Cross-sectional surveys of the amount of sugar, energy and caffeine in sugar-sweetened drinks marketed and consumed as energy drinks in the UK between 2015 and 2017: monitoring reformulation progress

Objectives To investigate the sugar, energy and caffeine content of sugar-sweetened drinks marketed and consumed as energy drinks available in the UK. Study design We carried out a cross-sectional survey in 2015 and 2017 of energy drinks available in the main UK retailers. Methods The sugar (sugars g/100 mL), energy (kcal/100 mL), caffeine (mg/100 mL) and serving size were collected from product packaging and nutrition information panels of energy drinks available in the nine main UK grocery retailers, three health and beauty retailers and one convenience store. Results The number of formulations (per 100 mL) and number of products (per serving) have fallen (from 75 to 49 and from 90 to 59) between 2015 and 2017, respectively. Energy drinks surveyed showed a 10% reduction in sugar, from 10.6 to 9.5 g/100 mL (P=0.011) and a 6% reduction in energy content (P=0.005) per 100 mL between 2015 and 2017. The average caffeine content of energy drinks, with a warning label, has remained high at 31.5±0.9 in 2015 and 31.3±1.0 mg/100 mL in 2017. Despite there being reductions, sugar, energy and caffeine content remain at concerning levels in 2017. Conclusions To reduce the harmful impact of energy drinks, further reduction in sugar and a reduction in caffeine by reformulation are urgently needed. Other measures such as ban on the sale of energy drinks to children and smaller product sizes should also be explored, while warning labels should be kept. A reduction in sugar, energy and caffeine content and overall energy drinks consumption could be beneficial in reducing sugar, energy and caffeine intake of consumers of energy drinks.

• This paper shows that the sugar content of energy drinks is high and there is a wide variation in the sugar content of energy drinks.
• These findings demonstrate that the amount of free sugars added to energy drinks could be reduced without technical issues.
• The study was based on the sugars content data provided on energy drinks packaging labels in store; hence we relied on the accuracy of the data provided on the label.

INTRODUCTION
Energy drinks are typically non-alcoholic drinks containing caffeine and sugar in combination with other ingredients known to have stimulant properties and are distinct from sports drinks and vitamin waters. Worldwide consumption of energy drinks has been increasing dramatically in recent years. The $50 billion global energy drinks market is projected to grow at an annual rate of 3·5% between 2015 and 2020, even though the soft drinks market is declining overall.
[1] The increase in the popularity of energy drinks has raised concerns among the scientific community, governments and the general public about the health effects of these products, particularly among children and adolescents. [2][3][4] In 2011, a study by the European Food Safety Authority (EFSA) on consumption data of energy drinks in 16 countries of the European Union found that 68% of adolescents (aged 10-18 years old), 30% of adults, and 18% of children (<10 years old) consumed energy drinks.
[2] Another study by EFSA found that young people in the UK consumed more energy drinks than their counterparts across the other EU countries (3.1 litres per person per month, compared with 2 litres). [5] Market research reports potential increase in purchase of energy drinks among 10-14 year olds. [6] The rise in their popularity and their high sugar content suggests it is contributing to obesity, type 2 diabetes and dental caries, [7][8][9][10][11][12][13][14][15] all are major public health problems in the UK,[16-23] with significant costs. [24] In July 2015, the Scientific Advisory Committee on Nutrition (SACN) recommended average intake, across the UK population, of free sugars should not exceed 5% of total energy intake. [25] This is in line with the World Health Organisation's new guidelines on free sugars intake. [15,26] 'Free sugars' includes all monosaccharides and disaccharides added to foods by the manufacturer, cook or consumer, plus sugars naturally present in honey, syrups and unsweetened fruit juices. Under this definition lactose (milk sugar) when naturally present in milk and milk products and sugars contained within the cellular structure of foods (particularly fruits and vegetables) are excluded. Furthermore, SACN advised that consumption of sugars-sweetened drinks, including energy drinks, should be minimised in children and adults.
Current average intakes of free sugars exceed recommendations in all age groups. The average sugars intake in adults was 59g per day and contributes to 12·1% of energy intake.
Children have a higher sugars intake. The average sugars intake was 61 and 74g per day in 4-10 and 11-18 year olds respectively, about three times the new recommendation in 11 to 18-year olds. [27] This is likely to be an underestimate of how much free sugars are consumed [28] because underreporting is highly prevalent in these types of surveys. [29][30][31][32] Soft drinks are the main contributor of free sugars intake in children (4-10 years), teenagers (11-18 years) and the second main contributor in adults (18-64 years), contributing to 30%, 40% and 25% of free sugars intake respectively. [27] Within soft drinks, energy drinks are an important contributor of free sugars intake. Sales of energy drinks in the UK have increased by 155% between 2006 and 2014, from 235 to 600 million litres, an average per capita consumption of 9·4 litres in 2014, and a total value of £1·48 million. [33] The leading brands' shares in the UK energy drinks market, by value in 2013/14, were Red Bull (25%), own-label (12%), Monster (10%), Relentless (6%), Rockstar (5%), Mountain Dew (2%), Boost (2%) and Emerge (2%). [6] In the UK, the government plans in 2018 to implement its own levy on sugary drinks with more than 5g of sugar per 100ml, with a higher rate for drinks (including energy drinks) with more than 8g per 100ml. The impact of these types of levies on sales of energy drinks remains to be seen. [34] While the potential impact of energy drinks in terms of their caffeine have been discussed in recent years[2, 35] little research has looked at the levels of sugars, calories and caffeine in  5 energy drinks which can impact obesity, type 2 diabetes and dental caries. Considering this, we carried two cross-sectional surveys to evaluate the sugars, calorie and caffeine content of energy drinks sold in the UK between 2015 and 2017, report the variability in sugar level, assess the sugar content in relation to the UK's new daily recommendation for sugar intake and assess the number of products that will be taxed based on the criteria suggested in the Soft Drinks Industry Levy.

Data Collection
The data was collected from product packaging and nutrient information panels in January 2015 and January 2017. The survey was designed as a comprehensive survey of all energy drinks available in a snapshot in time, using one large outlet per each of the nine main grocery retailers, three beauty and health food retailers and one convenience store.
For each energy drink, the data collected included the company name, product name, pack weight, serving size, sugars (g), energy (kcal) and caffeine (mg) per 100ml and sugars, energy and caffeine per serving. All data was double checked after entry, and a further 5% of entries were checked against the original source in a random selection of products. This study used the same study design and procedures as reported in our previous work. [36] Inclusion/exclusion criteria: Data were collected from each of the major UK supermarkets (Aldi, Asda, Lidl, Marks and Spencer, Morrisons, Sainsbury's, Tesco, The Co-operative and Waitrose) and three beauty, health food retailers (Boots, Superdrug and Holland and Barrett) and a Costcutter store to represent the amount of free sugars in energy drinks in the UK. We included supermarket own brand and branded products. Energy drinks products with sugars and calories information labeled were included. As the study focus is on product reformulation, products labelled called 'zero' or 'light' or 'no added sugar' etc. were excluded.
Energy drinks were defined as any drinks products with 'energy', 'energise', 'energising', 'caffeine' and 'stimulation' in the product name. We also excluded isotonic or sports drinks, which consist primarily of carbohydrates and electrolytes and are intended to help athletes The caffeine content of a small selection of products were collected in 2015 (n =11) so comparison over the two surveys of caffeine content were not made, as the sample size from 2015 was too small.
Serving size: The serving size was calculated from per 100ml nutrition information provided on packaging. Any products with up to 500ml can or bottle size was considered as one serving, regardless of what is stated on the packaging as a serving size e.g. often a 500ml bottle is split into two serving but we consider that most consumers consume these drinks as one serving. In 1 litre products, the serving size was standardised to 250ml. If a 250ml can of the same product exists then, the litre product was excluded.
High, Medium and low criteria for sugars content: The free sugars content will be compared to the UK front of pack colour-coded labelling for drinks. Portion size criteria applies to Data are reported as mean, SD, range as indicated. Significance in all tests carried out was deemed significant as being p<0·05. All data were analysed using SPSS Version 22.
In addition to the above analyses, the sugar content was compared to the UK's new recommendation for sugar intake for adults (30g/day) and 7-10 year old children (24g/day) and to the colour-coded labelling for sugars content.

Sugar, calorie and caffeine content per 100 ml in the 2015 survey
A total of 79 energy drinks met the inclusion criteria and were included in our analysis. The average sugar content in energy drinks was (±SD) 10·8±2·8g/100ml. There was a large variation in sugars content between different energy drinks ranging from 1·9 to 15·9g/100ml.
The average calorie content in energy drinks was 48±13 kcal/100ml. There was a large variation in calorie content between different energy drinks ranging from 10 to 72 kcal/100ml. There were no significant differences in sugar content between supermarket own brand and branded products (48±6 vs 48±14 kcal/100ml, P=0·938).  8 A total of 102 energy drinks products met the inclusion criteria and were included in our analysis, 89 products were branded and 13 were supermarket own-brand energy drinks (Table 2). The serving size varied from 150 to 500ml. Among the 102 products 39% were sold in 500ml cans/bottles.
There was a large variation in sugars content between different energy drinks ranging from 9·5 to 78·0g/serving.
On average, a serving of energy drink (40·8g/serving) contains more than the entire maximum daily recommendation for sugar intake in the UK for an adult (30g).
Among the 102 products 62% exceeded the maximum UK's recommendation for sugar intake per serving for an adult (30g/d). Additionally, 87% of the products exceeded the maximum daily recommendation for sugar intake for a child aged 7-10 years (24g/d).
The average calorie content in energy drinks was 184±79 kcal/serving. There was a large variation in calorie content between different energy drinks ranging from 50 to 350kcal/serving. Supermarket own brand (140±48kcal/serving) contained fewer calories than branded (190±82kcal/serving) products, but this was not statistically significant (p=0·069).
The energy drink product with the highest calorie content (350 kcal) can contribute up to 17·5% of a woman's daily energy intake (2000 kcal).
A total of 11 energy drinks formulations (per serving) had caffeine content labelled on pack and were collected in 2015. The average caffeine content was 133·4±50·7 mg/serving.

Sugar, calorie and caffeine content per 100ml in the 2017 survey
A total of 48 energy drinks products met the inclusion criteria and were included in our analysis, 39 products were branded and 9 were supermarket own-brand energy drinks  10 The serving size was significantly larger in branded verses supermarket own brand products (P=0·002), thus supermarket own brand contained on average lower levels of sugars compared to branded products per serving (26·3±14·9 vs 39·5±18·3g/serving, P=0·024) ( Figure 1).
The average calorie content in energy drinks was 173±75kcal/serving. There was a large variation in calorie content between different energy drinks ranging from 50 to 335kcal/serving. Supermarket own brand (112±58kcal/serving) contained significantly fewer calories than branded (183±73kcal/serving) products, (p=0·005) (Figure 2). The energy drink product with the highest calorie content (335 kcal) can contribute up to 17% of a woman's daily energy intake (2000 kcal).
A total of 41 out of 65 energy drinks products provided caffeine content per serving or we could calculate caffeine content per serving, 9 products were branded and 32 were supermarket own-brand energy drinks. The average caffeine content in energy drinks was 126·3±39·2 mg/serving. There was a large variation in caffeine content between different energy drinks ranging from 75·0 to 160·0 mg/serving. There was a significant difference in caffeine content between supermarket own brand and branded products (86·1±27·8 vs 137·6±34·3 mg/serving, P=0·001).
The total number of products with labelled caffeine content here is different to the number of products included per 100ml analysis because the 100ml data only included an example of one formulation regardless of the different serving sizes, while the serving data included all the different serving sizes available, even if it is the same formulation. The caffeine content per serving differed between products with the same formulation; therefore, it was necessary to include all the products regardless if they are the same formulation.

Comparing to the maximum daily recommendation for sugar intake
On average, a serving of energy drink (37·7g/serving) contains more than the entire maximum daily recommendation for sugar intake in the UK for an adult (30g).
Among the 65 products, 36 (55%) exceeded the maximum UK's recommendation for sugar intake per serving for an adult (30g/d). Additionally, 48 products (74%) exceeded the  11 maximum daily recommendation for sugar intake for a child aged 7-10 years (24g/d).

Comparison of the same products between the two years
The overall number of branded and supermarket own brand energy drinks has decreased from 2015 to 2017 (102 vs 65 products).
Note that these averages are slightly different from those when all products were included in each year and this trend analysis reflects reductions made in the same products rather than the overall products available.

DISCUSSION
Energy drinks consistently surveyed showed significant reduction of sugar content. Despite this, sugar content was still high. In 2017, the average sugar content in energy drinks was (±SD) 9·7±3·0g/100ml. There was a large variation in sugar content between different energy drinks ranging from 2·1 to 16·0g/100ml. The average sugar content in energy drink in both surveys per serving was more than an adult's entire maximum daily recommendation for sugar intake in the UK (40·8±17·6g vs 37·7 ±18·3g/serving). Furthermore, among the 65 products surveyed, 74% exceeded the maximum daily recommendation for sugar intake for a child aged 7-10 years. Therefore the claim that energy drinks can be consumed as part of a 'healthy balanced diet' is misleading. [25] Due to high sugar level, the calorie content in energy drinks is high with an average of 184±80kcal in 2015 and 173±74kcal/serving in 2017. The products with the highest sugars and calorie content could contribute up 17% of a woman's total daily energy intake. Whilst certain drinks manufacturers claim their products are a good source of energy, this is also misleading. The body generates energy from any food, such as fruits, vegetables, breads, pasta and rice and there is no need whatsoever for added sugar that are in energy drinks.
The term 'energy drinks' is chosen to communicate a falsehood in disguise, i.e. that these drinks are full of energy giving nutrients rather than sugar that can contribute to weight gain, if not burnt off. Therefore, the levels of sugar and calories in energy drinks can be an important contributor to obesity, type 2 diabetes and dental caries and therefore substantial healthcare costs [38][39][40].
Besides, the concerns around the sugar and calorie content of energy drinks, there is the concern around the levels of caffeine in them. In 2017, average caffeine content in energy drinks was 126·3±39·2 mg/serving among the 41 products labelled, almost equivalent to two cups of coffee. There is also some evidence to suggest that the caffeine content increases sugar-sweetened beverage consumption in a free-living population.
[41] It important to note that not all companies label the amount of caffeine in their energy drinks but among the companies that do, they also state a warning on pack, acknowledging the potential danger these products can possess to children. However, these products can be easily purchased and consumed by children.
Energy drinks also contribute to sugar intake in many other countries around the world, [2] showing that high levels of sugars as well as caffeine in energy drinks is a global national 'Get Healthy' campaign. In the UK, the government plans in 2018 to implement its own levy on sugary drinks with more than 5g of sugar per 100ml, with a higher rate for drinks (including energy drinks) with more than 8g per 100ml. The impact of these types of levies on sales of energy drinks remains to be seen. [34] Another option a country can adopt, to protect children and adolescents from energy drinks is implement a systematic, unobtrusive and gradual reformulation programme for manufacturers to reduce the amount of sugar and caffeine in energy drinks. [46] The reductions seen so far suggest that the UK market may be adopting this approach.
A similar successful reformulation programme has been implemented in the UK to reduce salt intake in the whole population. The key to the success of the UK salt reduction programme is setting incremental targets for each food group with a specified timeframe to be achieved using maximum and average or sales-weighted average targets [47]. Since there has been a gradual, progressive reduction in salt, the UK population has adjusted to the taste of lower salt levels. There has been no loss of sales or switching between products as a result of salt reduction, or addition of salt at the Alongside reducing the free sugars and caffeine content and governments can introduce regulation to restrict larger serving size. Recently, the Grocer has reported an increase in sales of larger cans of energy drinks on the UK market, which must be tackled to reduce intake of sugar, calories and caffeine from these products.
Our study was based on sugars, calorie and caffeine content data provided on the available energy drinks packaging labels in store on the dates of collection; hence we relied on the accuracy of the data provided on the label. Therefore, it is assumed that the manufacturers

CONCLUSION
The full impact of the rise in popularity of energy drinks has not yet been measured, but the increase in sales, [33,53,54] the high amount of caffeine and free sugars in energy drinks (measured in this study), the pervasive and misleading marketing targeted at young people,[55, 56] the ease of access, [3] combined with limited regulation have created an environment where energy drinks could pose a major public health threat.
This study indicates there has been a reduction in sugar content of energy drinks in the UK since 2015. Sugars content however remains high. This research demonstrates that sugars content of energy drinks in the UK is of concern, with a typical serving (500ml) exceeding an adult's entire maximum daily recommendation of sugar intake. More can and should be done. The benefits will be very large in reducing both sugars and calorie consumption and therefore a reduction in obesity, type 2 diabetes and dental caries.

COPYRIGHT
The Corresponding Author has the right to grant on behalf of all authors and does grant on behalf of all authors, a worldwide licence to the Publishers and its licensees in perpetuity, in all forms, formats and media (whether known now or created in the future), to i) publish,

TRANSPARENCY DECLARATION
The manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant) have been explained.

FUNDING
This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

ETHICS
Not applicable.

ROLE OF SPONSOR
Not applicable.         Conclusions: In order to reduce the harmful impact of energy drinks, further reduction in sugar and a reduction in caffeine by reformulation are urgently needed. Other measures such as ban on the sale of energy drinks to children and smaller product sizes should also be explored, while warning labels should be kept. A reduction in sugar, energy and caffeine content and overall energy drinks consumption could be beneficial in reducing sugar, energy and caffeine intake of consumers of energy drinks.

Strengths and limitations of this study
• This paper for the first time investigates and documents the sugar, energy and caffeine content of energy drinks sold in the UK. • These results demonstrate that the amount of sugar and energy has fallen, without technical issues, while caffeine levels remain high.
• This paper shows that serving sizes of sugary energy drinks are large and should be reduced.
• The study was based on the levels provided on energy drinks packaging labels in store; hence we relied on the accuracy of the data provided on the label.

INTRODUCTION
In July 2015, the Scientific Advisory Committee on Nutrition (SACN) recommended that average free sugars (sugar) intake, across the UK population, should not exceed 5% of total energy intake.
[1] This is in line with the World Health Organisation's new guidelines on sugar intake. [2, 3] SACN also advised that consumption of sugar-sweetened drinks, including energy drinks, should be minimised in children and adults, [1] because high intake of sugar is contributing to obesity, type 2 diabetes and dental caries, [3][4][5][6][7][8][9][10][11] major public health problems in the UK, [12-19] with significant healthcare costs. [20] Current average intakes of sugar exceed recommendations in all age groups.
[21] The average sugar intake in adults was 59g per day and contributes to 12·1% of energy intake.
Children have a higher sugar intake with an average of 61 and 74g per day in 4-10 and 11-18 year olds respectively, which is about three times the new recommendation in 11 to 18-year olds (24g/day). [21] Soft drinks are the main contributor of sugar intake in children (4-10 years) and teenagers (11-  The increase in consumption and popularity of energy drinks has raised concerns among the scientific community, governments and the general public about the health effects of these products, particularly among children and adolescents. [26, 33,34] There have been two policies to tackle sugar and caffeine content of energy drinks. In terms of sugar content, a two-tiered Soft Drinks Industry Levy (SDIL) on the production and importation of sugary drinks will be introduced in 2018.
[35] The policy is intended to drive product reformulation and lower sugar consumption; sugary drinks with more than 5g of sugar per 100ml are taxed at a lower rate, with a higher rate for drinks (including energy drinks) exceeding 8g per 100ml.
[35] The impact of the levy on sales of energy drinks remains to be seen.
[36] In terms of caffeine content, the EU Food Information Regulation requires specific labelling for high caffeine drinks (over 150mg per litre (mg/l).
[37] The warning states that the product is not recommended for children. However, these products can be easily purchased and consumed by children.

METHODS
The data was collected from product packaging and nutrient information panels in January 2015 and January 2017. The survey was designed as a comprehensive survey of all energy drinks available in a snapshot in time, using one large outlet for each of the nine main grocery retailers, three beauty and health food retailers and one convenience store. This study used the same study design and procedures as reported in our previous work on carbonated sugar sweetened drinks, which excludes 'energy' drinks). [38] Definition 'Energy' drinks were defined as any drink products with 'energy', 'energise', 'energiser', 'caffeine' and 'stimulation' in the product name or description e.g. Red Bull Energy Drink,

Monster Energy Drink, Relentless Origin Energy Drink and Tropicana Energy Mango and
Guava with Passion fruit. Additionally, products that are known to contain high caffeine content but are not described as 'energy drinks' (such as Mountain Dew) were also included.
Since the focus of this study is on product reformulation, products labelled called 'zero' or 'light' or 'no added sugar' etc. were excluded. We also excluded products described solely as 'sports' drinks, which consist primarily of carbohydrates and electrolytes and are intended for athletes to rehydrate after exercise e.g. 'Lucozade Sports'.

Data Collection
For each 'energy' drink the data collected included the company name, product name, pack weight, serving size, sugars (g), energy (kcal) and caffeine (mg) per 100ml and sugars, energy and caffeine per serving. Where data was not available per serving, it was calculated from pack size and per 100ml data. Caffeine content was only collected in a sub-sample of the 2015 products.
Where products only labelled carbohydrate content in 2015 but labelled sugar content in 2017 and the content of all nutrients and ingredients were the same, it was assumed that the sugar content from 2015 was the same as 2017.
All data was double checked after entry, and a further 5% of entries were checked against the original source in a random selection of products.

Stores
Data was collected from each of the major UK supermarkets in store (Aldi, Asda, Lidl, Marks and Spencer, Morrisons, Sainsbury's, Tesco, The Co-operative and Waitrose), which altogether hold 93.2% of the grocery market share.
[39] Three beauty, health and food retailers (Boots, Superdrug and Holland and Barrett) and a Costcutter (convenience store) were also included since they sell a wide variety of branded energy drinks.

Analysis
Per 100ml: Some brands sell the same formulation in different serving sizes. Therefore, the 100ml data only included an example of one formulation regardless of the different serving sizes.
Per serving: The per serving data included all the different serving sizes available apart from 1 litre bottles. Any product with up to 500ml can or bottle size was considered as one serving, regardless of what is stated on the packaging as a serving size e.g. often a 500ml bottle is split into two servings but we consider that most consumers drink these drinks as one serving. 1 litre products were excluded from the per serving analysis since it was  Caffeine: Separate analysis was conducted on the products with the high caffeine warning label (over 150mg per litre (mg/l).
High, Medium and Low criteria for sugars content: The sugar content was compared to the UK front of pack colour-coded labelling for drinks. Portion size criteria applied to portion/serving sizes greater than 150ml. Colour coding is based on the following front of pack colour-coded nutrition labelling criteria (Sugars -Red/High >13·5g/portion or >11·25g/100ml, Amber/Medium >2·5 to ≤11·25/100ml, Green/Low ≤2·25g/100ml). [40] In addition to the above analyses, the sugar content was compared to the UK's recommendation for sugar intake for adults (30g/day) and 7-10 year old children (24g/day) and to the colour-coded labelling for sugar content.

Comparison among products within each survey
Independent-Samples Mann-Whitney U Test was used to compare the levels of sugar between supermarket own label and branded products.

Comparison of the same products between the two years
For the purpose of this comparison, only the products with data available in both surveys were included in the analysis. Related-Samples Wilcoxon Signed Rank Test was used to examine whether there was a significant change in the sugar, energy and caffeine content of energy drinks from 2015 to 2017.
Data are reported as mean, SD, range as indicated. Significance in all tests carried out was deemed significant as being p<0·05. All data were analysed using SPSS Version 22.

Sugar, energy and caffeine content per 100 ml in 2015 and 2017
A total of 74 and 49 energy drinks met the inclusion criteria in 2015 and 2017 respectively.
The average sugar content was 10.6±2.9 g and 9.7±3.0 g/100ml, with a large variation in sugar content between different energy drinks ranging from 1·9 to 15·9g/100ml and 2·1 to 16·0g/100ml in 2015 and 2017 respectively (Table 1). There were no significant differences in sugar content between supermarket own label and branded products in 2015 (P=0·523) and 2017 P=0·245). The product with the highest and lowest sugar content per 100ml in 2015 and 2017 are highlighted in Table 2.
Among all of the manufacturers with three or more formulations of energy drinks in 2015, Lucozade and Rockstar product ranges contained the highest average sugar and energy per 100ml respectively (Table 3). However, in 2017 the Rockstar product range contained the highest average sugar and energy content per 100ml (Table 3).
Since not all 'energy drinks' have a high caffeine content, we analysed the caffeine, sugar and energy content of the products with the warning label separately . A total of 19 and 41 energy drinks formulations had caffeine content labelled on pack (there may have been more that were labelled but not all were collected in 2015). The average caffeine content in energy drinks was (±SD) 30·8±3.2 in 2015 and 31·6±0·8 mg/100ml in 2017 (Table 1). The levels of sugar, energy and caffeine were similar to the levels in the full sample of energy drinks.

Sugar, energy and caffeine content per serving in 2015 and 2017
A total of 89 and 59 energy drinks products met the inclusion criteria in 2015 and 2017 respectively ( Table 4). The total number of products available has fallen between 2015 and   Table 4).

Comparison of the same products between the two years
The overall number of energy drinks formulations has decreased between 2015 and 2017, from 74 to 49 products.

Per 100ml
There were 30 products surveyed repeatedly over the 2 years. The average sugars content per 100ml for these 30 products was 10·6±3·1g in 2015 and 9·5±3·3g in 2017. This represents a reduction of 10% from 2015 to 2017 (p=0·014). The sugar content has reduced in 12 products , remained the same in 16 and has increased in 2. The average calorie content per 100ml was 47±14 kcal in 2015 and 44±15 kcal in 2017, representing a 6% reduction (p=0·007).
There were 12 products surveyed repeatedly in 2015 and 2017 displaying the high caffeine warning label. The caffeine content of these products stayed the same, which is a better reflection of product reformulation.
Note that these averages are slightly different from those when all products were included in each year and this trend analysis reflects reductions made in the same products rather than the overall products available, for a full list of products refer to supplementary file 1.

DISCUSSION
This survey has documented the levels of sugar, energy and caffeine in energy drinks in the UK between 2015 and 2017, a topic which hasn't been previously documented for easy  Despite this, sugar content remains at concerning levels, with 95% would receive a 'red' (high) label for sugars per serving in 2017 respectively. There was also a large variation in sugar content between different energy drinks ranging from 2·1 to 16.0g/100ml, which suggests further reductions are possible.
As well as the concerns around the sugar and energy content of energy drinks, there are also concerns about caffeine levels, particularly in the products labelled as high caffeine. In 2017, average caffeine content in energy drinks was 125.3 ± 39.2 mg/serving among the 42 products labelled, almost equivalent to two cups of coffee or four cans of cola.
[37] There is also some evidence to suggest that the caffeine content increases sugar-sweetened drinks consumption further and therefore sugar intake too. [46] Since children and teenagers are the main consumers of energy drinks, manufacturers should consider reducing levels of caffeine -again through reformulation. Indeed a recent study showed that removing caffeine from sugar-sweetened drinks, along with 10·3% of sugar, has the potential to reduce body weight of adults by 0.6kg, without any change in sugar-sweetened drinks consumption. [47] It important to note that not all manufacturers label the amount of caffeine in their energy drinks and this is likely because their products do not exceed the cut-off for adding the warning label. Furthermore, serving sizes of energy drinks matter since the larger the serving size, the more sugar, energy and caffeine is consumed. Indeed, typical serving sizes of energy drinks are larger than other sugar-sweetened drinks; [38] this survey showed that the most common serving size was 500ml. The average sugar content in energy drink in both surveys per serving was more than an adult's entire maximum daily recommendation for sugar intake in the UK. Likewise, 87% in 2015 and 78% in 2017 of the products exceeded the maximum daily recommendation for sugar intake for a child aged 7-10 years (24g/d).
[1] The products with the highest sugar and energy content per serving could contribute up 17.5% of a woman's total daily energy intake. Therefore, in order to reduce the amount of sugar, energy and caffeine consumed from energy drinks, there is scope for introducing restrictions on larger serving sizes (500ml bottles and cans). This is a growing problem since the Grocer recently reported an increase in sales of larger cans of energy drinks on the UK market. [23] Our study was based on sugars, energy and caffeine content data provided on the available energy drinks packaging labels in store on the dates of collection; hence we relied on the accuracy of the data provided on the label. Therefore, it is assumed that the manufacturers

CONCLUSION
Factors such as an increase in sales, [24,25,30] concerning levels of sugar, energy and caffeine (assessed in this study), ease of access [33] as well as limited regulation have created an environment where energy drinks could pose a major threat to public health.
In order to reduce the impact of energy drinks, sugar (and therefore energy) reformulation should continue and begin regarding caffeine content. Other measures such as a ban on the sale of energy drinks to children should be explored, while warning labels should be kept.
Can and bottle sizes of energy drinks should also be reduced to reduce overall consumption of sugar, energy and caffeine. A reduction in sugar, energy and caffeine content and overall energy drinks consumption could be beneficial in reducing sugar, energy and caffeine intake of consumers of energy drinks.

COMPETING INTERESTS
All authors have completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years. KH

Strengths and limitations of this study
• This paper for the first time investigates and documents the sugar, energy and caffeine content of energy drinks sold in the UK. • These results demonstrate that the amount of sugar and energy has fallen, without technical issues, while caffeine levels remain high.
• This paper shows that serving sizes of energy drinks are large and should be reduced.
• The study was based on the levels provided on energy drinks packaging labels in store; hence we relied on the accuracy of the data provided on the label.  [36] The reformulated products become a replacement of an existing product (e.g.

INTRODUCTION
the same brand of energy drink with less sugar). The SDIL aims to incentivise manufacturers of sugar-sweetened drinks with more than 8g of sugar, per 100ml, to reduce the levels to below 8g and pay a lower tax rate and manufacturers of sugary drinks with more than 5g of sugar per 100ml, to lower the sugar levels to less than 5g and pay no tax. [35]

METHODS
The data was collected from product packaging and nutrient information panels in January 2015 and January 2017. The survey was designed as a comprehensive survey of all energy drinks available in a snapshot in time, using one large outlet for each of the nine main grocery retailers, three beauty and health retailers and one convenience store. This study used the same study design and procedures as reported in our previous work on carbonated sugar sweetened drinks, which excluded energy drinks. [38] Definition Energy drinks were defined as any drink products with 'energy', 'energise', 'energiser', 'caffeine' and 'stimulation' in the product name or description e.g. Red Bull Energy Drink,

Monster Energy Drink, Relentless Origin Energy Drink and Tropicana Energy Mango and
Guava with Passion fruit or products with high caffeine warning label including products not described as energy drink (such as Mountain Dew). Since the focus of this study is on product reformulation, products labelled 'zero' or 'light' or 'no added sugar' etc. were excluded. We also excluded products described solely as 'sports' drinks, which consist primarily of carbohydrates and electrolytes and are intended for athletes to rehydrate after exercise e.g. 'Lucozade Sports'.

Data Collection
For each energy drink the data collected included the company name, product name, pack weight, serving size, sugars (g), energy (kcal) and caffeine (mg) per 100ml and per serving.
Where data was not available per serving, it was calculated from pack size and per 100ml data. Caffeine content was only collected in a sub-sample of the 2015 products.
Where products only labelled carbohydrate content in 2015 but labelled sugar content in 2017 and the content of all nutrients and ingredients were the same, it was assumed that the sugar content from 2015 was the same as 2017 (applied to only one product).
The data were double checked after entry into excel spreadsheet, and a further 5% of entries were checked against the original source in a random selection of products by the lead author.

Stores
Data was collected from each of the major UK supermarkets in store (Aldi, Asda, Lidl, Marks and Spencer, Morrisons, Sainsbury's, Tesco, The Co-operative and Waitrose), which altogether hold 93.2% of the grocery market share in the UK.
[39] Three beauty, health and food retailers (Boots, Superdrug and Holland and Barrett) and one convenience store (Costcutter) were also included since they sell a wide variety of branded energy drinks.

Analysis
Per 100ml: Some brands sell the same formulation in different serving sizes. Therefore, the 100ml data only included an example of one formulation regardless of the different serving In addition to the above analyses, the sugar content was compared to the UK's recommendation for sugar intake for adults (30g/day) and 7-10 year old children (24g/day).
[40] The highest energy containing products were compared to women's daily energy intake (2000 kcal).
[41] Also, assess the number of products that will be taxed based on the criteria suggested in the SDIL. [35]

Comparison among products within each survey
Independent-Samples Mann-Whitney U Test was used to compare the levels of sugar between supermarket own label and branded products.

Comparison of the same products between the two years
For the purpose of assessing reformulation, only the products with data available in both surveys were included in this analysis. Related-Samples Wilcoxon Signed Rank Test was used to examine whether there was a significant change in the sugar, energy and caffeine content of energy drinks from 2015 to 2017.
Data are reported as mean, SD, range as indicated. Significance in all tests carried out was deemed significant as being p<0·05. All data were analysed using SPSS Version 22.

Sugar, energy and caffeine content per 100ml in 2015 and 2017
A total of 75 and 49 energy drinks met the per 100ml inclusion criteria in 2015 and 2017 respectively. The average sugar content was 10.6±2.9 g and 9.7±3.0 g/100ml, with a large variation in sugar content between different energy drinks ranging from 1·9 to 15·9g and 2·1 to 16·0g/100ml in 2015 and 2017 respectively (Table 1) Table 2.
Among all of the manufacturers with three or more formulations of energy drinks in 2015, Lucozade and Rockstar product ranges contained the highest average sugar and energy per 100ml respectively (Table 3). However, in 2017 the Rockstar product range contained the highest average sugar and energy content per 100ml (Table 3), which suggests Lucozade have been reformulating their products.   (Figure 1 and Table 4).

Comparison of the same products between the two years
The overall number of energy drinks formulations has decreased between 2015 and 2017, from 74 to 49 products.

Reformulation
There were 30 products surveyed repeatedly in 2015 and in 2017. The average sugar content per 100ml for these 30 products was 10·6±3·1g in 2015 and 9·5±3·3g in 2017. This represents a reduction of 10% (p=0·014). The sugar content has reduced in 12 products, remained the same in 16 and has increased in 2. The average energy content per 100ml was 47±14 kcal in 2015 and 44±15 kcal in 2017, representing a 6% reduction (p=0·007).  11 There were 11 products surveyed repeatedly in 2015 and 2017 displaying the high caffeine warning label. The caffeine content of these products stayed the same, apart from one product, where the caffeine content was reduced, from 32mg to 32mg.
Note that these averages are slightly different from those when all products were included in each year and this trend analysis reflects reductions made in the same products rather than the overall products available, giving a better reflection of reformulation, for a full list of products refer to supplementary file 1. This survey may suggest the early success of the SDIL on bringing down sugar levels in energy drinks but we are yet to see if the SDIL will have an impact on overall consumption of energy drinks by reducing sales. [44] Other countries have imposed taxes on soft drinks, including energy drinks, and have seen reductions in sales.

DISCUSSION
[45] [46]  of a woman's total daily energy intake. Therefore, in order to reduce the amount of sugar, energy and caffeine consumed from energy drinks, there is scope for introducing restrictions on larger serving sizes (500ml bottles and cans). This is a growing problem, that must be tackled, particularly since the Grocer recently reported an increase in sales of larger cans of energy drinks on the UK market. [23] Our study was based on sugar, energy and caffeine content data provided on the available energy drinks packaging labels in store on the dates of collection; hence we relied on the accuracy of the data provided on the label. Therefore, it is assumed that the manufacturers provide accurate and up to date information in line with EU Regulations. However, future studies should include sugar and energy content determined through laboratory analysis to achieve a better understanding of the true sugar and energy content, and breakdown of the types of sugars.
Also, the data collection process was slightly different in the two surveys. In 2015, caffeine was collected in a sub-sample of products, since the aim was not to collect caffeine at the time, but the data that was collect was included in the analysis, this may not give a full reflection of caffeine content of all products on the market in 2015.
Still, the results of this study are relevant and serve to document the sugar, energy and caffeine content of energy drinks sold the in UK, underpinning future studies and providing evidence for the Treasury and the soft drinks industry, who are reformulating. [50, 51]

CONCLUSION
Factors such as an increase in sales, [24,25,30] concerning levels of sugar, energy and caffeine (assessed in this study), ease of access [33], large serving sizes, as well as limited regulation have created an environment where energy drinks could pose a major threat to public health.
In order to reduce the impact of energy drinks, sugar (and therefore energy) reformulation Can and bottle sizes of energy drinks should also be decreased to reduce the overall consumption of sugar, energy and caffeine. A reduction in sugar, energy and caffeine content and overall energy drinks consumption could be beneficial in reducing sugar, energy and caffeine intake of consumers of energy drinks.

COMPETING INTERESTS
All

Strengths and limitations of this study
• This paper for the first time investigates and documents the sugar, energy and caffeine content of products marketed as energy drinks sold in the UK. • These results demonstrate that the amount of sugar and energy has fallen, without technical issues, while caffeine levels remain high.
• This paper shows that serving sizes of energy drinks are large and should be reduced.
• The study was based on the levels provided on energy drinks packaging labels in store; hence we relied on the accuracy of the data provided on the label.
[21] The mean sugar intake in adults was 60g per day and contributes to 12% of daily energy intake.
Children have a higher sugar intake with an average of 54g (13.4%) and 73g (15.1%) per day in 4-10 and 11-18 year olds respectively. [21] Soft drinks are the main contributor of sugar intake in children (4-10 years) and teenagers (11- The increase in consumption and popularity of energy drinks has raised concerns among the scientific community, governments and the general public about the health effects of these products, particularly among children and adolescents. [26, 33,34] There have been two policies to tackle sugar and caffeine content of energy drinks. In terms of sugar content, a two-tiered Soft Drinks Industry Levy (SDIL) on the production and importation of sugarsweetened drinks, including energy drinks, will be introduced in 2018.
[35] The policy is intended to drive product reformulation and lower sugar consumption. Reformulation is commonly described as efforts by the manufacturer to lower the 'unhealthy' components (e.g. saturated fat, trans fats, sugar, salt) of food and drink product at the time of production, without worsening the profile of other ingredients (e.g. increasing calorie content).
[36] The reformulated products become a replacement of an existing product (e.g. the same brand of energy drink with less sugar). The SDIL aims to incentivise manufacturers of sugar-sweetened drinks with more than 8g of sugar, per 100ml, to reduce the levels to below 8g and pay a lower tax rate and manufacturers of drinks with more than 5g of sugar per 100ml, to lower the sugar levels to less than 5g and pay no tax.
[35] This cross-sectional survey has been conducted to document levels of sugar and energy in energy drinks, as well as caffeine in the UK. The data available to monitor the energy drinks market, including whether reformulation is taking place, is generally the preserve of

METHODS
The data was collected from product packaging and nutrient information panels in January 2015 and January 2017. The survey was designed as a comprehensive survey of all energy drinks available in a snapshot in time, using one large outlet for each of the nine main grocery retailers, three beauty and health retailers and one convenience store. This study used the same study design and procedures as reported in our previous work on carbonated sugar sweetened drinks, which excluded energy drinks. [38] Definition Energy drinks were defined as any drink products with 'energy', 'energise', 'energiser', 'caffeine' and 'stimulation' in the product name or description e.g. Red Bull Energy Drink,

Data Collection
For each energy drink the data collected included the company name, product name, pack weight, serving size, sugars (g), energy (kcal) and caffeine (mg) per 100ml and per serving.
Where data was not available per serving, it was calculated from pack size and per 100ml data. Caffeine content was only collected in a sub-sample of the 2015 products.
Where products only labelled carbohydrate content in 2015 but labelled sugar content in 2017 and the content of all nutrients and ingredients were the same, it was assumed that the sugar content from 2015 was the same as 2017 (applied to only one product).
The data were double checked after entry into excel spreadsheet, and a further 5% of entries were checked against the original source in a random selection of products by the lead author.

Stores
Data was collected from each of the major UK supermarkets in store (Aldi, Asda, Lidl, Marks and Spencer, Morrisons, Sainsbury's, Tesco, The Co-operative and Waitrose), which altogether hold 93.2% of the grocery market share in the UK.
[39] Three beauty, health and food retailers (Boots, Superdrug and Holland and Barrett) and one convenience store (Costcutter) were also included since they sell a wide variety of branded energy drinks.
[40] The highest energy containing products were compared to women's daily energy intake (2000 kcal).
[41] Also, we assessed the number of products that will be taxed based on the criteria suggested in the SDIL. [35]

Comparison of the same products between the two years
For the purpose of assessing reformulation, only the products with data available in both surveys were included in this analysis. Paired t test was used to examine whether there was a significant change in the sugar and energy content of energy drinks from 2015 to 2017.
Data are reported as mean, SD, range as indicated. Significance in all tests carried out was deemed significant as being p<0·05. All data were analysed using SPSS Version 22.

Sugar, energy and caffeine content per 100ml in 2015 and 2017
A total of 75 and 49 energy drinks met the per 100ml inclusion criteria in 2015 and 2017 respectively. The average sugar content was 10.6±2.9 g and 9.7±3.0 g/100ml, with a large variation in sugar content between different energy drinks ranging from 1·9 to 15·9g and 2·1 to 16·0g/100ml in 2015 and 2017 respectively (Table 1) Table 2.

Comparison of the same products between the two years
The overall number of energy drinks formulations has decreased between 2015 and 2017, from 74 to 49 products.
Note that these averages are slightly different from those when all products were included in each year and this trend analysis reflects reductions made in the same products rather than the overall products available, giving a better reflection of reformulation, for a full list of products refer to supplementary file 1. [43]

DISCUSSION
This survey may suggest the early success of the impending SDIL. Even though the SDIL is yet to come in, manufacturers have started to reformulate their products in advance, in order to avoid paying the levy. However, we are yet to see if the SDIL will have an impact on  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  Furthermore, serving sizes of energy drinks matter since the larger the serving size, the more sugar, energy and caffeine is consumed. Indeed, typical serving sizes of energy drinks  13 are larger than other sugar-sweetened drinks; [38] this survey showed that the most common serving size was 500ml. The average sugar content in energy drink in both surveys per serving was more than an adult's entire maximum daily recommendation for sugar intake in the UK. Likewise, 86% in 2015 and 78% in 2017 of the products exceeded the maximum daily recommendation for sugar intake for a child aged 7-10 years (24g/d).
[1] The products with the highest sugar and energy content per serving could contribute up 17.5% of a woman's total daily energy intake. Therefore, in order to reduce the amount of sugar, energy and caffeine consumed from energy drinks, there is scope for introducing restrictions on larger serving sizes (500ml bottles and cans). This is a growing problem, which must be tackled, particularly since the Grocer recently reported an increase in sales of larger cans of energy drinks on the UK market. [23] Our study was based on sugar, energy and caffeine content data provided on the available energy drinks packaging labels in store on the dates of collection; hence we relied on the accuracy of the data provided on the label and the availability of products in store. It is assumed that the manufacturers provide accurate and up to date information in line with EU Regulations. However, future studies should include sugar and energy content determined through laboratory analysis to achieve a better understanding of the true sugar and energy content, and breakdown of the types of sugars. Furthermore, an analysis of what are the best selling products in the UK would allow for an accurate estimate of the potential compact of reformulation.
Also, the data collection process of caffeine content was slightly different in the two surveys. In 2015, caffeine was collected in a sub-sample of products, since the aim was not to collect caffeine at the time, but the data that was collected was included in the analysis, this may not give a full reflection of caffeine content of all products on the market in 2015.
Still, the results of this study are relevant and serve to document the sugar, energy and caffeine content of energy drinks sold the in UK, underpinning future studies and providing evidence for the Treasury and the soft drinks industry, who are reformulating. [50,51]

CONCLUSION
Factors such as an increase in sales, [24,25,30] concerning levels of sugar, energy and caffeine (assessed in this study), ease of access [33], large serving sizes, as well as limited regulation have created an environment where energy drinks could pose a major threat to public health.
In order to reduce the impact of energy drinks, sugar (and therefore energy) reformulation should continue and begin regarding caffeine content. Other measures such as a ban on the sale of energy drinks to children should be explored, while warning labels should be kept.
Can and bottle sizes of energy drinks should also be decreased to reduce the overall consumption of sugar, energy and caffeine. A reduction in sugar, energy and caffeine content and overall energy drinks consumption could be beneficial in reducing sugar, energy and caffeine intake of consumers of energy drinks.

COMPETING INTERESTS
All authors have completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years. KH