Many thanks to Dr Melissa Stoneham and the Public Health Advocacy Institute of Western Australia for this latest instalment of Journal Watch.


Melissa Stoneham writes:

Vending machines – they are certainly versatile. You can get almost anything in them today. In Perth and Adelaide right now, 4 hot chip vending machines are being trialled. In Italy in March this year, a dough vending machine was launched allowing you to make pizza from scratch! Occasionally you see a healthy vending machine like the Queensland based FruitBar that was launched in Mackay by a local fruit growing family. But in most cases vending machines are full of sugary and refined products.

Yet vending machines are convenient – they are generally accessible and quick to use. It is true though that most of the vending machines are full of energy dense foods, so it is likely they are contributing to the obesity epidemic in Australia. The prevalence of overweight and obesity has been steadily growing among Australians for the past 30 years. In 2011–12, around 60% of Australian adults were classified as overweight or obese, and more than 25% of these fell into the obese category.

So bearing in mind the convenience of a vending machine and the increasing rate of obesity, the question to be asked is: if vending machines provided nutritional information, would this influence the buyer to select lower kilojoule items?

A study in North Carolina, led by Deidre Dingman from the University of North Carolina, tested this very question and included nutrition information, interpretive labels, and promotional health communication at vending machines to see if this would lead consumers to choose lower calorie snacks that contained less salt, sugar and saturated fat. The study, which was conducted in a university setting, aimed to identify if the average calories sold per snack reduced, as well as identifying the proportion of snacks sold that contained fewer calories and less saturated fat, sugar and sodium.

The study covered 18 vending machines located in University residence halls, servicing 3,850 students. There were more males than females and a higher proportion of first year students living in the residences. The study nominated 9 intervention and 9 control vending machines to allow comparisons throughout the 8 week study. During Week 5 of the study,  a poster board was placed next to the vending machines and listed the Nutrition Facts Panel (as required on packaged food in the US) for eac hproduct in that vending machine. They also highlighted 5 items in the machine that met certain per package nutrition criteria (less than 200 calories, 2 grams or less of saturated fat, 0 grams of trans fat, 7 grams or less of sugar and less than 300mg of sodium per package). An email was forwarded to all students advising that this information was now available.

Snacks were defined using criteria (similar to those recommended by the Institute of Medicine) to define individual snacks as a Better Choice (BC). Stickers with the letters BC were placed inside the machine next to these snacks. Available snacks ranged in calories from 100 to 470 each.

The Nutrition Facts Panel labels, a BC symbol, and the criteria were all placed on the posters that were located next to the vending machine.

Sales data on each snack item sold was provided by the vendor. At the end of the data collection period, a link inviting students to participate in a supplemental survey was forwarded. The survey asked if the student had received the original email and also enquired as to whether they recall seeing the information at the vending machine.

From the sales data, the average calories per snack sold and the proportion of Better Choice snacks sold were calculated. It was found that the average calories per snack sold across the 9 intervention machines for the 4 weeks of pre-intervention sales was 252 and for the 4 weeks of post-intervention sales, the average was251. The average calories per snack sold across the 9 control machines at the pre-intervention time point was 217 and at post intervention the average was 225. Nothing major there!

For the survey, 45% of the students responded. Of these 670 students, only 16% recalled getting the initial email relating to the availability of the health information at the vending machines (n=106). Of those only 63% (n=67) said they had read the email.  Fifty-six per cent of students living in the intervention residency halls (n=364) said that they noticed the on-site nutrition information, but 60% (n=192) said it did not influence their purchasing decisions.

So this study found that providing nutritional information including nutrition information, an interpretive label, and a health communication/promotional message did not reduce average calories per snack item purchased.

I suspect there are many reasons for these findings. The target group, being time poor, broke and stressed university students, may simply have opted for the easiest rather than the healthiest option. Or maybe it is food labels themselves. I am an advocate for food labels as they allow people to make healthy and safe food choices. But for food labelling to have an impact, consumers must first read the label, understand the information and how to use it, and then make decisions about their food consumption based on the information.  At the moment,there’s so much nutritional information on food labels that they can sometimes be confusing. In fact, they can be so unclear that even Food Standards Australia has designed a poster on how to read food labels.

Another option to combat poor vending machine choices may be to do Illinois has done, where a Bill was introduced in the state legislature urging the Illinois Department of Transportation to add healthy snacks to vending machines in rest areas along state highways. The Bill states that there would have to be at least three healthy snacks for every ten snacks in each vending machine and two out of the three healthy snacks would need to be made in Illinois.

Of course, the holistic approach to address issues such as the one identified in this research is to advocate for environments that don’t actually promote obesity – including strategies such as getting people active, the promotion of riding or walking to school or work, the integration of green space and vegetable gardens into communities, fewer fast food premises in communities, reducing fast food marketing and giving people choices so that healthy options are easy to pick – in all settings – communities, schools, health services and workplaces. We need implementable policies that commit long term funding to ensuring there’s plenty of fruit, water, and healthy take-away food – not just high-fat, high-salt, high-sugar alternatives. Now that would really give people the opportunity to make the healthy choice, the easy choice.


Does providing nutrition information at vending machines reduce calories per item sold? Deirdre Dingman, Schulz MRWyrick DLBibeau DLGupta SN; Journal of Public Health Policy; Vol 36, Issue 1; pages 110-122.

About JournalWatch

The Public Health Advocacy Institute WA (PHAIWA) JournalWatch service reviews 10 key public health journals on a monthly basis, providing a précis of articles that highlight key public health and advocacy related findings, with an emphasis on findings that can be readily translated into policy or practice.

The Journals reviewed include:

Australian & New Zealand Journal of Public Health (ANZJPH)

Journal of Public Health Policy (JPHP)

Health Promotion Journal of Australia (HPJA)

Medical Journal of Australia (MJA)

The Lancet

Journal for Water Sanitation and Hygiene Development

Tobacco Control (TC)

American Journal of Public Health (AMJPH)

Health Promotion International (HPI)

American Journal of Preventive Medicine (AJPM)

These reviews are then emailed to all JournalWatch subscribers and are placed on the PHAIWA website. To subscribe click to Journal Watch click here.

(Visited 32 times, 1 visits today)