Sugar-sweetened beverage consumption from 1998-2017: Findings from the health behaviour in school-aged children/school health research network in Wales.

To date no study has examined time trends in adolescent consumption of sugar-sweetened beverages and energy drinks, or modelled change in inequalities over time. The present study aimed to fill this gap by identifying historical trends among secondary school students in Wales, United Kingdom. The present study includes 11-16 year olds who completed the Health Behaviour in School-aged Children (HBSC) survey and the Welsh School Health Research Network (SHRN) survey between 1998 to 2017. Multinomial regression models were employed alongside tests for interaction effects. A total of 176,094 student responses were assessed. From 1998 to 2017, the prevalence of daily sugar-sweetened beverage consumption decreased (57% to 18%) while weekly consumption has remained constant since 2006 (49% to 52%). From 2013 to 2017, daily consumption of energy drinks remained stable (6%) while weekly consumption reports steadily decreased (23% to 15%). Boys, older children and those from a low socioeconomic group reported higher consumption rates of sugar-sweetened beverages and energy drinks. Consumption according to socioeconomic group was the only characteristic to show a statistically significant change over time, revealing a widening disparity between sugar-sweetened beverage consumption rates of those from low and high socioeconomic groups. Findings indicate a positive shift in overall consumption rates of both sugar-sweetened beverages and energy drinks. Adolescents from a low socioeconomic group however were consistently shown to report unfavourable sugar-sweetened beverages consumption when compared to peers from high socioeconomic group. Given the established longer term impacts of sugar-sweetened beverage and energy drink consumption on adolescent health outcomes, urgent policy action is required to reduce overall consumption rates, with close attention to equity of impact throughout policy design and evaluation plans.

Introduction

Consumption of sugar-sweetened beverages (SSBs), including energy drinks (EDs) represents a significant public health problem, with consumption rates linked to an increased health risk of type II diabetes [1] and dental erosion [2]. Soft drinks contribute an estimated 40% of sugar intake among adolescents [3], of which EDs make up an increasing proportion [4]. In 2017, one study found that 95% of EDs would receive a ‘red’ (high) label for sugars per serving [5]. This poses a concern as dietary patterns track from adolescence into adulthood [6], and this period represents a crucial phase in the life-course for the development of various diseases [7].

SSBs, including EDs are widely available and promoted. Marketing strategies have actively targeted certain communities, for example, using outdoor advertisements within deprived areas, and increased television exposure among young people within minority ethnic and low-income communities [8]. It is estimated that 1 billion litres of soft drinks are produced globally each year [9] with the soft drinks industry contributing £11 billion to UK economic growth [10]. The EDs market is estimated to be worth over £2 billion in the UK [11] and $50 billion globally, with a projected annual growth rate of 3.5% between 2015 and 2020 [12]. Concerns around ED consumption primarily relate to the high caffeine content, however some large ED cans may contain up to 21 teaspoons of sugar [13], over three times the daily recommendation [14].

While calls for a reduction in SSB consumption date back to 1942 in the United States (US), only in recent years have EDs received increasing attention from policy-makers and health experts [5, 15]. Some countries have banned sales of EDs and others have introduced sales and labelling regulations [11]. At present there remains no UK-wide legislation relating specifically to EDs. In 2018, most major UK supermarket chains enforced a ban on ED sales to under 16s [16] and 2019 saw a ban on ED sales to under 16s in all NHS sites in Scotland [17]. An ED ban for under 16s was proposed by the UK government in 2018, but has not yet been executed [18]. Instead, devolution in each UK nation has resulted in a number of consultations. The Scottish Government recently closed a consultation on ending the sale of EDs to under 16’s [19]. In December 2019, the Welsh Government set out plans to ban sales of EDs to all children and young people by 2030 as part of a nationwide strategy [20]. Thus, present UK sales remain unregulated with a voluntary code of practice to avoid deliberate marketing of products to under 16’s [11], and stakeholders and health experts across all levels continue to call on industry and government to introduce a ban on such sales.

To drive product reformulation and reduce sugar content, a two-tiered Soft Drinks Industry Levy (SDIL) was introduced in the UK in 2018 [21]. Including EDs, the policy concerns the production and importation of SSBs and aims to incentivise manufacturers to lower sugar content though the lowering of tax rates, referred to as the ‘Sugar Tax’. Recent UK findings show this is having a favourable impact on the sugar content in drinks [22].

Trends in SSB consumption have been documented widely across young people in the US [23] and recently Denmark [24], revealing declines in daily consumption over time. A 2017 report on daily SSB consumption rates, involving 32 European countries also noted a decline over time, yet results were limited to two time-points, 2002 and 2014, with no ED data [25]. As such, no study to date has examined periodic time trends in SSB and ED consumption among UK nations in the lead up to the introduction of the SDIL.

Little attention has been paid specifically to EDs, with only one US-based study to date exploring adolescent ED consumption trends with no time trend reports by demographic characteristics [26].

The aim of this study is to examine the consumption frequency of SSBs and EDs among 11–16 year olds over time. With use of national data collected between 1998 and 2017, we examined overall consumption and reports among sociodemographic subgroups. Data available for SSB consumption spanned a 20-year period (1998–2017) and ED consumption across five years (2013–2017).

Materials and methods

Study sample

Student self-report data from the Health Behaviour in School-aged Children (HBSC) survey and the School Health Research Network (SHRN) surveys in Wales, from 1998 to 2017 were used. Surveys were conducted approximately every two years from 1998 to 2017. Data are appended over the years to create a repeated cross-sectional dataset as in previous studies [27]; Data on SSBs were available from 1998–2017, and EDs from 2013–2017. The HBSC survey, a collaborative cross-national survey, is administered every four years and currently involves 50 countries and regions across Europe and North America. The SHRN survey is administered every two years and is based on the HBSC survey allowing integration of the two surveys every four years. Over-time the SHRN survey sample size has grown due to the increasing number of schools in Wales agreeing to conduct the survey. Details on study sampling strategies and procedures can be accessed elsewhere [28, 29].

Sociodemographic characteristics

Gender (response options: ‘Boy’ and ‘Girl’) and school year were reported in all survey years. School year and corresponding age groups were: Year 7 (age 11–12), Year 8 (age 12–13), Year 9 (age 13–14), Year 10 (age 14–15) and Year 11 (age 15–16). An indicator of socioeconomic status (SES) was available from 2002, using the Family Affluence Score (FAS) [30, 31] which comprised measures of: car and computer ownership, bedroom occupancy and family holidays. From 2013 onwards, two additional measures (dishwasher and bathroom ownership) were included [32]. Scores for each of the four/six survey items were summed for a total score, whereby a higher score indicated greater affluence. This score was split at the median in each survey year to achieve ‘low’ and ‘high’ SES.

Definitions of outcome variables

SSB consumption

A question on SSB consumption, included in every survey year, asked; ‘How many times a week do you usually drink Coke or other soft drinks that contain sugar?’ (response options: ‘Never’, ‘Less than weekly’, ‘weekly’, ‘2–4 times a week’, ‘5–6 times a week’, ‘Daily’ and ‘More than once a day’). In the first two survey years, ‘Never’ and ‘Less than weekly’ formed one category. For each survey, responses were recoded into a three-category variable indicating: ‘Never or less than weekly’ (includes ‘Never’ and ‘Less than weekly’), ‘Weekly’ (includes ‘Weekly/Once a week’, ‘2–4 times a week’ and ‘5–6 times a week’) and ‘Daily or more’ (includes ‘Daily’ and ‘More than one a day’).

ED consumption

A question on ED consumption, included in 2013, 2015 and 2017, asked ‘How many times a week do you usually drink energy drinks (such as Red Bull, Monster, and Rockstar)?’ (Response options: ‘never’, ‘less than once a week’, ‘once a week’, ‘2–4 days a week’, ‘5–6 days a week’, ‘once a day, every day’ and ‘every day, more than once’). Responses were recoded to form a three-category variable; ‘Never or less than weekly’ (included ‘Never’ and ‘Less than weekly’), ‘Weekly’ (included ‘Weekly/once a week’, ‘2–4 times a week’ and ‘5–6 times a week’) and ‘Daily or more’ (included ‘Daily’ and ‘More than one a day’).

Inclusion and exclusion criteria

For SSB data, 1.6% (n = 2,840) were missing. Following introduction in 2013, missing data for ED questions ranged between 0.2% and 2.2%. Analyses focused on students in Years 7 to 11 (i.e. aged 11–16). In the years 1998–2002 and 2006, Years 8 and 10 were not available at the time of analysis. Hence as a sensitivity analysis, analyses were conducted with only Year 7, 9 and 11. As trends did not differ with either method, data for all year groups were retained. In 2017, gender included an additional response category (‘prefer not to say’). As there was only one year of data on this group, the students selecting this response were set as missing (2%; n = 2,261).

Ethical approval and consent to participate

Schools signed and returned a commitment form to participate in the HBSC study; parents were sent information sheets and had the option of withdrawing their child from the study. Before the survey, participants were assured of anonymity and confidentiality and asked to provide written active assent. All students had the opportunity to withdraw from data collection at any time. The survey was approved by Cardiff University Social Sciences Research ethics committee.

Statistical analysis

Statistical analysis was conducted using STATA 15. Descriptive data are presented as frequencies and percentages. Analyses of SSB and ED consumption over time were performed using multinomial regression models with time (variable year) included as a covariate to measure the effects of time on consumption. Multinomial logistic regression models examined associations between sociodemographic characteristics and the three-category variables for SSB and ED consumption (reference category—‘Daily consumption or more’). Models were first tested for SSBs and EDs separately using the predictors of gender, school year, SES, and survey year. Coefficients are reported as relative risk ratios (RRR’s).

Interaction effects with the variable ‘year’ and the characteristics of school year, gender and SES were also investigated to estimate change over time. All variables were mean-centred where applicable, i.e. not binary indicators, to limit multi-collinearity in analysis. Models were performed separately to test for change over time among characteristics of interest (e.g. gender). Interactions were estimated using multinomial regression and predictive margins, and graphed using these estimates. Models were conducted using complete case analysis; other options were considered–see S1 File.

Results

The total sample comprised 176,094 student responses (S1 Table provides a breakdown across each survey wave). Sample demographics (Table 1) were evenly split in terms of gender and SES. The largest school year group was Year 7 and the smallest Year 10. For SSBs, approximately one in two students reported weekly consumption (52%) overall, whereas just over a quarter never consume them (27%), and around one fifth reported daily or greater consumption (22%). For EDs, most students reported never consuming them, or consumption less than weekly (77%), whereas approximately 1 in 6 (17%) reported weekly consumption. Only 6% reported ED consumption daily or more. Cross-tabulations (S2 Table) showed a relationship between SSB and ED consumption, with daily SSB consumption being largely related to daily ED consumption, and vice-versa for never consumption of SSBs and EDs (χ2 = 24000, p<0.05, n = 140,470).

Table 1

Sample characteristics of the study participants (11–16 years) between 1998 and 2017 (n = 176,094).

Variable	N		Total (n, %)	Missing (n, %)
Gender	173 957			2337 (1%)
		Boy	85 919 (49%)
		Girl	88 038 (51%)
School year	176 094			0 (0%)
		Year 7	40 358 (23%)
		Year 8	34 467 (20%)
		Year 9	39 200 (22%)
		Year 10	30 570 (17%)
		Year 11	31 499 (18%)
Socioeconomic Status	161 779			14315 (8%)
		Low	78 880 (49%)
		High	82 899 (51%)
Sugary drink use	173 254			2,840 (2%)
		Never, or less than weekly	46 257 (27%)
		Weekly use	89 228 (52%)
		Daily use or more	37 769 (22%)
Energy drink use	141 154			34,940 (20%)
		Never, or less than weekly	109 208 (78%)
		Weekly use	23 937 (17%)
		Daily use or more	8 009 (6%)

Time trends

Time trend analysis results are shown in Figs 1 and 2. The proportion of students consuming SSBs daily decreased steeply from 2000 to 2006, appearing to plateau from 2009, dropping from 57% to 18% across the time series in 2017. Similarly, the proportion reporting never or less than weekly consumption increased four-fold from 7% in 1998 to 29% in 2017. Weekly SSB consumption increased steadily since 2000 and has remained constant since 2006 at 49% to 52% in 2017. More detailed analysis showed that the ‘Once a week’ and ‘2–4 times’ a week were mostly attributable to the increase (See S13 Table). Regression analyses indicate that relative to the highest consumption category (i.e. daily consumption), consumption of SSBs never or less than weekly increased significantly over time (RRR 1.08, p<0.05, CI 1.08–1.09), while weekly consumption also increased compared to daily consumption (RRR 1.05, p<0.05, CI 1.04–1.05). Hence, findings indicate an overall trend toward declining SSB consumption over time, indicated by increasing movement of the population toward lower consumption categories.

Fig 1

Reported SSB consumption between 1998 and 2017.

Fig 2

Reported ED consumption between 2013 and 2017.

Since 2013, daily ED consumption has remained stable at 6%. Weekly ED consumption has steadily decreased from 23% in 2013 to 15% in 2017. Conversely, a steady increase in reports of never or less than weekly consumption is estimated, 71% in 2013 vs. 79% in 2017. Upon further inspection, this steady increase reflects the increase in reports of ‘Never’ consumption over the years (see S15 Table). Regression analyses indicated that relative to daily consumption, reports of never or less than weekly remained unchanged over time relative to daily consumption (RRR 1.00, p<0.05, CI 0.98–1.02) although weekly consumption decreased over time relative to daily consumption (RRR 0.89, p<0.05, CI 0.87–0.91); note that when year was treated as categorical, RRR’s showed an increase for never and less than weekly consumption (2015 RRR = 1.05, 2017 RRR = 1.03), and a decrease for weekly consumption (2015 RRR = 0.92, 2017 RRR = 0.68). Hence, while very regular consumption remains stable, the proportion of adolescents consuming EDs has fallen.

SSB consumption and demographics

Gender

Against the reference category of daily consumption, girls were more likely to consume SSBs never or less than weekly compared to boys (RRR 1.75, p<0.05, CI 1.70–1.80). Boys were substantially less likely than girls to report never or less than weekly consumption (22% vs 31%) and slightly more likely to report daily consumption (24% vs 20%; S2 Table).

Socioeconomic status

Lower SES groups were less likely to report never or less than weekly consumption (RRR 0.68, p<0.05, CI 0.66–0.70) and weekly consumption (RRR 0.78, p<0.05, CI 0.76–0.80) when compared to daily consumption. Lower SES groups were more likely to be in the daily consumption group over-time compared to high SES groups (e.g. 21% vs 16% in 2017; S4 Table).

School year

Older students were less likely to consume SSBs never or less than weekly, compared to daily consumption; Year 9s and 10s were the least likely to report SSB consumption as never or less then weekly (RRR 0.70, p<0.05, CI 0.67–0.73/0.74). Year 8s had the highest likelihood of consuming SSB’s never or less than weekly (RRR 0.86, p<0.05, CI 0.82–0.90). For weekly SSB consumption, a similar pattern was observed but Year 10’s and 11’s had the highest consumption risk (RRR 0.79, p<0.05, CI 0.76–0.83) and Year 8’s had the lowest (RRR 0.95, p<0.05, CI 0.91–0.99). For the full model, see Table 2. Over time, 21% of Year 7’s consumed SSBs daily, compared to 24% of Year 9’s, and 25% of Year 11’s (S6 to S10 Tables).

Table 2

Multinomial regression of SSB consumption and sociodemographic characteristics with daily use as the reference category (n = 157,564).

				Confidence Intervals
	RRR	Std. Err	p	Upper bound	Lower Bound
Daily use (base outcome)
Never or less than weekly use
Gender
Girl	1.75	0.03	<0.001	1.70	1.80
School Year
Year 8	0.86	0.02	<0.001	0.82	0.90
Year 9	0.70	0.02	<0.001	0.67	0.73
Year 10	0.70	0.02	<0.001	0.67	0.74
Year 11	0.72	0.02	<0.001	0.68	0.75
Socioeconomic Status
Low	0.68	0.01	<0.001	0.66	0.70
Year	1.08	0.00	<0.001	1.08	1.09
Weekly use
Gender
Girl	1.11	0.01	<0.001	1.08	1.14
School Year
Year 8	0.95	0.02	0.02	0.91	0.99
Year 9	0.81	0.02	<0.001	0.77	0.84
Year 10	0.79	0.02	<0.001	0.76	0.83
Year 11	0.79	0.02	<0.001	0.76	0.83
Socioeconomic Status
Low	0.78	0.01	<0.001	0.76	0.80
Year	1.05	0.00	<0.001	1.04	1.05

ED consumption and demographics

When comparing ED consumption to daily consumption, girls were more likely to report consumption as never or less than weekly compared to boys (RRR 1.85, p<0.05, CI 1.77–1.95), however this was not apparent for weekly consumption reports (RRR 1.00, p = 0.91, CI 0.94–1.05) (Table 3). Over time, 7% of boys consumed EDs daily compared to 4% of girls; with little change across years (see S3 Table).

Table 3

Multinomial regression of ED consumption and sociodemographic characteristics with daily use as the reference category (n = 135,712).

				Confidence Intervals
	RRR	Std. Err	p	Upper bound	Lower Bound
Daily use (base outcome)
Never or less than weekly use
Gender
Girl	1.85	0.05	<0.001	1.77	1.95
School Year
Year 8	0.81	0.03	<0.001	0.75	0.87
Year 9	0.61	0.02	<0.001	0.57	0.66
Year 10	0.58	0.02	<0.001	0.53	0.62
Year 11	0.66	0.03	<0.001	0.61	0.72
Socioeconomic Status
Low	0.67	0.02	<0.001	0.63	0.70
Year	1.00	0.01	0.99	0.98	1.02
Weekly use
Gender
Girl	1.00	0.03	0.91	0.94	1.05
School Year
Year 8	1.04	0.05	0.43	0.95	1.13
Year 9	0.97	0.04	0.49	0.89	1.06
Year 10	0.93	0.04	0.09	0.85	1.01
Year 11	1.00	0.05	0.93	0.91	1.09
Socioeconomic Status
Low	0.83	0.02	<0.001	0.78	0.87
Year	0.89	0.01	<0.001	0.87	0.91

Lower socioeconomic groups were less likely to report never or less than weekly consumption compared to daily consumption (RRR 0.67, p<0.05, CI 0.63–0.70). Similarly, they were less likely to report weekly consumption compared to daily consumption (RRR 0.83, p<0.05, CI 0.78–0.87). High socioeconomic groups were more likely to report never or less than weekly consumption of EDs over time compared to low SES (80% vs 75%); see S5 Table.

For school year, Year 10s were the least likely to report never or less than weekly ED consumption (RRR 0.58, p<0.05, CI 0.53–0.62), with Year 8s being the most likely (RRR 0.81, p<0.05, CI 0.75–0.87). For weekly consumption Year 8s were more likely to report ED consumption (Year 8: RRR 1.04, p = 0.43, CI 0.95–1.13 vs. Year 10: RRR 0.93, p = 0.09, CI 0.85–1.01), although differences were not significant. Over time, 4% of Year 7’s consumed ED’s daily, compared to 6% of Year 9’s, and 6% of Year 11’s (see S6, S8 and S10 Tables).

Demographic patterning of SSB and ED consumption overtime

The association between SES and SSB consumption changed over time. Where models used the reference category of ‘daily consumption’, lower socioeconomic groups were less likely to respond ‘never or less than weekly’ in SSB consumption compared to higher socioeconomic groups (RRR 0.91, p<0.05, CI 0.88–0.95). Descriptive data (S4 and S5 Tables) indicate that in 2002, consumption was very similar for higher and lower SES groups (e.g. 37% daily consumption for both groups), and while consumption has fallen for both groups, it has done so fastest in higher SES groups, leading to increased inequality (i.e. in 2017, 21% of young people from poorer families report daily consumption vs 16% of those from more affluent families). Regression models show that whilst both groups increased their reports of ‘never or less than weekly’ SSB consumption, the rate of increase was slower for low socioeconomic groups, indicating greater movement toward non-consumption in more affluent groups (S1 Fig). Likewise, lower socioeconomic groups were less likely to consume SSBs weekly compared to high socioeconomic groups (RRR 0.92, p<0.05, CI 0.89–0.95). S2 Fig shows that the socioeconomic pattern has changed over time in this model. In 2002, lower socioeconomic groups were more likely to consume SSBs weekly, however by 2004 this reversed with higher socioeconomic groups being more likely to using them weekly (compared to daily). As a result, the gap between low and high socioeconomic groups has widened, with lower socioeconomic groups being more likely to consume SSB’s daily compared to weekly. This trend has changed in most recent years, with weekly consumption decreasing since 2015 for both socioeconomic groups.

The absence of time varying effects via other characteristics such as gender, school year and SES suggests that individual characteristics of SSB consumers have remained relatively stable over time, with RRR’s being 1.00–1.01; likewise, for EDs consumption, with wider confidence intervals observed. For interaction estimates see S12 Table.

Discussion

Present findings provide a profile of national trends over the past two decades of self-reported SSB and ED consumption among adolescents in Wales. This is the first large study to examine such consumption rates over time and by multiple demographic characteristics.

Overall consumption trends

Almost 80 years since the first calls for a reduction in SSBs [5, 15] our findings provide an encouraging outlook on trends in SSB consumption among adolescents. We found that consumption reports since 1998 indicate a positive shift for daily SSB consumption with approximately 40% fewer adolescents reporting daily consumption in 2017 compared to 2000. A noticeable upward trend was observed for the number reporting never or less than weekly SSB consumption. ED consumption was not measured prior to 2013, but showed small decreases over time with one in four young people using EDs at least weekly in 2013. Recent findings in Denmark also displayed a decrease in daily SSB consumption between 2002 and 2018, albeit lower prevalence rates were observed in 2018 among Danish adolescents at 6.4% [24]. Compared to other HBSC countries, daily SSB consumption rates are somewhat lower than Malta, Belgium and Bulgaria at 34–37% [25]. The overall ED consumption rates for Wales are comparable to findings among Canadian adolescents [33] while lower rates were previously reported among a Korean population (11.4%) [34]. Comparable weekly SSB consumption rates were recently reported among Australians aged 15 or older [35]. While the sampling of differing populations, data collection tools and analyses makes direct comparisons of prevalence difficult, present findings provide the first insights into consumption trends of young people in Wales.

A number of environmental and policy changes may have contributed to the observed reduction in daily SSB consumption between 2000 and 2009. In 1996, the UK was reported to have one of the highest proportions of overall food advertisements worldwide, with 79% of adverts devoted to sweet or high fat foods. Since, more stringent advertising guidelines have been introduced to reduce young people’s exposure to advertising of unhealthy food products. Between 2001 and 2007, a series of school food policies were also introduced across Wales in an attempt to improve the nutritional standards of food and drink provided in schools [36, 37]. Despite revealing a substantial decrease in daily SSB consumption rates, 52% of adolescents continue to consume SSBs on a weekly basis and 6% still consume EDs daily. As such, further political action is required to maintain downward trajectories, notwithstanding any impacts which may have since occurred because of the 2018 SIDL. Furthermore, as the global ED market is forecast to reach a net worth in excess of $84 billion by 2025 (projected 7% increase in sales) [12], it is of public importance that consumption trends among young people continue to be monitored.

Socioeconomic patterning in consumption

We found clear patterning of SSB and ED consumption according to SES, observing higher consumption rates among young people from lower socioeconomic groups. These findings echo those of wider studies concerning adolescent SSB [25, 38, 39] and ED [38-40] consumption rates. With a lack of current legislation, the present findings are a potential reflection of the current marketing and availability landscape, with EDs being as affordable as SSBs [40] and some marketing trends disproportionately aimed at minority youth consumers [41]. A rapid UK-based review highlighted that ‘own brand’ EDs are often available at a cheaper price than water with young people preferring cheaper, less well-known varieties [42].

Our time trends analyses indicate that inequalities in SSB consumption have increased over time. While in 2002 there was no socioeconomic difference in SSB consumption, and declines have been observed for both groups, these have been faster among children from more affluent families. Hence, whilst actions to date may have led to an improvement in SSB consumption at the population level, actions may have also inadvertently contributed towards growing inequality. A key rationale for introducing a SDIL in 2018 was the expected equitable impact on population health (observing greater health gains in those with the worst health problems). With earlier UK-based models projecting a potential widening of inequalities due to the SDIL however [43], unearthing the impacts of the SIDL among young people is vital as are urgent policies aimed at reducing inequalities.

Present findings are in line with the current obesity landscape, which also reveals a persistent widening of inequalities, as the gap between child obesity prevalence in the most and least deprived areas of Wales continues to broaden [44]. It is widely accepted that considerable effort will be required to halt the growing inequalities in obesity rates as child poverty is likely to increase and in turn inequalities will persist or worsen. As part of a nationwide strategy [20], the Welsh Government seeks to reduce the impact of ill health and inequality, which includes a reduction in the diet inequality gap between the most and least deprived communities. Present findings can inform the divergent trends noted among obesity in young people, yet at present, actions towards a UK-wide legislation for ED sales appear to have become stagnant [18] and the evaluation of the 2018 Sugar tax is ongoing [22].

Our findings have important implications for practitioners and policy makers alike, demonstrating how secular consumption trends are disparate between socioeconomic groups, an area which is pivotal for the introduction or modification of responsive interventions. Pinpointing the underlying factors which contribute to such socioeconomic differences is key to ensuring policy interventions facilitate healthy food choices for all population groups [23]. Future work will look to examine any differences in consumption rates across socioeconomic gradients in light of the introduction of the soft drinks levy in 2018.

Limitations

There are several limitations. First, despite the strength of utilising a large scale, national survey, findings are reliant upon self-reported data and therefore are subject to reporting bias. Second, data are derived from cross-sectional surveys and slight changes in survey questions have resulted in manipulation of data into categories. For example, socioeconomic data were not collected in 1998 and 2000 with SSB regression models estimated from 2002 and data include two extra measures of socioeconomic status from 2013 onwards. Third, the use of a binary measure of socioeconomic status may limit interpretations, despite being widely used. Fourth, we do not account for the variation in sample size which increases over-time. Fifth, as only three consumption categories are used, some detail is lost, but trends have been explored in S13–S16 Tables. Lastly, while SSBs and EDs are increasingly varied in terms of amounts of sugar and ingredients included, our single item measures treat these as homogeneous products. Our ability to comment on the content or volume of drinks consumed is limited, with a reliance on consumption frequency data only. This poses potential implications for the interpretation of our findings as despite a common perception that portion sizes have widely increased [45], UK trends in soft drink portion sizes over time remain unclear [46].

Conclusion

Whilst overall reductions in SSB consumption are encouraging, study results indicate widespread continued consumption among adolescents and growing socioeconomic disparities in SSB consumption. There remains an urgent need for policy action to reduce adolescent consumption of SSBs, including EDs, and for these to be designed and evaluated with close attention to equity of impact.

Information on missing data.

(DOCX)

Click here for additional data file.

(DOCX)

Click here for additional data file.

Boys and girls SSB consumption over-time.

(DOCX)

Click here for additional data file.

Boys and girls ED consumption over-time.

(DOCX)

Click here for additional data file.

High and low SES SSB consumption over-time.

(DOCX)

Click here for additional data file.

High and low SES ED consumption over-time.

(DOCX)

Click here for additional data file.

Year 7’s SSB and ED consumption over-time.

(DOCX)

Click here for additional data file.

Year 8’s SSB and ED consumption over-time.

(DOCX)

Click here for additional data file.

Year 9’s SSB and ED consumption over-time.

(DOCX)

Click here for additional data file.

Year 10’s SSB and ED consumption over-time.

(DOCX)

Click here for additional data file.

Year 11’s SSB and ED consumption over-time.

(DOCX)

Click here for additional data file.

Cross-tabulation of sugary drink consumption and energy drink consumption.

(DOCX)

Click here for additional data file.

Interactions adjusted for other confounders; gender, school year and socioeconomic status (estimates in bold = p<0.05).

(DOCX)

Click here for additional data file.

SSB over-time before recoding.

(DOCX)

Click here for additional data file.

SSB over-time after recoding.

(DOCX)

Click here for additional data file.

ED over-time before recoding.

(DOCX)

Click here for additional data file.

ED over-time after recoding.

(DOCX)

Click here for additional data file.

Never or less SSB time trends according to socioeconomic grouping.

(TIF)

Click here for additional data file.

Weekly SSB time trends according to socioeconomic grouping.

(TIF)

Click here for additional data file.

11 Dec 2020

PONE-D-20-21401

Sugar-sweetened beverage consumption from 1998-2017: findings from the health behaviour in school-aged children/school health research network in Wales

PLOS ONE

Dear Dr. Morgan,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Jan 25 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Jane Anne Scott, PhD, MPH Grad Dip Dietetics, BSc

Academic Editor

PLOS ONE

Additional Editor Comments :

Lines 155 -160 Ethical approval  There is no reference to the SHRN surveys in this section

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For more information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially sensitive information, data are owned by a third-party organization, etc.) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: I Don't Know

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Enjoyed reading this piece of work. Overall this is an interesting piece of work considering change of SSB and ED in adolescents using cross sectional data. There are a number of changes noted below for the authors to consider and a few questions to consider.

Abstract:

Can you add some figures to back up key findings? Is there an overall conclusion for paper re implications for PH?

Intro:

Line 60: Rather than ‘it is linked to’ can you make this sentence a little clear to say consumption of these drinks are linked to dental erosion and Type II diabetes

Line 62: Can you add an actual figure where you say ‘of which Eds make up a contribution’ as you have done for SSBs.

Line 68: Because you are considering both SSB and Eds can you also provide the context for SSBs re economic contribution.

Lines 100-102: The aims and objectives need to be made clearer and more specific for this paper. Make reference to the data you explored, the specific time periods, and also the key variables explored.

Overall comment: As you start with SSB discussion – can you restructure the introduction to discuss aspects related to SSBs then EDs to make it more succinct and flow better. You tend to focus more on EDs than SSBs in parts of intro so the rationale in lines 98-99 only capture the ED aspect as opposed to the importance of exploring the effect of SSB consumption in this age group as well. Also, is this analysis based on total daily consumption of SSBs and EDs?

Methods:

Lines 108-110: I appreciate you have referenced the data are appended as per previous studies and but can you provide more context? For example, how frequently are these surveys carried out between 1998-2017 – yearly, every 2 years etc?

Lines 107-110: I think the authors need to include more detail about the survey methods. In particular the method used to obtain dietary data (FFQs). This can then be linked to the section on ‘limitations’ where they mention the reliance on self-reported data.

Socio-demographics – this section includes more than that so may be better as participant characteristics as you have gender, year etc included

Line 116 and 121: FAS method, I’m not overly familiar with this method but can you explain why you only have two groups - low and high for SES effect? You should also note the limitations of this method for SES in the discussion. In this section you also mention SES was only available from 2002 – what did you use as SES categorisation for data pre-2002? Or was this excluded for these data? This needs noted. Did the change in SES coding have any impact (2002 and then in 2013 changed – if not tested that’s okay but then need to note in limitations categorisation of SES did change between these time points)

Line 129: Here you mention in the first two survey years for SSBs - can you provide the actual survey years and periods you have from 1998-2017 (I see this in S1 table but it would help the reader to be included in the text) and be specific that the ED data only includes 2013-2017; that is quite different (4y verse 20y)

Line 164: change to ‘using’ Stata15

Results:

Currently are lengthy. I’m not convinced all the supplementary tables are required as a lot are descriptive.

A few suggestions would be to condense the descriptive stats and focus more on the overall key findings from the more complex model results, once you have factored in year, SES and gender. Otherwise they get lost. Also you have significant interactions – can you provide more detail/explanation on the interaction effects rather than just refer to a supplementary table – this is one of the supplementary tables that may be better in the results findings.

Are you able to provide actual p-values as opposed to reporting as p<0.05; appreciate you have CI but if using the p-value too then would be better.

Re presentation of results: use more detailed subheadings, for example, as the effect of year on both SSBs and ED; and then another subheading noting the effect of gender, SES and year group and just highlight the key points.

Discussion

Can the authors provide some detail as to why they think there has been such a reduction in daily SSBs since 2002/9 – here there was a steep reduction in comparison to other survey periods, prior to the sugartax etc. Anything related to changes in school food and drink purchasing (relates back to a prior question about is this data total dietary data?) Switch from SSB to diet drinks?

Limitations:

• Need to acknowledge point about the SES coding, change from 2002,

• I wonder if the authors can comment on the robustness of the SES categorisation, in particular for this piece of research? I.e. possible changes in the ‘sensitivity’ of FAS over the time periods. In addition, the paper referenced (Currie et al, 2008) states “FAS associations are strong for health out-comes that are related to family culture and behaviour, but less so for some behaviours where peer norms area potentially powerful influence”. I wonder if the authors can comment on this in terms of the influence of peers in SSB consumption and the subsequent use of FAS to categorise SES?

• Acknowledge the SSB drinks aren’t categorised into different sugar contents, cross sectional survey – not individual-child changes.

• Acknowledgment of the limitations of FFQs (in adolescents) should be included.

Line 313: What does “other HBSC countries” mean? Is the same survey conducted in other countries? Just needs a brief explanation, or perhaps include more detail about the HBSC and SHRN surveys in the methods (as mentioned above).

Two other questions: does the model account for the variation in sample size? For example, in one of the later surveys roughly over 40,000 by comparison to much smaller survey numbers previously?

Reviewer #2: Thank you for this well-written manuscript. This research investigates trends in SSB consumption frequency among 11-16 year olds in Wales, via national school surveys, over a period of almost 20 years; and ED consumption over a 5-year period. Sociodemographic determinants of SSB and ED consumption frequency are explored. This research has a large sample size, and 9 data collection points over the time period. The work contributes a comprehensive report of the current and historical SSB consumption trends among adolescents in Wales, and highlights the socioeconomic inequality of this behaviour.

My main concern is around the grouping of response options. Around 50% of responses were categorised as ‘weekly’, defined as consuming SSB/ED between 1 and 6 times per week. There is a big difference between these frequencies. This should be discussed as a limitation, and the disaggregated response data provided as a supplementary table to assist with interpretation of findings. I have given some more specifics to this (points 5 and 7 below), along with some other minor recommendations (in order of appearance through the manuscript).

1. The aim should be clarified, to differentiate between the amount of SSB and ED data you have (ie 19 years vs 5 years), and the addition of the word ‘frequency’, to describe the consumption pattern that you are measuring.

2. In the methods, please give a brief overview of the study sampling strategies and procedures for each survey; including sampling methods, frequency of administration and year groups sampled at each time point. Were the same schools sampled every time, or new schools approached at each data collection wave? Also describe the approach that led to larger sample sizes at the 2015 and 2017 waves.

3. Please also indicate in the methods whether questionnaires were completed solely by the students, a parent proxy or a combination.

4. In the results (line 211) please revise the statement ‘remained stable at 6% each year’ – ED data were collected at 3 time points in the period, not each year.

5. Could the disaggregated SSB & ED response data be made available as a supplementary table and mentioned in your results? In particular, it would be useful to see the spread of intake responses that were grouped into the ‘weekly’ category at each time point.

6. The abstract and discussion both mention that the data is ‘nationally representative’, but this is not established in the methods or results. Please do so in at least one of those sections, and then add a comment to the discussion about the representativeness of the samples and generalisablty of your findings.

7. Please add a discussion point about whether these categorical groupings might lead to an overestimation of the scale of the decrease in SSB consumption. Moving from daily to weekly SSB consumption is a great outcome, however the true decline may be much more modest. For example, is it possible that this decline is the result of SSB intakes decreasing from 7 to 6 times per week? Is it also possible that this is reflected by the replacement of one of the SSBs per week with an ED? The recommended disaggregated data table (see point 5) might help you explore the likelihood of this limitation.

8. A further limitation to consider is the reporting of only frequency rather than incorporating the amount consumed at any occasion. As portion sizes of commercial foods appear to have increased over time, is it possible that total SSB consumption (eg in mls) has not declined?

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Gemma Devenish

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

25 Jan 2021

We would like to thank both reviewers for their time to review our manuscript and for their comments and suggested changes. Please find a point-by-point response document attached along with an updated manuscript.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

8 Mar 2021

Sugar-sweetened beverage consumption from 1998-2017: findings from the health behaviour in school-aged children/school health research network in Wales

PONE-D-20-21401R1

Dear Dr. Morgan,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Jane Anne Scott, PhD, MPH Grad Dip Dietetics, BSc

Academic Editor

PLOS ONE

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: (No Response)

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: (No Response)

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: (No Response)

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: (No Response)

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Many thanks to the authors for working through and responding to all the comments. These have all been considered and addressed, thank you. The responses and additions have helped clarify any initial queries. I have enjoyed reviewing and reading the topic of the paper and it is of PH importance.

Reviewer #2: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Gemma Devenish

22 Mar 2021

PONE-D-20-21401R1

Sugar-sweetened beverage consumption from 1998-2017: findings from the health behaviour in school-aged children/school health research network in Wales

Dear Dr. Morgan:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Jane Anne Scott

Academic Editor

PLOS ONE