select search filters
briefings
roundups & rapid reactions
Fiona fox's blog
A study published in Human Reproduction looks at ultra-processed food (UPF) consumption and fertility and embryonic development.
Prof Jules Griffin, Director of the Rowett Institute, University of Aberdeen, said:
“The paper by Celine Lin and colleagues is a well conducted observational study of reproductive health in a Dutch population of new parents that identifies associations between ultra-processed food and male fertility as well as with certain aspects of the development of the foetus. This study adds to the growing body of evidence that diets high in ultra-processed food are associated with a range of poor health outcomes, but as with other observational studies it is difficult to be definitive about the causes. This means we must interpret the results with some caution. In this study, the group with the highest intake of ultra-processed food is also the group with the highest average body mass index, higher energy intake, high carbohydrate intake and lower markers of income – this makes it more complicated to understand whether it is ultra-processed food driving the changes they observe or other markers of lifestyle. The authors of the study use UPF classifications to describe dietary intake but we don’t know what components of dietary intake are responsible or indeed whether there are other descriptions of our diet that might work better to explain the data – I would be very interested to know how much intake of free sugars, salt and saturated fat explain the patterns they see in the data.”
Prof Kevin McConway, Emeritus Professor of Applied Statistics, Open University, said:
“In their conclusions, the researchers who wrote this new paper don’t really go beyond saying that their findings indicate that the health and lifestyle of fathers-to-be might need to be considered as well as those of mothers-to-be in terms of the outcomes for pregnancy and for their babies. But I do seem to recall that that advice was already around in some form forty years ago when my partner and I were trying to get pregnant.
“They also say that more research needs to be carried out to see whether similar results can be found in other populations, and if they are replicated elsewhere, to try to work out what might be the bodily mechanisms behind them. The new paper does give some possible reasons for those mechanisms, though this particular study doesn’t directly throw light on them.
“In other words, I’d say that the new study doesn’t really go further than giving very limited evidence about a possible correlation between the future parents’ diet and outcomes for fertility and for the babies, and raising hypotheses that it can’t confirm on its own but might possibly worth researching further in the future.
“The research paper and the quote from Dr Gaillard in the press release make it clear, rightly, that the study is observational and so can only look for associations, that is, correlations. It can’t establish whether those associations are ones of cause and effect.
“So the correlation that has the firmest level of statistical evidence in their findings is that couples where the male partner consumes higher amounts of ultra-processed foods (UPFs) tend to be more likely to be ‘subfertile’, that is, to take a year or more to get pregnant after starting to try to conceive. But there’s no clear implication that it is the male partner’s diet that causes the increased difficulty for the female partner to get pregnant. That’s a possibility, but there are others. Couples who differed in UPF consumption also differed in terms of many other factors, and it could be one or more of these other factors that are actually causing differences in the risk of subfertility.
“The researchers were, naturally, aware of this, and made several statistical adjustments to make some allowance for the effects of other factors. But they report that they couldn’t make an adjustment for levels of physical activity, because they had no data on it [see Supplementary Figure S2], and there may well also have been other factors that also couldn’t be adjusted for. So we just can’t be sufficiently sure what causes what.
“An important problem with this research is that the numbers of couples involved in most of the comparisons that were made were relatively small, as these things go. Therefore the statistical margins of error associated with the estimates that they made of the ratios and differences in their results, shown by the so-called 95% confidence intervals associated with those estimates, are often relatively wide.
“In most cases, in the tables in the research paper, the ranges of plausible values (confidence intervals) include the value 1 (for ratios, in Table 2), or 0 (for differences, Tables 3 and 4), which indicate that it remains plausible that there is no difference in these measures between the outcomes for parents who consumed different amounts of UPFs. I’m not saying that those differences really must be zero, only that that possibility (alongside other possibilities) can’t be ruled out by the data from this study.
“Some of the ranges of plausible values don’t include 1 (for ratios) or 0 (for differences). But generally even those go pretty all the way up to 1 or to 0, and so provide only weak statistical evidence that there’s a correlation of the outcomes with UPF consumption.
“In those cases, if the researchers could have adjusted for some factor on which they actually didn’t have data, it’s very plausible that those ranges might have indicated that there could really be no correlation between UPF consumption and outcomes for fertility and for the growth of the unborn baby.
“But really the issue is that the study couldn’t include enough couples to give statistically clear findings about the size of most correlations it looked at, let alone about what might be causing what. There’s just too much statistical uncertainty.
“That’s why, in principle, it makes sense for the researchers to suggest that more research should be done to throw a clearer light on what might be happening. However, I also have doubts on what that further research ought to look like. For instance, the paper recommends that these further studies “should use a randomized controlled trial design”.
“It’s true that such an RCT design might throw more light on cause and effect – but what sort of RCT do they actually mean? It would be neither practical nor ethically sound, I think, to run a trial where couples were randomised to eat particular diets throughout the whole period when they were trying to get pregnant and in early pregnancy.
“It would, potentially, be feasible to run a trial where different couples were randomised to get different advice on what to eat, which they could choose to follow or ignore. But that would need a good deal of preliminary work where different approaches to persuasion were developed.
“Further, the authors of the paper are keen on the approaches, at the level of whole populations, to reducing UPF consumption, that were advocated in the recent Lancet series of papers on UPFs. But if those changes are going on, it would be even harder to come up with a set of different approaches that individual couples could be randomised to in an RCT.
“It’s all very well saying that it would be good to use an RCT, but without actually being more specific on what kind of RCT, just advocating that one should be used is rather empty in my view.
“One set of factors that did differ between people who consumed different UPF amounts were the consumptions of different broad food components (carbohydrates, protein, fat and fibre). [See Supplementary Table S2.] It’s not surprising that (on average) those who consumed higher quantities of UPFs also consumed more carbs and fat, and less fibre. This raises the question of the extent to which any correlations between UPF consumption and the outcomes were due to the way the food was processed, or to the amount of some dietary components that tend to be high in UPFs.
“In the paper, the researchers generally seem to be equating dietary quality with the proportion of UPFs consumed. There are other aspects too. Some other research on UPF consumption has looked at estimates of correlations with UPFs after adjusting for things like fat and fibre consumption (as well as without making such adjustments). That could potentially have thrown more light here too – though it’s possible that there just weren’t enough participants to give a clear result on this.”
Prof Channa Jayasena, Professor of Reproductive Endocrinology, Imperial College London, said:
“We know that over-nutrition and obesity are major causes of both male and female fertility. However, the role of UPF in fertility has not been studied previously. This study looked at hundreds of couples trying to get pregnant. The authors find that UPF consumption in women was associated with slightly smaller embryo development. They also found that men with UPF consumption had slightly lower fertility.
“These results are interesting, but there are several reasons to be caution interpreting them. First, we have no way of knowing whether it is UPF itself, or some other behaviour that is linked with the things they observed. Secondly, the differences observed are tiny, and hardly significant compared with measures such as weight loss. This means that even if UPF are causing reproductive problems, their impact on individuals appears very small indeed.
“We know from previous research that in general all couples should prioritise a healthy diet, exercise, and smoking cessation when trying to get pregnant. Whether avoiding UPF will offer additional benefits remains unresolved.”
Prof Gunter Kuhnle, Professor of Nutrition and Food Science, University of Reading, said:
“Making bold claims based on single observational studies is always worrying – especially when the results are likely to cause distress to some people. Fertility is an important but very sensitive topic – and should therefore be handled accordingly.
“The study has investigated the association between ultra-processed food intake and fertility. However, the methods used in the study to estimate ultra-processed food intake have several limitations: first of all, the questionnaire used does not appear to have been developed or validated for ultra-processed food. Second, the authors had to make assumptions based on average food consumption data. While this approach can be used to estimate population intake, it has severe limitations for studies such as this.
“For example, the authors assumed that everyone who consumed ‘grains for porridge’ had “79.7% breakfast product Brinta, 16.4% oatmeal, 2.3% Breakfast product Albona 7-grains-energy breakfast, and 1.6% rice flour Bambix” – when in reality, some people might have had no oatmeal at all while others consumed nothing but. In consequence, there are no reliable data on how much ultra-processed foods were actually consumed by individuals, making it impossible to estimate the impact on health.
“Given that the assessment of ultra-processed food intake has severe limitations, the conclusions of the study, and the recommendations, need to be interpreted carefully. The authors point out that their study is observational so cannot prove causality, but then they also suggest that “a diet low in UPFs would be best for both partners”, which could be interpreted as them suggesting they do assume causality. A concern is that such result my cause distress and self-blame in people who have unsuccessfully tried to conceive – and that it is used to give advice to people trying to conceive.”
‘Periconceptional ultra-processed food consumption in women and men, fertility, and early embryonic development’ by Celine H.X. Lin et al. was published in Human Reproduction at 00:05 UK time on Tuesday 24 March 2026.
DOI: 10.1093/humrep/deag023
Declared interests
Prof Jules Griffin: “I am a consultant for Sitryx, a company specialising in designing drugs to target immunometabolism. The company is looking at drugs for irritable bowel disease and psoriasis.
I have received funding from the European Union to investigate endocrine disrupting chemicals and hold a grant from UK Research and Innovation examining the health benefits of a fish diet.
I hold shares in GlaxoSmithKline and Haleon plc.”
Prof Kevin McConway: “I have no conflicts of interest to declare.”
Prof Channa Jayasena: “No conflicts.”
Prof Gunter Kuhnle: “• Current funding: BBSRC Transforming UK Food System (FoodSEqual);