Analysis: nutrition apps have the potential to be a useful support in improving behaviour when it comes to food, but more research is required

By Cathal O'Hara, UCD

Now that the new year is not so new, how are those resolutions doing? We all want to change our behaviour; to exercise more or eat more healthily. If we know these things are good for us, why is change so difficult? In short, knowing something is good for us isn't enough. Knowledge is simply a tool in behaviour change and it’s not all that useful on its own. Knowledge is good, of course, and we need to know what’s good for us and what’s not, but we have to put this knowledge into action if we want to change our behaviour. This can be a difficult thing to do.

So what helps? In the area of diet and nutrition, self-monitoring, or tracking what you eat, has been found to be useful in behaviour change. The other thing we know is that dietary advice is far more effective when it is personalised to be relevant specifically for the person receiving that advice

Nutrition apps can potentially provide a convenient way to both track diet and provide personalised feedback. But to be helpful, the app needs to get the right information about the user in order to personalise advice. This is more challenging than you might think and making it easier for people could be hugely helpful when it comes to diet and behaviour change.

We need your consent to load this rte-player contentWe use rte-player to manage extra content that can set cookies on your device and collect data about your activity. Please review their details and accept them to load the content.Manage Preferences

From RTÉ Radio 1's Today with Claire Byrne, Louise Reynolds from the Irish Nutrition and Dietetic Institute on how nutrition and diet affects your heart health

The popularity of these apps may be surprising to some. A survey of 2,382 European adults found that over half had used a nutrition app of some description. Nearly 30% of respondents reported tracking their diets and more than 40% of these used apps to do so. Nearly 30% of those who reported never having used a nutrition app said they were too time consuming to use. Users valued ease of use above everything else and wanted the app to be certified, validated and intuitive to use.

Analysis of data from 41 different studies with over 6,300 people using nutrition apps found small to medium positive changes were seen on people’s nutrition behaviours such as improved fruit and vegetable intake. Similar positive changes were seen in physical measures such as blood pressure and cholesterol levels. These changes were similar to those seen in face-to-face (non-app) approaches used to improve nutrition. The advantage of an app is that it is much cheaper and more accessible than other options.

The use of nutrition apps for improving health holds promise, though there is a lack of transparency about accuracy and efficacy among existing apps. Further research and development of these apps that is transparent and open to scrutiny by regulatory authorities and independent experts is necessary to ensure that only the highest quality and evidence-based services are being provided.

The average accuracy in correctly identifying foods across a range of different foods is up to 65% and more work is needed here

Diet tracking doesn’t have to be onerous. In fact, too much focus on tracking and numbers is unlikely to be helpful. Tracking our diets for between three to seven days gives a reasonably good idea of our typical intakes. Recording for any longer is unlikely to provide a useful amount of additional information relative to the time and effort required. This should provide enough information to choose two or three positive changes that can be maintained. While there will always be good days and bad days when making changes, choosing a smaller number to focus on will make things easier. If we wish to make further changes, this process of tracking can be repeated periodically.

My research revolves around making this kind of diet tracking easier. Many apps ask users to take pictures of their meals. The apps then use computer vision to analyse what individual foods are in those meals and analyse the resulting data. The average accuracy in correctly identifying foods across a range of different foods has been reported as up to 65% and further work is required if this method is to become commonplace.

I am currently working to see if we can make this process simpler. Can we train our computer vision technology to look at a photo of a meal as a whole and provide feedback on that, rather than focusing on breaking the meal into individual foods and providing a calculation based on adding all those components together? This could allow a greater focus on the patterns of the meals we consume and achieving a healthy variety in those patterns, rather than placing such great emphasis on the tracking of specific nutrients. We are currently seeking volunteers to participate in a study that will help lay the foundations for this work and they will be offered free personalised feedback on their dietary intakes.

Having been identified as a priority in the digital transformation at both EU and national levels, it is inevitable that digital health tools will become more integrated in time into our existing healthcare systems, playing a greater role in how we interact with healthcare professionals in the future. Ideally, we should use nutrition apps in a way that allows us to achieve long-term and positive changes to our diets without becoming dependent on an app for daily use over prolonged periods.

More work is needed to establish approval and regulatory processes to ensure that such apps are safe and effective. But with continued research, they have the potential to be a useful support in the effort to improve behaviour when it comes to food.

Cathal O'Hara is a PhD student at the Insight SFI Research Centre for Data Analytics, the UCD Institute of Food and Health, and UCD School of Agriculture and Food Science.


The views expressed here are those of the author and do not represent or reflect the views of RTÉ. If you have been affected by issues raised in this article, support information is available online