diet soda consumption in the Australian adolescent population, and the degree to which knowledge of soda and beliefs regarding diet soda vary according to sex, socio-economic status, age and consumption of these beverages. The results showed that general knowledge of health risks of soda consumption was high (87 %), consistent with a previous study of adolescents(Reference Lundeen, Park and Onufrak26), and did not vary significantly by demographics, which indicates a near universal understanding among the demographic groups assessed that there is a general health risk associated with regular consumption of soda. Levels of knowledge of the specific health effects associated with soda consumption were relatively lower than general knowledge, ranging from 56 % to 76 % for diabetes, weight gain, tooth decay and heart disease, similar to a previous study of adolescents(Reference Lundeen, Park and Onufrak26). Reporting cancer as a health risk was relatively lower (19 %), indicating this disease is not as commonly perceived to be a risk of consuming sodas, similar to results of studies of adults(Reference Miller, Ettridge and Wakefield30).
Beliefs that the same health effects are a result of diet soda consumption followed a similar pattern, though were somewhat lower such that they ranged from low to moderate (39–56 %) for diabetes, weight gain, tooth decay and heart disease, but were equivalent for cancer (17 %). Two-thirds of adolescents indicated they believed sodas and diet sodas were about the same when it came to ‘healthiness’. Adolescents may acknowledge that while both diet and regular sodas are not healthy per se, consuming sodas results in a greater risk of the health effects assessed in this study compared to diet soda. Adolescents may also consider consuming diet sodas to be associated with different health risks than those assessed in this study. Notably, around one-third indicated they were unsure of health effects associated with diet soda consumption, compared to 12 % that were unsure for soda.
In this study, adolescents had relatively lower nutritional knowledge of soda (nutritional composition, jogging required to eliminate sugar consumed) than general knowledge of the health risks and health effects of soda consumption. These results are similar to other studies that have also found higher general knowledge (e.g. general health risks, such as weight gain/obesity or overall knowledge a drink is sugary), and lower levels of more specific aspects of knowledge (e.g. sugar content, caloric composition, exercise equivalent)(Reference Gase, Robles and Barragan15,Reference Lundeen, Park and Onufrak26,Reference Rampersaud, Kim and Gao28,Reference Miller, Ettridge and Wakefield30–Reference Park, Onufrak and Sherry33) . It is also consistent with qualitative findings among adults, that people have an overall sense that SSB are associated with health risks, but level of understanding of risk and actual physical implications beyond this initial sense is limited(Reference Miller, Braunack-Mayer and Wakefiled47).
Consumption of soda and diet drinks did not vary with general knowledge of future health risk with daily consumption of soda. This is likely due to adolescents’ perceptions of the distal nature of the health risk to themselves. This is similarly found in the area of tobacco among young adults, where long-term health risks of smoking have been found to be less compelling than short-term consequences(Reference Drovandi, Teague and Glass48,Reference Hoek, Hoek-Sims and Gendall49) . However, soda consumption was very much in line with their knowledge of the specific health effects or diseases associated with soda consumption and their nutritional knowledge of soda (teaspoons of sugar or amount of jogging required for 600 ml soda). Having higher knowledge of the specific health effects of soda consumption and nutritional knowledge was significantly associated with lower soda consumption, even after adjustment for sex, age and socio-economic disadvantage. Conversely, reporting no health effects of soda consumption was significantly associated with higher consumption. Overall, this pattern of results was similar to findings of previous studies of adolescents that indicate specific aspects of knowledge tend to be significantly associated with consumption rather than more general knowledge(Reference Grosso, Mistretta and Turconi24,Reference Irwin, Speechley and Gilliland25,Reference Nelson, Lytle and Pasch27) .
Taken together, the results indicate the potential for increasing several aspects of knowledge among adolescents; raising general awareness that daily soda consumption leads to health problems later in life, raising awareness of specific health effects associated with consumption and raising awareness of nutritional aspects of sodas (e.g. teaspoons of sugar, exercise equivalents). While higher soda consumers were less knowledgeable, the extent to which fulfilling these knowledge gaps would facilitate reduced soda consumption would need to be further evaluated and tested.
We identified several demographic subgroups which may assist in determining adolescents who would benefit most from intervention to increase knowledge of the health effects of soda consumption. While general knowledge of health risk did not vary by demographics (as previously described), knowledge of specific health effects and nutritional knowledge (calories, sugar content, exercise equivalent) did. Generally, slightly lower levels of knowledge were reported by males, younger adolescents and those from areas of greater disadvantage. An exception was that underestimating the sugar content of a 600 ml soda increased with age. This is particularly concerning given older adolescents may have more autonomy in purchasing and consuming these drinks. These associations are somewhat similar to those found among adolescents in two previous studies(Reference Irwin, Speechley and Gilliland25,Reference Lundeen, Park and Onufrak26) . The gender difference is also reflective of demographic differences found in adolescents’ consumption behaviour in a 2012–2013 national survey of Australian adolescents(Reference Scully, Morley and Niven41), perhaps suggesting increasing knowledge may reduce consumption and this demographic disparity between males and females. Associations between demographics with knowledge observed among adolescents in this study are also somewhat evident in adulthood, that is, among adults, knowledge has also varied by sex and level of education(Reference Miller, Ettridge and Wakefield30). This is a novel finding and may indicate demographic disparities in knowledge commence when young but continue into adulthood.
To the best knowledge of the authors, this is the first study to explore beliefs regarding health risks of consuming diet sodas among adolescents and how these beliefs vary by demographic characteristics and level of diet drink consumption. Largely, demographic associations with beliefs regarding the health effects of consuming diet sodas were consistent with that found with sodas: lower health risks perceived among males, younger adolescents and those from areas of greater socio-economic disadvantage. A novel finding was that knowledge of soda health effects was significantly associated with diet drink consumption, in the same general pattern as that found with soda consumption across most knowledge indicators. For example, those who perceived specific health effects associated with soda consumption (such as diabetes) were also less likely to be high consumers of diet drinks. Notably, there were some associations that were stronger with diet drink consumption, although causality cannot be assumed due to the cross-sectional design of this study. Diet drink consumption was higher (over twice the odds) among those who considered diet soda to be healthier than soda, but it remains unknown whether health beliefs influenced, or were the consequence of, consuming diet drinks. Diet drink consumption was also higher among those with lower knowledge of the amount of jogging to eliminate energy consumed from sodas (underestimate or did not know), and the amount of sugar in sodas (overestimate or did not know). The results may indicate that some adolescents who are trying to avoid overconsumption of sugar may consume more diet drinks, or that their lack of knowledge is because they are less familiar with regular sodas. These results are interesting in light of substitution effects observed in some studies where consumption of ‘lite’ and no sugar beverages increased when strategies were introduced to reduce SSB consumption(Reference Teng, Jones and Mizdrak11). The findings of this study provide preliminary insight into how beliefs of beverages may relate to consumption. More research is required to determine the effects of educating consumers regarding sugar content of sodas on consumption of both soda and diet drinks. Results may also point to an overall higher level of nutritional knowledge/health literacy among low consumers of both sodas and diet drinks.
Limitations of this study include the cross-sectional design (no causality between knowledge and consumption can be inferred), and issues relating to the self-report nature of the survey rather than measurement of consumption via structured food recall interviews, and the associated potential risk for response biases. The knowledge measures utilised in this survey were all based on existing measures, however, would be strengthened with validation. Knowledge measures were focused on sodas, although it is noted that sugar and nutritional content can vary according to SSB type. It was not feasible to assess knowledge of nutritional content of the many types of SSB available within this study. Some questions of the survey were asked in reference to different volume sizes (e.g. 250 ml or 600 ml). Furthermore, while beliefs were asked about diet soda, consumption was assessed in relation to diet drinks. While ideally, volumes and drink types would be kept consistent across questions, this was due to different origins and purposes of questions. Specifically, consumption questions are important monitoring questions for the NaSSDA survey, and as such are kept historically consistent. Overall, this study presents data from a large sample of adolescents across Australia and explores associations that can inform future interventions and research.
Conclusion
The knowledge gaps identified among adolescents in this study are informative for public health interventions, demonstrating both the need for intervention and aspects of knowledge that could be targeted to assist in reducing consumption. This study also identified specific demographic subgroups who may benefit most from intervention. Addressing these knowledge gaps will assist adolescents to make informed choices and likely encourage reduced consumption in some adolescents in the short term, and it will provide others with a greater understanding of potential harms, which they can draw upon as they become adults. In the context of inconclusive evidence on the health effects of ASB, there is a need to monitor the extent to which adolescents may consider diet drinks as an alternative beverage when reducing SSB consumption in response to public health interventions.
Acknowledgements
Acknowledgements: The authors acknowledge Ms Joanna Caruso for assistance constructing data tables. Financial support: The 2018 National Secondary Students’ Diet and Activity (NaSSDA) survey was principally funded by Cancer Council Australia, and this study was undertaken with additional funding from a Heart Foundation Future Leader Fellowship for CM, and the financial support of Cancer Council’s Beat Cancer Project on behalf of its donors, the State Government through the Department of Health and the Australian Government through the Medical Research Future Fund. The funding sources were not involved in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the article for publication. Authorship: J.D. and C.M. conceptualised survey questions for this study, with assistance from K.E., M.S. and B.M. M.S. and B.M. oversaw data collection, and compiled the national data set. K.E. conducted data analysis, and C.M. and K.E. drafted the manuscript, with assistance from J.D., B.M. and M.S. J.D., C.M., K.E., M.S., B.M. provided final approval of the manuscript version to be published. Ethics of human subject participation: This study was conducted according to the guidelines laid down in the Declaration of Helsinki and all procedures involving research study participants were approved by the Human Research Ethics Committee of Cancer Council Victoria (HREC 1117). Written informed consent was obtained from parents/caregivers and students. Approval was also obtained from education authorities in each state/territory, as well as from school principals.
Conflicts of interest:
C.M., J.D., B.M., M.S. and K.E. declare funding to conduct this survey was received from Cancer Council Australia, with additional funding for this study from a Heart Foundation Future Leader Fellowship for CM, and the financial support of Cancer Council’s Beat Cancer Project on behalf of its donors, the State Government through the Department of Health and the Australian Government through the Medical Research Future Fund.
Open access
