select search filters
roundups & rapid reactions
before the headlines
Fiona fox's blog

expert reaction to new study into fruit and veg consumption and mortality

Research published in the Journal of Epidemiology and Community Health 
suggested eating at least seven daily portions of fruit and vegetables may confer the best chance of staving off life-threatening diseases. An accompanying editorial suggested the UK’s current recommendation of five daily portions may require review. A before the headlines analysis accompanied these roundup comments.


Prof Brian Ratcliffe, Professor of Nutrition, Robert Gordon University, Aberdeen, said:

“This paper adds some helpful insights to the possible benefits of consuming plenty of fruit and vegetables. The people eating the highest levels of fruit and vegetables in this study were also likely to display other healthy characteristics such as lower prevalence of overweight, higher activity levels and fewer smokers.  The authors have applied statistical corrections to account for this. It is interesting to note that their findings indicate the importance of both fruit and vegetables, especially the latter and this has some significance for public health advice since, generally, people find it easier to increase consumption of fruit rather than vegetables. It was a problem for the authors to consider tinned and frozen fruit products together since these are not the same nutritionally and they acknowledge some other short-comings such as lack of information about the participants’ energy intake. The call to increase the 5-a-day advice seems a little premature considering that most people do not achieve this target and the differences between 5-a-day and 7-a-day in this study are small.”


Dr Nita Forouhi, MRC Epidemiology Unit, University of Cambridge, said:

“Confirming that eating more fruit/vegetables has benefits to reduce mortality, the current research is credible and strong to support the 5-a-day message. This is great because the public can be re-assured that we have got the public health recommendations right, and there is neither cause for confusion or concern.

“However, it is too early to change the current 5-a-day message to 7 or more a day on the basis of this study; the results are based on a single assessment of food intake in the previous 24 hours, adjustment for total calorie intake was not possible, accounting for other dietary factors was also not possible, and there is little evidence of a linear dose response. It is important to remember that the current message is already to consume at least 5-a-day.

“Current efforts will therefore be better spent in getting the population intake to meet the guideline of eating at least 5-a-day, which offers a win-win for all. This means those who don’t yet meet the 5-a-day recommendation, can be encouraged with confidence now to aim for that goal. This applies to about 70% of the adult population and 90% of the younger age groups in the UK. As for those who already meet the 5-a-day goal, this study supports they can increase their intake. The other good news is that eating some fruit/veg daily, even if under the 5-a-day, still has health benefits compared with eating none or under 1 portion per day.

“As for the findings of possible higher risk of death in those consuming more frozen/tinned fruits, it is too early to draw any conclusions. It provides a signal for more research that measures diet in much more detail and is able to account for many relevant factors that might be systematically different about people who eat tinned fruits versus those who don’t. As for reducing added sugars in foods (including therefore sugar in tinned fruit with syrups or juice), there is increasing research evidence to support that, but this study does not directly address this issue.”

Additional information from Dr Forouhi:

Currently only fewer than about a third of the UK adults meet the 5-a-day recommendation, and in youngsters aged 11-18 it’s only about 10% meeting the 5-a-day (source NDNS – National Diet and Nutrition Survey).


Professor Susan Jebb, Nuffield Department of Primary Care Health Sciences, University of Oxford, said:

“This new study confirms previous observations that people who choose diets with plenty of vegetables have a reduced risk of cardiovascular disease and cancer. A key outstanding question is whether this is entirely attributable to these specific foods, or whether they are acting as a marker of a broader dietary pattern associated with improved health. This is difficult to study here because the dietary information collected in the Health Survey for England is limited. However, the broader dietary pattern may well explain the apparent differences between fruit and vegetables and between fresh and tinned fruit.  From this data we can only speculate on a host of possible explanations for the effects and there is no evidence here to conclude this is specifically related to sugar. Meanwhile these data should provide further encouragement to the public to include more vegetables in their diet. While there may well be benefits to individuals of consuming more than 5 a day, for the nation as a whole the key challenge is to encourage and enable those people currently consuming just 1 or 2 portions of fruit and vegetables to reach their 5 A Day.”


Prof Patrick Wolfe, Professor of Statistics, University College London, said:

“This is an important study that adds to our understanding of the association between heavy daily consumption of fruits and vegetables, and decreased mortality. 

“It describes the results of a study on fruit and vegetable consumption via a large sample comprising “a nationally representative population in England.”

“As the authors note, “This study has found a strong association, but not necessarily a causal relationship.” 

“The authors are clear in discussing the limitations of their study (primarily: (1) lack of frequent, regular direct measurement of fruits/vegetables consumed, and (2) the possibility for response bias in reporting amounts consumed), along with potential confounding factors, particularly in the finding that “consuming frozen/canned fruit was associated with an increased risk of mortality.” 

“The commentary on this study is, however, problematic.  It appears to assume a causal relationship, in spite of the authors’ specific statements to the contrary.”


Professor Tom Sanders, Head of Diabetes and Nutritional Sciences Division, School of Medicine, Kings College London, said:

“This study reports an association between eating lots of fruit and vegetables and being much less likely to die prematurely. However, it is already well known that people who report eating lots of fruit and vegetable are usually health conscious, educated and better off. We also know that health conscious people report eating more fruit and vegetables (often more than they really eat).

“Among the group reporting eating 7+ portions of fruit and vegetables only 10% reported being current smokers compared with 39% in the lowest category, they were also more highly educated. Smoking is such a lethal lifestyle habit that the results are badly confounded by this. It could be equally argued that going to university reduces your risk of death. You cannot extrapolate from this kind of information to make sensible pronouncements about what people should eat.

“However, what it does show is that following a healthy lifestyle and being female means you are less likely to die young; this is particularly relevant as young women are now those likely to smoke often because they think it will keep them thin. While advice to eat five portions of fruit and vegetable a day is well accepted, there is little other evidence to suggest that eating more is better. We were unable to show any effects on blood pressure or cardiovascular risk from eating larger amounts of fruit and vegetables in a controlled trial(1).”

(1) Berry SE, Mulla UZ, Chowienczyk PJ, Sanders TA (2010) Increased Potassium Intake from Fruit and Vegetables or Supplements Does Not Lower Blood Pressure or Improve Vascular Function in Uk Men and Women with Early Hypertension: A Randomised Controlled Trial. Br J Nutr 104, 1839-1847.


Prof Naveed Sattar, Professor of Metabolic Medicine, BHF Glasgow Cardiovascular Research Centre, University of Glasgow, said:

“One must always be careful with observational data since some unmeasured confounders may account for both greater fruit and vegetable intake and lower risk of death. Having said that, the general trends from this paper suggest that eating more fruit and vegetables, especially in their natural form, is likely beneficial for health and this finding concurs with emerging evidence elsewhere. By contrast, taking fruit in a processed form, whether as fruit juice or potentially as sweetened tinned fruit may not give the same benefits to health.   These latter findings, linked to recent viewpoint published in the Lancet Diabetes and Endocrinology from ourselves, suggest we need to urgently:

  1. Revisit whether fruit juice should be considered as one of the five a day
  2. Examine seriously the proposal to increase recommended intake to 7 a day. Such an implementation could encourage many to aim for greater fruit and vegetable intake and send out a clear message that many of us need to try harder to add fruit and vegetables to our diets. This message, in turn, could have secondary benefits, for example, by encouraging schools to increase their fruit and vegetable offerings to children.

“To implement a 7 a day message would be really challenging for many in society and would require governmental support such as subsidising the cost of fruit and vegetable (perhaps by taxing sugar-rich foods) and making available high quality products to all in society.”


Professor Tim Key, Oxford University’s Cancer Epidemiology Unit, said:

“These results are broadly in line with previous studies of this topic and it is helpful to have data from a fairly representative sample of people in England.

“Both the paper and the commentary refer to “effects” of fruit and vegetables, whereas the researchers have only observed associations in this study. We need better understanding of mechanisms before we can be confident about effects.

“Technically, there are a few things in the way the analysis has been carried out which I would have asked about if I had refereed the paper. For example, the adjustment for smoking is relatively crude (simply considering whether people were current smokers without using number of cigarettes smoked), and they did not exclude participants who were unwell at the start of the study. And there is a lot of potential for residual confounding – that is, not completely accounting for other factors that might influence the association seen – for example, there is about a fourfold variation in education and smoking between those who consumed low to high amounts of fruit and vegetables.”


Professor Richard Tiffin, Director of the Centre for Food Security at the University of Reading, said:

“Other countries already recommend more than five a day, but I don’t think we should change the message every time new evidence emerges. The underlying key message behind ‘five a day’ is that we all need to eat more fruit and vegetables. The evidence suggests that this is one of the more successful campaigns at improving behaviour around diet and health. Changing such a successful message in this way runs the risk of undermining its strength. If we keep changing such messages, people could be forgiven for thinking that scientists can’t make up their minds, or, worse for public health implications, that they don’t know what they’re talking about.”


Dr Rachel McCloy, a psychologist at the University of Reading, said:

“Rushing to change the guidelines on ‘five-a-day’ too quickly might have an adverse effect on trust in scientific advice. By increasing the recommendation to seven-a-day, or even 10-a-day, public health officials might succeed at pushing up average consumption, as motivated individuals seek to follow the most up-to-date health advice. But it could also demotivate those who already struggle to get to five-a-day, by making the target even further out of reach. These are the people who are likely to benefit most from an improvement in diet, and so cannot be forgotten in efforts to improve public health.

“Advice on healthy lifestyles might work on paper, but human behaviour is more complex than that. You can’t just pull a lever and expect to change people’s behaviour overnight. If we really want to encourage people to eat more fruit and vegetables, we need to find ways to make the healthy choices easier to make.”


Dr Gunter Kuhnle, a food scientist at the University of Reading, said:

“This is an excellent study. It is one of the first pieces of research to clearly show that eating fruit and vegetables lowers the risk of cancer among the British public. Previously the evidence for this was far more ambiguous – this is a game-changing advance in food science. It also reinforces the well-understood benefits of eating fruit and veg on lowering risks of cardiovascular disease. The fact that the research is based on a representative sample of the UK population makes this even more relevant to British consumers.

“The question about tinned or frozen fruit is due to problems with the method. The researchers were not able to distinguish between the two in their survey, making it impossible to make a distinction in later analysis. It’s possible that eating tinned fruits are an indicator of high sugar intake, but it might also be a marker of poverty or lower socio-economic class – as there is no data, it’s simply not possible to speculate.

“Currently, most British people don’t even eat five portions of fruit and vegetables a day – the average is two portions of fruit and 1.5 portions of vegetables – so any efforts that will encourage people to eat more are welcome. But we should be very careful in how these messages are communicated.”


Dr Janice E. Drew, Senior Research Fellow, University of Aberdeen, Rowett Institute of Nutrition and Health, said:

“Research supports that further benefits in health could be achieved by increasing fruit and veg consumption above the minimum guidelines set at 5 a day. Also the detrimental effects of high levels of sugar would be best avoided even if in a fruit based drink. Fruit juice is also lacking in the complex mixture of components in whole fruit that contribute to the health effects.

“Irrespective of diet, the genetic make-up of each individual does contribute to that individual’s risk of developing cardiovascular disease or cancer. Consequently the beneficial effects of consuming fruit and vegetables will vary from one individual to another. Other lifestyle factors can also impact on the requirement to replenish the protective components supplied by dietary fruit and vegetables. The quality of the fruit and vegetables consumed is also a factor and many nutrients will be depleted with storage. This does make it difficult to establish absolute amounts of fruit and vegetable consumption required to ensure health and prevention of cardiovascular disease or cancer in each individual. However, the evidence to date would support that an increase from minimum guidelines set at 5 a day would be beneficial to the majority of individuals.

“A strength of this current study is the recruitment of participants from The Health Survey for England generating a more representative cross section of a free-living population. Previous studies reporting health benefits of consuming fruit and vegetables have often used cohorts that are potentially more health conscious than the general population, being recruited from health service personnel or individuals attending clinics for specific tests relating to chronic diseases associated with diet and lifestyle. This has been reported to introduce bias in previous reports. However, it is also necessary to appreciate that the general population within England is just that and the conclusions drawn from this study cannot be directly representative of individuals from other countries.

“There are a number of other diet and lifestyle factors that are associated with fruit and vegetable consumption. The authors have attempted adjustments based on physical activity and BMI. However, further statistical analysis of interactions of diet and lifestyle factors contributing to the outcomes would be of interest. For example do individuals consuming greater amounts of frozen/canned fruit eat less fresh fruit and vegetables.

“The data regarding frozen/canned fruit is certainly intriguing and warrants further investigation. Further analysis is required though since this category was classified as “frozen/canned fruit”. These two different dietary components require to be examined separately.

“Overall I think the editorial provides a balanced view of the paper.”


Catherine Collins, Principal Dietitian, St George’s Hospital NHS Trust & spokesperson for The British Dietetic Association, said:

“As a dietitian it’s good to see justification of the robust recommendations on ‘5-a-day’ for fruit and veg intake being proven for our England population.

“The research also showed that whatever your body weight your health significantly benefits from eating more fruit and veggies.

“Eating 5 a day reduces your risk of dying from heart disease or cancer by almost 30% compared to those eating none. Eating more than five-a-day may be additionally beneficial for those concerned about heart disease.

“All fruits and vegetables count towards healthy eating, but it seems vegetables may be slightly more beneficial than fruits in terms of health effects. However, it’s worth remembering that those with the highest intakes (over 7 portions a day) on average ate twice as many fruit as vegetable portions (see below), so we mustn’t undermine fruit – however we eat it – as useful in our diets.

“I’d be worried if the public took a message that ‘tinned fruit increases risk of death’ from the editorial accompanying this study, which is based on flawed arguments and limited comparative data. Fruit, like vegetables, provide not only vitamins, minerals, and dietary fibres, but also phytochemicals, plant substances that help maintain good health. Bottom line – just eat more!”

Additional information from Catherine Collins:

Whilst dietitians and health researchers have robustly promoted the ‘five a day’ nutrition message for fruits and vegetables for decades, an increasing number of commentators have challenged the nearly 25yo WHO guidelines exhorting us to eat at least 400g of fruit and vegetables a day from which this slogan was derived.

This study of more than 65,000 middle-aged people in England evaluated the risk of death from heart disease, cancer, and ‘all-cause’ mortality compared to a 24h recall of the previous days fruit and vegetable intake.

As a dietitian it was heartening to see that the majority of those surveyed managed just under 4 portions a day, which their analysis showed to reduce risk of death from heart disease or cancer by around 30%, compared to those who ate no fruit or vegetables at all. The protective effect of fruits and veggies seemed slightly more pronounced for those managing 5-7 portions a day, and with a very slight additional benefit in reducing deaths from heart disease in those managing more than 7 portions a day (but for cancer mortality this effect plateaud at 5-7 portions a day).

Vegetables seemed more protective than fruit in preserving health, which is consistent with other dietary surveys, so the goal should be at least 3 of your five-a-day should be raw or cooked vegetables, or salad. This is consistent with current advice.

With 59% of the population group classified as overweight or obese (based on BMI values) compared with 25% deemed ideal body weight, a higher mortality may have been expected in this ‘larger’ group (both in terms of number in sample, and their body weight). However, mortality in both groups was reduced in those who reported eating more fruits and vegetables and also in those who reported being more physically active in both groups. The small group of healthy weight who ate more than 7 portions of fruit and veg daily had a mortality almost half of those who ate none (HR 0.52), whereas for the overweight or obese mortality reduced almost 40% from baseline (HR 0.63) if 5 or more portions of fruit and veg were eaten daily (Table 2).

Whatever your body size, adopting more healthful diet and lifestyle messages will reduce your risk of death from the two most common causes of death in the UK.

One significant criticism, which is likely to form the bulk of some journalistic copy, was the finding that a mere single portion of frozen or canned fruit increased risk of mortality by 17%. The authors suggest that this could be due to the ‘high sugar intake’ of tinned fruit, derived from the added sugars present in syrup or the natural sugars in the fruit juice present with the tinned fruit.

However there are several issues in relation to tinned fruit:

(1) The sugar content of canned fruit is highly variable. Tinned apple, for example, has negligible liquid content and is almost 100% apple. Tinned fruits in syrup may have light syrup or heavy syrup, have a fruit juice/fructose syrup or pure fruit juice alone as part of the product, which usually accounts for 25-35% of weight of the product. People also vary as to whether they would usually consume the juice/syrup, consume part, or eat drained fruit only. This had not been evaluated.

(2) The definition of a portion within the question can only refer to the fruit itself, not to the syrup – how do you get a ‘heaped or full tablespoon of syrup’? In which case, why include the syrup issue in the hypothesis about sugar consumption, given that, say, tinned pineapple has the same carb content as fresh?

(3) Frozen fruits are a premium priced product. They are usually berries, melon, fruit salad, mango etc – none of which have added sugar. In terms of sugar content they should have been qualified along with fresh fruits. That they weren’t is a major nutritional oversight and one which will have contaminated any effect of tinned fruits on overall health.

(4) Tinned fruits are cheaper than the fresh fruit equivalent, and they also ‘keep’ for longer periods. It is highly likely (but not confirmed) that these would be a preferential fruit source for those on lower incomes. Income is a proportional indicator of cardiovascular and cancer mortality risk. The authors do state in their conclusion “Fruit and vegetable consumption is related to household income”.

The authors give a thoughtful qualification of what can be interpreted from their study, giving the limitations such as 15% of the sample not having their BMI data collected, that a single 24h recall may not be representative of habitual diet over seasons and years, and that they found fruit and vegetable intake inversely proportional to household income, all of which are valid influences on dietary quality and healthy lifestyle determinants.

Problems with the study as it discussed in the editorial:

This thoughtful survey is let down by the editorial, which focuses on one small and likely unreliable finding, and points the finger at sugar in relation to the effects of frozen or tinned fruit increasing mortality risk by 17%.

The differences in sugar content between frozen and different forms of canned fruit make such claims inaccurate. In addition, the way in which data was collected for this study means that this result is likely unreliable. As a dietitian I encourage all to eat more fruit and vegetables, including canned versions which may be cheaper, easier to store without waste, and that contribute to your five-a-day total.

In addition, for the ‘healthiest’ people eating 7+ portions of fruit and veg a day (mean around 10 portions), this was constructed of a mean 6.1 portions of fruit, and 3.1 portions of veg, ie fruit intake double that of veg. This does not correlate with the editorial berating the sugar content of dried fruit, smoothies, fruit juice and frozen/canned fruit as being a health issue.


‘Fruit and vegetable consumption and all-cause, cancer and CVD mortality: analysis of Health Survey for England data’ by Oyebode et al. published in Journal of Epidemiology and Community Health on Monday 31 March 2014.


in this section

filter RoundUps by year

search by tag