select search filters
briefings
roundups & rapid reactions
before the headlines
Fiona fox's blog
A conference abstract presented at the European Congress on Obesity looks at the effects of vegan diets on cardiometabolic health.
Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:
“It’s pretty difficult to assess the quality and potential impact of this piece of research, because (so far) rather few details are available. We have a media release, a fairly brief summary (abstract) of the findings, and a poster that will be presented at a conference, giving some selected details. The release points out that the full research paper on this study has just been accepted by a scientific journal, and we’ll know quite a lot more after that comes out.
“The research is a systematic review and meta-analysis of findings from previous randomised trials that compared various outcomes in people who were overweight, or who had Type 2 diabetes (or prediabetes). In all the trials that were considered, some people were on a vegan diet for at least 12 weeks, and others were on some sort of control diet, though exactly what that control diet was did differ quite a lot between the trials. The researchers looked at average differences between those on the vegan diet and the control diet, on several different measures (listed in the media release). They found that, on average, those on the vegan diets had lower body weight and BMI than those on a control diet. There were also differences favourable to the vegan diets on a standard blood sugar test (related to type 2 diabetes) and on two measures of cholesterol in the blood, though the media release is careful to point out that the differences on these measures were ‘rather small’.
“Overall, from the limited information available, it does look as if there is evidence that reductions on body weight and in BMI were, on average, greater for those on the vegan diets than those on the control diets. But I’d advise a lot of caution in interpreting those findings. There’s nothing here to say that these differences were caused by the fact that the vegan diets contained only plant-based foods, while the control diets did not. For one thing, we don’t know much about what the control diets actually were. More importantly, the quote from Anne-Ditte Termannsen in the release, the final paragraph of the release, and the ‘conclusion’ section on the poster, both indicate that these advantages of being on the vegan diets could, in part at least, be because they differed from the control diets in calorie intake, in fat content, and in the amount of dietary fibre. There were also differences between diets, including between the vegan diets in different studies, in carbohydrate, protein and fat content. That all makes it difficult to disentangle what’s going on. Would the advantages of vegan diets in terms of body weight and BMI still be there if they were compared with a non-vegan diet that matched in terms of these broad classes of nutrients? As the researchers point out, they can’t tell from this study.
“Another inevitable issue is that, on randomised trials like those reviewed in this research, where the diets went on for at least 12 weeks, the participants would generally have known which diet they were on. So expectations about which diet might work, as well (in some cases) as their preferences for different foods, might have been reasons for some of the differences in body weight and BMI reduction, rather than any specific properties of the foods involved.
“In statistical terms, as far as I can tell from the available materials, the review and meta-analysis were carried out in a satisfactory and conventional way. (Meta-analysis is the statistical process of finding an appropriate statistical summary of the results of the previous trials.) But the trouble with assessing any meta-analysis is that the devil is always in the detail. You have to check whether the studies that are being meta-analysed are similar enough, that they are of good enough quality (because putting together the results of a set of bad studies isn’t going to give helpful overall measures), and look at the detail of who was studies in each of the trials and of what the treatments (the different diets) actually were. And we can’t do that because we don’t have the full research report yet.
“The poster does give a few more details of the findings on two outcomes – the blood sugar test (HbA1c) and body weight. On the blood sugar test, the average difference between the effects of the diets in trials involving people with type 2 diabetes was larger than in the trials involving overweight people. That’s hardly surprising since type 2 diabetes is a disorder of the metabolism of blood sugar.
“More interesting is the detail on body weight loss. In trials where the control group were not on any special diet but just continued on what they normally ate, on average those in the vegan diets lost around 7kg more than those who just continued their normal eating. In trials where the control group were also on a diet different from what they usually ate, but not a vegan one, those on the vegan diet lost about 3kg more, on average, than those on the control diets, so less than half as much as in trials involving people’s normal diets.
“It’s perhaps not really surprising that the difference was bigger when the control diet was what people normally ate, than when it involved some other special diet. But what’s also interesting is that the statistical measures of heterogeneity in this table on the poster are rather large, for both types of control diet. That’s telling us that there is evidence that the reductions in body weight in different studies where the control diet was normal eating, were different from one another to a statistically important degree, and that the same is true of the different studies where the control diet was something other than normal eating. In these circumstances, it’s less obvious that the average weight loss across each of the types of control diet has a very clear meaning. It seems to depend on some details of the individual studies. So even here, the overall conclusion is far from clear, I’d say. The vegan diets do, overall, seem to lead to higher weight loss on average, but how much higher depends on the details of the individual randomised trial. That’s not a good basis for recommending just any vegan diet. (Note that the column headed ‘Weight’ in both of these detailed tables on the poster is referring to a statistical concept of a weight in any meta-analysis, that has nothing direct to do with body weight.)”
Dr Frankie Phillips, Registered Dietitian and British Dietetic Association spokesperson, said:
“This study provides a useful insight into the potential role that a vegan diet may have on obesity and diabetes. Whilst it’s clear that some of the studies used in the meta analysis were small, there seems to be some clinically relevant improvements with regards weight loss and blood sugar.
“The detail needs to be presented in a full paper, as it is not clear whether the vegan diets were tightly prescribed, or self-prescribed, which could have a major impact on compliance and nutritional balance. The study does not give detail about compliance to a vegan diet by the participants, and this may be a factor in long-term compliance for clinical gain. It would also be useful to establish whether lifestyle might have played a role in the outcomes for this study, again this was not reported.
“Vegan diets are generally lower in energy content, fat and higher in dietary fibre and long-term observational studies have shown that vegans tend to be leaner than meat-eater controls. Since overweight is one risk factor for type 2 diabetes it is likely that vegans would be expected to have lower risk of type-2 diabetes. However, switching to a vegan diet for a fixed period of time is not the same as following a long-term vegan diet.
“Given the enormous cost of obesity and diabetes then this study suggests that following a vegan diet might play a useful role in supporting overweight adults and those with Type 2 diabetes to reduce body weight and possibly to help improve control of blood glucose, but support would be needed to ensure that nutrition is not compromised, particularly with respect to vitamin B12, vitamin D and iodine.”
Dr Ian Johnson, Nutrition researcher and Emeritus Fellow, Quadram Institute Bioscience, said:
“Many studies have shown that vegetarians tend to have a lower BMI and a reduced risk of metabolic diseases, compared to control subjects consuming omnivorous diets, and that the differences are greatest for vegetarians who minimise their intake of animal products. This systematic review of randomised controlled trials in which overweight patients or those with diabetic symptoms were given vegan diets and compared with controls consuming a range of different non-vegan diets, is generally consistent with these previous findings.
“It is difficult to draw any conclusions as to the mechanisms of action at work in these trials because no details of the diets are given, and vegan diets are complex, and varied in composition. However, as the authors suggest, the most likely reason for the observed weight losses and improved metabolic parameters is a prolonged reduction in energy intake associated with the vegan diets. Diets that are free of animal products but rich in fruits, vegetables and whole grains tend to have a relatively low energy density, and a much higher fibre content than conventional “western” diets.
“This review suggests that even over a relatively short period of time, vegan diets can offer important health benefits, though it is important to remember that they can also be low in certain micronutrients, such as iron and vitamin B12.”
Prof Gunter Kuhnle, Professor of Nutrition and Food Science, University of Reading, said:
“Diets tend to reduce weight, at least in the short term. The difficulty is often maintaining the weight loss and compliance with the dietary regime. In this study, the authors investigated the weight loss following a vegan diet, and unsurprisingly found that they can be successful.
“The authors conducted a systematic review of intervention studies and found that compared with no dietary interventions, vegan diets showed the strongest association with body weight reduction. When comparing vegan diets with other dietary interventions – such as the Mediterranean diet – the association was much weaker.
“It is very difficult to attribute the observed effects to a plant based diet as the diets differed in their composition: vegan diets generally have less fat and more fibre than other diets and are often less energy dense. However, it is possible to achieve the same with a diet that is not exclusively plant based.
“While this study provides very useful information for research, it does not suggest that adopting a vegan diet will automatically result in weight loss.”
Prof Janice Thompson, Emeritus Professor of Public Health Nutrition & Exercise, University of Birmingham, said:
“This preliminary study employed a systematic review and meta-analysis to explore the impact of vegan diets on weight loss and various clinical metabolic parameters in adults who were overweight or had type 2 diabetes. One issue is combining studies and samples, in that not all participants were overweight and living with type 2 diabetes. The limitations identified are valid, in that studies did not compare diets with similar energy (calories), fat, carbohydrate and protein content. This limitation can be useful to guiding the design of future studies to tease out if, and how, vegan diets may impact on weight loss more effectively than other healthy eating diets and eating plans.”
Prof Ian Givens, Professor of Food Chain Nutrition, Director, Institute of Food, Nutrition and Health, University of Reading, said:
“The benefits of weight loss to diabetics/pre-diabetics are well known and so this study is clearly of interest. However a key question is whether the findings of this study were due to the vegan diets or a reflection of a lower energy intake, in other words were the control diets isoenergetic (providing the same number of calories) with the vegan diets. This seems not to have been the case as the authors note in their poster conclusion ‘Some of this effect may be contributed to differences in the macronutrient composition and energy intake in the vegan diets versus control diets’. Clearly any differential effect of diet composition (vegan) and energy intake needs to be resolved.”
Dr Duane Mellor, Registered Dietitian and Senior Teaching Fellow, Aston Medical School, Aston University, said:
“Changing your diet from a typical diet to one which excludes a large number of foods (such as a vegan diet) is likely to lead to weight loss, and in people living with type 2 diabetes this is likely to improve their blood glucose levels and blood pressure. This, if well planned can be healthy, but cutting out a whole range of foods without thinking how nutrients are replaced, in the case of a vegan diet that includes nutrients such as vitamin D, vitamin B12, iron and iodine, can lead in the long term to worse health.
“This review only considers studies for up to 12 weeks and does not really consider dietary adequacy. The studies it used would have advised participants to eat a varied and balanced diet – something if people are thinking of trying a change to a vegan diet also need to think about and get advice from a health professional, and if they are living with a health condition such as type 2 diabetes they should ideally speak to a dietitian.
“It is also important to remember that not all vegan foods are healthy, after all sugar is vegan, and as more people choose a plant based diet, there are increasingly more foods which are highly processed which are being produced to see this demand. A vegan diet can be healthy, but it needs to be varied so that adequate intakes of all nutrients are achieved.
“It is also important to remember this is an abstract presented at a scientific meeting and has not been peer reviewed or published as a full and detailed paper, so not all the information on quality and risk of bias of the study are available to be able to fully critically appraise it.”
Abstract title: ‘Effects of vegan diets on cardiometabolic health: A systematic review and meta-analysis of randomized controlled trials’ by A. Termannsen et al. was presented at the European Congress on Obesity.
This work is not peer-reviewed and there is no paper.
Declared interests
Prof Kevin McConway: “I am a Trustee of the SMC and a member of its Advisory Committee. My quote above is in my capacity as an independent professional statistician.”
Dr Frankie Phillips: “No conflicting interests.”
Dr Ian Johnson: “No conflicts.”
Prof Gunter Kuhnle: “No conflicts to declare.”
Prof Janice Thompson: “No conflicts of interest.”
Prof Ian Givens: “I have no interests to declare in relation to this study.”
Dr Duane Mellor: “Through personal choice I am a vegetarian and have worked with the Vegan Society to produce materials in the past as a member of the British Dietetic Association.”