Interventions to reduce sugar‐sweetened beverage consumption using a nudge approach in Victorian community sports settings

Abstract

Objective: To assess the effectiveness of interventions using a nudge approach to reduce sugar-sweetened beverage purchases in community sports settings.

Methods: A total of 155 community sporting organisations participating in VicHealth funded programs were invited to nominate a nudge based on a traffic light approach to drinks classification. These included limit red drinks, red drinks off display, water the cheapest option, and meal deals. Sales data was collected for a predetermined period prior to and following the introduction of the nudge. Nudges were classified initially on whether they were implemented to VicHealth standards. Appropriately implemented nudges were classified as successful if they achieved a relative decrease in sales from drinks classified as red.

Results: In all, 148 organisations trialled 195 nudges; 15 (7.7%) were successful and 20 (10.3%) were appropriately implemented but unsuccessful. Limit red drinks was the most frequently attempted nudge (30.8%). Red drinks off display had the greatest rate of success (20.0%).

Conclusions: Red drinks off display was the simplest and most successful nudge.

Implications for public health: Guidelines limiting the display of sugar-sweetened beverages may be an effective means of altering consumer behaviour.

One in four Australian children and two-thirds of Australian adults are overweight or obese, placing them at higher risk of developing type 2 diabetes and cardiovascular disease.1, 2 Diets high in energy-dense foods strongly contribute to the development of obesity.3 Sugar-sweetened beverages (SSBs) are a significant contributor to sugar and energy intake and have consistently been shown to be associated with increased body weight in the Australian population.4-6 Low socioeconomic status (SES) has also been consistently associated with SSB consumption in both children and adults and the mean SSB consumption continues to rise in a subset of the population who have a lower SES.4, 7-11 Australian males, young people, and those of lower SES have reported a perception that SSBs are better value than water, with males and young people more likely to purchase SSBs as part of a ‘meal deal’.12

A 2017 survey of 3,430 Australian adults found nearly half (47.3%) had consumed SSBs in the previous week, with 13.6% consuming SSBs daily.13 In that study, higher soft drink consumption was associated with obesity, heart disease and depression, prompting the authors to recommend a comprehensive approach to reduce consumption, including the active promotion of water. There is compelling evidence that a reduction in SSB consumption can reduce the prevalence of obesity and obesity-related diseases.14

Environmental interventions including traffic light labelling (TLL) of beverages, reducing the availability of SSBs and increasing the price of SSBs are all associated with reductions in SSB sales and/or consumption.15 The TLL system classifies foods and beverages based on their nutrient composition: green represents the healthiest options; amber represents choices with some nutritional value; and red represents choices that are high in sugar, salt or unhealthy fats.16 Interventions using the TLL classification may categorise foods to support consumers to make informed choices or to guide policies on the display of foods and drinks and their availability for purchase. In Victoria, Australia, the State Government's Healthy Choices Guidelines includes sample policies, implementation plans and toolkits for sport and recreation settings.16

In community sports settings, energy-dense food and drink choices, including SSBs, are often the norm.17, 18 However, non-SSB consumption has been shown to increase in response to the increased availability of non-SSB drink options in Canadian and Australian community recreation settings.19, 20 An Australian intervention incorporating multiple strategies (availability and positioning of SSBs vs. non-SSBs, pricing incentives, meal deal promotions, signage, posters, policies, and promotion through coaches) across 85 football clubs was successful in increasing non-SSB sales and consumption.21 However, it is uncertain whether this was related to any particular strategy used, and all outcomes were self-reported with no study-specific measurement of sales data. Detailed sales data were collected from 16 aquatic and recreation centres in Victoria, Australia, and found that removal of all SSBs classified as red drinks led to a decrease in red and total drink sales, but was not accompanied by an increase in sales of green drinks despite the use of complementary strategies including drink placement, traffic light labels at point of sale and promotional posters.22 Despite promising results, these studies leave gaps in our understanding of which strategies are most acceptable and successful in reducing SSB consumption.

Many of the interventions described above may be classified as nudges. In behavioural economics, a nudge is a term used to describe how one can influence others’ choices by organising the context in which they make decisions.23 Nudges align with the mid-levels of the Nuffield intervention ladder where choice is guided by changing the default, or through the use of incentives or disincentives.24 However, nudges must also be easy and cheap to avoid, and not go so far as to restrict or eliminate choice.25 Thus, interventions using a traffic light approach to classify drinks and manipulating the display or promotion of products to influence individual purchasing behaviours are considered nudges, while interventions removing SSBs from sale altogether are not. Nudges have been shown to be acceptable to both consumers and retailers, perhaps because healthy options are promoted without restricting choice.26, 27 Nudges have been used most frequently in public health interventions to influence eating and drinking behaviour.28 However, the quality of studies performed has typically been limited by small sample sizes, minimal methodological detail and a lack of use of reporting guidelines.28, 29

Practical solutions are needed to address concerning SSB intake among Australians. Community sport settings, where SSBs are widely available, present an opportunity to trial nudges in a targeted setting. However, it is not clear which nudges are most practical to implement and successful in reducing SSB consumption. Furthermore, other studies have been conducted in limited settings (Australian football clubs; aquatic centres), and have used either multiple combined nudges21 or restricted choices.22

The Victorian Health Promotion Foundation (VicHealth) adopted a nudge approach to its initiatives that aimed to reduce SSB consumption in community sports settings. Halpern's TEST (target, explore, solution, trial) framework was used to guide the development and implementation of a trial.30 Ultimately, Victorian community sporting clubs/venues were invited to undertake projects using one of four nudge approaches that aimed to increase the consumption of water and reduce the purchase of SSBs. The aim of this research was to assess the effectiveness of interventions using a nudge approach to reduce SSB purchase as an indicator of consumption in community sports settings.

Methods Framework This study was designed using Halpern's TEST framework.30 This involved the following steps:

Target (define the outcome): To reduce SSB consumption in community sports settings (using purchase as a proxy measure)

Explore (understand the context): Allow individual organisations to design and select an intervention that they believe will best suit their context

Solution (build your intervention): Refine to four evidence-based nudges that align with organisation preferences and capabilities

Trial (test, learn and adapt): Formal trial, dissemination of results and continued support

Participants

Sporting clubs in Victoria, Australia, involved in the State Sport Program (SSP; April 2016 – June 2017), Regional Sport Program (RSP; April 2016 – September 2017) and Water Initiative Program (WIP; January – August 2017) (n = 155 total) were invited to participate. Henceforth, these clubs will be referred to as the ‘intervention organisations’, for ease of reporting and to distinguish them from the ‘funded organisations’, i.e. the State Sporting Associations, Regional Sports Assemblies, and Local Government Authorities (LGAs) that received funding from VicHealth to implement the program. All intervention organisations were deidentified and allocated a numeric code for analysis and reporting. The socioeconomic status of the population attending the club/facility was represented using the Socioeconomic Indexes for Australia (SEIFA) Index of Relative Socio-economic Advantage and Disadvantage.31 This scale ranks areas identifiable by postcode into deciles, with 10 being the most advantaged.

Intervention

Intervention organisations were asked to nominate their own nudge initiative(s). The first group of intervention organisations, in the explore phase, used a range of self-determined strategies based on broad guidance from VicHealth. This was refined following the first program evaluation in April 2017 at the solution phase, with new intervention organisations choosing from one of four possible options that had been developed and documented by VicHealth and La Trobe University for the trial phase. In the Limit red drinks initiatives, clubs restricted the number of red drinks available to no more than 20% of drinks for sale, with green drinks displayed at eye level. The Red drinks off display required clubs to keep beverages classified as red drinks hidden from consumers, for example, behind fridge decals or posters or behind the counter, but permitted the sale of these beverages where requested. Clubs could choose to make Water the cheapest option by reducing the price of water or increasing the sales price of other beverages accordingly. The Meal deal involved packaging water with a food item to provide a relative cost saving to consumers. All options were designed to provide a nudge to consumers with beverage classification based on the Victorian Government's Healthy Choices Guidelines traffic light system.16

Intervention organisations were supported by their funded organisation to implement the nudge, and also had direct access to two members of the research team for further support. A representative of the funded organisation met in person with a member of the intervention organisation to explain the nudge concept and process. To encourage participation, some funded organisations offered intervention organisations an incentive. Incentives varied depending on the funded organisation and included direct payments, nutrition education sessions, sporting equipment, canteen equipment and canteen stock (for example, a supply of green drinks). Intervention organisations were supported with resources to assist with the implementation and evaluation of the nudge, such as the ‘Nudge Toolkit’, which comprised:

a simple table detailing the research process for the stocktake method∗ for both baseline and nudge implementation periods (∗stock was calculated by counting all stock on display and in storage at the start of the specified period, adding all stock purchased during the specified period, then subtracting all stock on display and in storage at the end of the specified period);

‘Facility Information’ identifying minimum criteria for site selection;

a ‘Photo Taking Guide’; and

simple documentation for recording baseline and nudge data per project location.

The resources in the Nudge Toolkit were further supported by communication tools developed by VicHealth regarding drink classification and the nudge interventions. Education sessions with external nutrition experts were provided to upskill intervention organisations prior to nudge implementation.

Data collection

Intervention organisations were asked to collect baseline sales data from their canteen or retail outlet for a pre-determined period of time (e.g. four weeks) while they continued business as usual. Where an intervention organisation had suitable baseline sales data from a previous period of operation of the same duration as the nudge, which allowed drinks to be classified accordingly, this was used.

The selected intervention was then deployed and implemented for a duration equivalent to the baseline data collection. Where multiple nudges were trialled by the same intervention organisation, they were implemented in sequence (i.e. one at a time), and each nudge was implemented for a duration equivalent to the baseline data collection. Photos were taken to document the intervention at baseline and once the nudge was implemented, and sales data were collected from the canteen or retail outlet throughout the intervention period. The minimum duration of baseline and intervention was related to the context of the site, with no changes in operation or pricing made throughout the baseline and intervention phases, and no major events or playoffs (finals) held during either phase. All sites were required to provide data on total attendance and daily weather (temperature in degrees Celsius) throughout both phases to identify any confounding variables. Only drinks sold from the fridge were included in analyses; hot beverages (tea, coffee, hot chocolate) were not included in the nudges or the analyses. Hot drinks were excluded because: i) canteens were unable to confirm the volume of beverage served relative to the TLL rating; ii) many canteens had only one item on the register for all hot drinks, making it impossible to determine the breakdown of the type of hot drinks sold; and iii) some venues had private coffee vans that would set up during games/matches, and as such the majority of hot beverages were sold by a third party provider that was not part of the intervention or the evaluation.

Evaluation

At the end of the intervention, a member of the research team reviewed all data collected, including sales data, attendance, weather and photographic evidence of the intervention. Projects were classified based on their implementation initially, and then on their outcomes (Figure 1):

image

Categorisation of Nudge initiatives.

Successful: implemented the intervention to VicHealth standards and had an increase in the percentage of total sales from green/amber drinks and/or a decrease in the percentage of total sales from red drinks compared to the baseline period.

Unsuccessful: implemented the intervention to VicHealth standards and did not have an increase in sales of green/amber drinks and/or a decrease in sales of red drinks relative to total sales compared to the baseline period (including projects with no change reported).

Not implemented according to VicHealth standards: based on photographic evidence

Unable to be determined: due to unclear dates, missing or poor quality photographs, missing sales, attendance or weather data, or low volumes of sales (<150 drinks per week).

Discontinued: intervention not completed due to stakeholder breakdown or interference from product company sales representatives.

Statistical methods

No a priori power calculation was performed as the sample size was determined by the total number of eligible organisations willing to participate in the study. All analyses were performed using Microsoft Excel version 16.45. The types of intervention organisations are reported using frequency analysis. The frequency of each type of nudge was stratified by evaluation status (successful, unsuccessful, not implemented to standards, unable to be determined, discontinued).

Characteristics of successful and unsuccessful nudges reported at baseline and post-intervention include total drink sales, total number and proportion of sales from green, amber and red drinks, outdoor temperature, attendance, profit margin, total profit and profit per week (reported for each intervention and as mean± standard deviation), and SEIFA deciles (reported for each intervention and as median and interquartile range). Characteristics of nudges classified as not implemented to standards, nudges where success could not be determined and discontinued nudges, were not reported.

For each characteristic (other than SEIFA), the change from baseline to post-intervention was calculated and expressed in total value for temperature, and as percentage change for all other variables. Differences between baseline and post-intervention values were assessed using paired samples t-tests. Differences in percentage change between groups were assessed using independent samples t-tests.

Ethics approval for the conduct of this study was provided by the La Trobe University Human Ethics Committee (E15/081). Participation by intervention organisations was voluntary and all provided consent to participate in this research.

Results Sample

A total of 148 sporting clubs, associations and facilities (i.e. intervention organisations) participated in the nudge trials; 63 of these were football (AFL) and/or netball clubs (Table 1). In total, the 148 intervention organisations trialled 195 nudges by implementing one (n=105), two (n=39) or three (n=4) nudges in sequence.

Table 1. Intervention organisation breakdown for Nudge trials.

Club/Association/Facility

Total

Baseball

3

Basketball

9

Bowls

6

Football (AFL)

8

Football (AFL) and Netball

45

Gymnastics

8

Hockey

4

Indoor multipurpose

19

Netball

10

Outdoor swimming pool

7

Soccer/Football

10

Tennis

4

Touch Football

6

Other

9

TOTAL

148

Note: ∗Includes: Adventure playground (1); Cricket (1); Cycling (1); Golf (2); Motor vehicle (1); Squash (2); and Volleyball (1) Evaluation of interventions

Fifteen of the 195 nudges implemented the intervention to VicHealth standards and had an increase in sales of green/amber drinks and/or a decrease in sales of red drinks; as such, were deemed successful (Table 2).

Table 2. Success of each type of nudge intervention.

Limit red drinks

Red drinks off display

Water the cheapest option

Meal deals

Guidelines

Promotions

Not selected

Total

Successful

5

5+1

1

2

1

0

0

15 (7.7%)

Unsuccessful

3

1

8

3

0

5

0

20 (10.3%)

Not implemented to standards

4

1

0

0

0

0

0

5 (2.6%)

Unable to be determined

43

20+1

28

14

4

26

0

136 (69.7%)

Discontinued

5

1

3

0

0

0

10

19 (9.7%)

Total

60 (30.8%)

30 (15.4%)

40 (20.5%)

19 (9.7%)

5 (2.6%)

31 (15.9%)

10 (5.1%)

195

Success rate

8.3%

20%

2.5%

10.5%

20%

0%

0%

Notes: ∗removed red drinks instead of red drinks off display ∼ 10 projects were discontinued before a Nudge was selected Frequency and success of nudges

Limit red drinks was the most frequently attempted nudge intervention (30.8%), while red drinks off display and guidelines experienced the greatest rate of success (both 20.0%) (Table 4). Water the cheapest option and meal deals were less popular (20.5% and 9.7% of nudges, respectively), and when chosen as a nudge, also less successful (success rates 2.5% and 10.5%, respectively). Promotions were used only in the pilot (explore) group of nudges, were inconsistent in their implementation, and had no successful outcomes.

Characteristics of successful and unsuccessful nudges

The characteristics of successful and unsuccessful nudge projects are presented in Table 3.

Table 3. Results of successful and unsuccessful nudge projects.

Code

Nudge type

SEIFA decile

Initiative duration

Change in % of total beverage sales

Attendance (n, total across the period)

Temperature (mean °C)

Total drink sales (n)

Profit margin ($)

Profit ($)

Green

Amber

Red

baseline

Nudge

% change

baseline

Nudge

change in mean °C

baseline

Nudge

% change

baseline

Nudge

% change

baseline

Nudge

% change

Successful Nudge projects

1

LRD

9

6

3.8

12.4

−16.2

1,366

1,232

−9.8

19

12.8

−6.2

823

659

−19.9

1.99

2

0.5

1,636.2

1,315.6

−19.6

2

LRD

2

4

8.5

−1.1

−7.4

17,937

18,554

3.4

IV

IV

IV

500

427

−14.6

1.55

1.63

5.2

1,679.4

1,787.7

6.4

3

LRD

7

2

17.2

−15.7

−1.5

600

600

0

18

8

−10

1,531

661

−56.8

N/A

N/A

N/A

N/A

N/A

N/A

4

LRD

8

2

1.9

0.3

−2.3

1,550

750

−51.6

18.5

12

−6.5

508

199

−60.8

N/A

N/A

N/A

N/A

N/A

N/A

5

RDOD

7

4

7.2

10.9

−18.1

1,250

2,050

64

16.6

15.5

−1.1

484

510

5.4

1.75

1.85

5.7

851.6

944.9

11

6

RDOD

4

6

11.8

1.8

−13.6

21,576

23,178

7.4

21.1

20.6

−0.6

2,079

1,950

−6.2

留言 (0)

沒有登入
gif