A paid-for ad on Instagram for DNAfit, seen on 29 September 2019, featured an animated double helix and captions that stated “We’re DNA. We know all about your body. Fast twitch muscle fibres give you power. Slow give you endurance. And that’s not all we can tell you. Order your kit now at DNAfit.com”. A caption under the animation stated “Unlock the secret to your ideal diet, vitamin need and exercise response”.

Issue

The complainant, who believed DNA testing could not be used to determine an individual’s diet, vitamin and exercise needs, challenged whether the claim “Unlock the secret to your ideal diet, vitamin need and exercise response” was misleading and could be substantiated.

Response

DNAfit Life Sciences Ltd (DNAfit) said the basic operation of the test involved consumers collecting a DNA sample using a swab provided as part of the DNAfit kit. The swab was then analysed in a lab to ascertain changes in specific genes – single nucleotide polymorphisms (SNPs) – that allowed DNAfit to produce a report detailing consumers’ specific nutritional and fitness needs and advising lifestyle changes they may wish to adopt to suit those needs.

They said there was a body of evidence showing ‘nutrigenetic’ information – information that related to the effect of genetic variation on dietary response – could lead to improved outcomes and better adherence compared to standard dietary and fitness regimes. By way of substantiation for that claim, they provided 31 peer-reviewed papers. They said the ad was no longer appearing.

Assessment

Upheld

The ASA considered consumers would understand the claim “Unlock the secret to your ideal diet, vitamin need and exercise response” to mean DNAfit were able to provide consumers with effective personalised exercise and nutrition information based on sequencing of their DNA and that if they made changes in line with that information they would achieve health benefits. While the ad did not refer to any specific benefits, it mentioned different types of muscle responses, and we considered that the overall impression of the ad was that the personalised information provided by DNAfit would help consumers improve their general health and fitness.

We noted DNAfit offered a range of packages focused on, among other things, weight loss, and improving fitness and strength, and the provision of advice in the form of genetically-matched training programmes and personalised meal plans and recipes. Consumers received action points pertaining to the information received, and also the option of a complimentary phone call with an expert who could advise them further. We therefore considered, given the broad nature of the claim in the ad and the number of different packages on offer, that DNAfit should hold scientific evidence in support of the claimed link between DNA sequencing and effective personalised exercise and nutrition advice, and that evidence should measure tangible benefits related to fitness, for example weight loss or strength. We considered that DNAfit’s audience would likely consist of individuals with a range of fitness levels and goals, including those who were motivated to improve their health and fitness, and those who were already reasonably fit.

DNAfit provided four relevant studies and a systematic review they said evidenced the effectiveness of implementing genetic information in diet programmes, termed ‘nutrigenetics’. They provided a paper of guidelines that could be used to evaluate scientific validity and evidence for nutrigenetic dietary advice. They also provided a number of studies that related to the functions of different SNPs which they believed demonstrated links between different genotypes and the nutrient and exercise requirements. Those informed the range of genes on which they did their testing. We noted the guidelines on evaluating scientific evidence for nutrigenetic dietary advice, which DNAfit said they used to inform their decisions about which genes to include in their testing. DNAfit said for each gene included in their tests, a minimum of three peer-reviewed studies using human subjects that showed an effect modifiable by an environment or lifestyle change was required. We reviewed the studies related to SNPs and how they related to the claim under investigation and understood DNAfit had provided them as a means of demonstrating the link between different genotypes used by DNAfit in their testing and consumers’ nutrient and exercise requirements. Some studies referred to links between disease risk and certain genotypes – for example, one study reported that men with the SOD2 Ala/Al genotype were more likely to have an increased risk of prostate cancer, and that increased consumption of selenium could mitigate this.

We understood that randomised controlled trials were not necessarily the most appropriate methodology to measure incidence of disease, because of the length of time it would take to measure health outcomes in many cases. However, we considered that other indicators of improved health and fitness would be measurable in the shorter term, and DNAfit should therefore be able to demonstrate through relevant, controlled trials that personalised nutrigenetic advice resulted in improved outcomes in relation to such indicators. One study from 2007 related to the use of nutrigenetics in improving long-term weight loss. The study used a test group of 50 participants from a weight loss clinic who were given nutrigenetic diet advice, and a control group of 43 individuals from the same clinic who were advised to follow a standard low-glycaemic Mediterranean diet. Both groups were given exercise recommendations. The study found a significantly greater reduction in Body Mass Index (BMI) in the intervention group compared to the control group at 300 days. We noted that the results at 100 days reported no significant difference. We also noted there was also a significantly greater reduction in fasting blood glucose level in those at pre-diabetic levels, in favour of the intervention group.

We considered that while the study suggested positive results, there were a number of key flaws. There was no blinding of the participants involved, and the difference between the groups could be explained by one group being motivated by having a personalised diet rather than the diet itself being objectively better. Additionally, dietary adherence was not described.

A more recent study from 2017 examined whether the inclusion of genetic information to personalise a patient’s diet could improve long-term weight management outcomes and used two groups, of 53 and 61 participants respectively, who self-selected either a ketogenic or a nutrigenetic diet. The self-selection indicated the study was not blinded, which in turn could have had an influence on the results and conclusions drawn. The study almost exclusively used obese participants ? with one participant in the overweight category. The study found a significant difference between the groups, after two years, in participants’ cholesterol and weight levels, compared to their weight and cholesterol levels at the beginning of the study. However, the authors stated it was uncertain whether that was due to biological mechanism or dietary adherence, so the reason for the significant difference between the two groups was unknown. Where dietary adherence was concerned, we noted that had not been assessed as part of the study, so it was unclear how well participants had adhered to the diet in both the test and follow-up periods. We considered the “control” diet would have placed a greater burden on participants, making them more likely to divert from it. Another, double-blind, study examined the effect of disclosing genetic data on its participants’ nutrient intake. The participants consisted of an intervention group of 92, who were provided with nutrigenetic advice, and a randomised control group of 46, who were given general dietary advice. We noted the nutrients subject to the study were caffeine, vitamin C, sodium, and added sugar. We noted the study found a significant decrease in sodium intake after 12 months in the intervention group, compared to the control group. The study was not designed to measure health outcomes, but looked at whether receiving personalised information made people more likely to comply with dietary advice.

The systematic review looked at the existing literature on the use of genetic testing in lifestyle behaviour change. The review concluded that, while it was possible to facilitate behaviour change through the provision of genetic intervention, but because of the multifaceted nature of behaviour and behaviour change, the authors could not make any broad statement about the impact of nutrigenetic information on dietary behaviour change. We also noted the author of the review concluded that the majority of the studies were of low quality. DNAfit provided two studies they said evidenced the effectiveness of implementing genetic information in exercise programmes, termed “lifestyle genomics”. We noted both studies used small sample sizes – one of 28 and 29 athletes across two study groups, and the other 42. The participants were exclusively male, athletic and between 16 and 20 years of age. The first study gave a group of participants fitness advice that matched their genetics, and another group mismatched advice. Participants were blind to whether or not they were receiving genetic-matched advice. Each group was then assessed on its explosive power and aerobic fitness. The study found that the matched group significantly improved on both of those aspects, compared to the mismatched group among the footballers. The study concluded that further research using untrained and unfit participants was needed.

We considered that the study showed some positive results, but was small in scale and did not account for potential confounding factors such as additional training the participants may have been doing at the time of the study. We also considered that without further robust results, a study of that size was unlikely to be sufficient to substantiate the claims in isolation. The second study divided a group of 42 football players into three groups – high, medium and low expected improvements in adaptation to aerobic exercise – based on genetic modelling. Participants were not blinded and there was no control group because the researchers felt it would be unethical to ask a control group of football players to refrain from training for the study. The study found a significant difference in post-training performance between the ‘high’ and ‘low’ groups. However, we noted that the study stated that further research with, among other things, female participants was needed. The study cited the small sample size used as being a limitation. We noted the study had been carried out by, among others, employees of DNAfit. We also noted the lack of a control group in the study and considered the researchers could have put the control group on a programme based on standard training, which would have mitigated the “ethical” concerns surrounding asking players to refrain from training completely.

We welcomed DNAfit’s assurance that the ad was no longer appearing. However, overall, we considered that the body of evidence provided by DNAfit was not sufficient to substantiate the claim “Unlock the secret to your ideal diet, vitamin need and exercise response” as consumers were likely to understand it in the context of the ad. We concluded that the ad was misleading.

The ad breached CAP Code (Edition 12) rules 3.1 (Misleading advertising), 3.7 (Substantiation), and 12.1 (Medicines, medical devices, health-related products and beauty products).

Action

The ad must not appear again in the form complained about. We told DNAfit Life Sciences Ltd that their future marketing communications must not state or imply they could provide consumers with effective personalised exercise and nutrition information or advice based on sequencing of their DNA that would result in improved health and fitness outcomes unless they held documentary evidence to that effect.