Implementation and sales impact of a capacity building intervention in Australian sporting facility food outlets: a longitudinal observational study

Abstract

Background: Few studies have explored behavioural and financial impacts of retail initiatives after 2 years to address the unhealthy food environments common in local government sporting settings.

Aims: To evaluate the impact of a 2-year local government capacity building intervention in sporting facility food outlets on (1) the healthiness of refrigerated drink choices available and visible to customers, (2) healthiness of refrigerated drinks sold and (3) refrigerated drink revenue.

Methods: 52 sporting facilities within 8 local governments from Victoria, Australia, participated in an intervention between March 2018 and February 2020 by limiting 'red' (least healthy) drinks to ≤20% of refrigerator display and increasing 'green' (healthiest) drinks to ≥50% of display. Mixed models assessed changes in mean percentage of 'red', 'amber' and 'green' drinks displayed over time, compared with baseline.Facilities provided electronic weekly itemised sales data (December 2015 to February 2020). Weekly volume of 'red' or 'green' drinks sold as a proportion of total drinks sold, and total refrigerated drinks revenue were compared preimplementation and postimplementation using mixed models (seasonal facilities), and mixed-effect interrupted time series models (non-seasonal facilities).

Results: Display of 'red' drinks decreased by mean -17.1 percentage points (pp) (95% CI -23.9 to -10.3) and 'green' drinks increased 16.1 pp (95% CI 9.30, 22.9) between baseline and 18-month audits.At nine seasonal facilities, compared with the summer preimplementation, the mean volume of 'red' drinks sold decreased by -19.0 pp (95% CI -28.6, to -9.51) and refrigerated drink revenue decreased by-AU$81.8 (95% CI -AU$123 to -AU$40.8) per week. At 15 non-seasonal facilities, by February 2020, the volume of 'red' drinks sold decreased on average by -11.0 pp (95% CI -21.6 to -0.41) with no change in drink revenue.

Conclusion: Reducing the display of unhealthy drinks can be an effective public health policy to improve the healthiness of customer purchases, provided there is consideration of potential impacts on revenue.

Keywords: Dietary patterns.

Figure 1

‘Water in Sport’ project implementation and data collection timeline. ^aProject officer employment start dates ranged between 1 March and 18 June 2018, and employment finish dates ranged between 20 December 2020 and 30 October 2021.

Figure 2

Flow diagram of facilities in the ‘Water in Sport’ evaluation. ^aNumber of facilities that expressed an interest to participate in the intervention and were included in the initial application for funding to VicHealth by local governments. LGA, local government area.

Figure 3

Percentage ^a of refrigerated drinks on display by traffic light classification over time^b, in 44 facilities. ^a Marginal means and 95% CI estimated from multilevel models adjusted for season, seasonality of facility and outlet type. ^bHealthy Choices guidelines target were to display no more than 20% ‘red’ drinks (eg, sugary soft drinks), at least 50% ‘green’ drinks (eg, sparkling waters) and the remainder ‘amber’ drinks (eg, diet soft drinks).

Figure 4

Weekly sales of ‘red’ drinks and drinks revenue, by summer season, in nine seasonal facilities. ^aMarginal means and 95% CI estimated from multilevel model adjusted for calendar month, maximum weekly temperature. ^bMarginal means and 95% CI estimated from multilevel models adjusted for calendar month, maximum weekly temperature, size of facility.

Figure 5

Weekly sales of percentage of ‘red’ drinks sold ^a with sales time trend, in 15 non-seasonal facilities. ^a Predicted values and sales trend using weekly data, estimated at mean values of the covariates from models adjusted for calendar month and maximum weekly temperature. Study periods: Preimplementation (4 January 2016 to 4 March 2018; Implementation (5 March 2018 to 30 September 2018); Postimplementation (1 October 2018 to 24 February 2020).