select search filters
briefings
roundups & rapid reactions
before the headlines
Fiona fox's blog
A study published in Nature Metabolism looks at food-seeking behaviour and skin ultraviolet exposure.
Dr Duane Mellor, Registered Dietitian and Senior Teaching Fellow, Aston Medical School, Aston University, said:
“This is an interesting series of five experiments exploring the effect of sunlight on eating behaviour. One of these was in mice, two looked at cells in a test tube, along with a small study in humans looking at appetite in people with skin disorders who were receiving UVB light therapy, and an observational study looking at how much food 3000 people in Israel ate at different times in the year. The data from the mouse study suggested in an experimental mouse model UVB light may be linked to increased food intake and weight gain. The data from the human studies did not look at weight gain, instead the data suggested that having UVB light therapy was associated with males with skin disorders reporting feeling more hungry, but females having the same treatment did not report a change to their appetite. The population study looked at what 3000 people ate on a single day, as this data was collected over the period of two years, it meant they could report average energy intake at different times in the year. Therefore it is not possible to say if the same male will eat more food (energy or calories) in summer than they would in winter as this data is only cross-sectional study. Additionally there is no data on sun exposure and therefore exposure to UVB in this study.
“So, this study gives five pieces of a very interesting scientific jigsaw, but unfortunately due to gaps in the data, these pieces do not quite fit together.
“What it does show is the potential mechanism of how UVB can influence hormone metabolism (which is similar to how it acts to help vitamin D production in our skin), and how this may be associated with an increase in males the appetite hormone ghrelin, at least in mice. It is important to recognise this paper does not claim that sunlight and UVB exposure will lead to weight gain in human males. Instead it provides some interesting insights in how moderate UVB exposure could be linked to health benefits including reduced cardiovascular risk and inflammation (as ghrelin has anti-inflammatory effects).
“What it is also important to remember is that when most people have increased UVB exposure from sunlight there may also be being more physically active, this could partially explain greater food intake although there is no data to support this from this current study.”
Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:
“This is a solid, informative piece of research, but it does need care in interpretation so as to be clear on what it’s actually informing us about.
“Much of this research report deals with investigations of the role of the hormone ghrelin in the skin, and how that relates to exposure to ultraviolet radiation (from the sun or otherwise). I did find that interesting and can imagine its importance, but the physiology and endocrinology is not my expertise, so I won’t comment further.
“Instead I’ll comment on just one broad aspect – the relevance of these findings to humans in real life outside the lab. Most of the detailed findings in the research come from studies of laboratory animals (mice) and in cultures of cells, including human cell cultures, in the lab. Of course, one can find out a very great deal about what might be going on inside human bodies from that kind of research, and I’m not at all criticising those aspects of the work, though it’s always necessary to check how findings from those sources might apply to humans. But the researchers also carried out studies in humans, and I think there are various problems of interpretation about those.
“Overall, though some (but not quite all) of the findings in humans in real life are consistent with the findings in lab mice and in cell cultures, I don’t think that they entirely support the possibility that the same is going on in these humans as in the lab results, for reasons that I’ll describe. This does not mean that the lab results have no value in describing what happens inside real human beings – I’m in no position to judge that. However, I do think it means that the prominence given to the findings from the national survey, that I’ll come to next, is misplaced in the context of the research as a whole.
“The first set of findings in humans that is discussed in both the research paper and the press release uses data from about 3,000 people, whose diets were recorded as part of a national survey carried out on random samples of people from the Israeli population, in 1999-2001. The researchers found that men, on average, increased their intake of calories from food in the summer months, when solar radiation is highest, but women did not on average show a similar summer peak of calorie intake. This aspect of the research is an observational study – there was no intervention to get people to change what they ate or how they behaved. The issue here is that nothing in this specific part of the work actually tells us that it is the solar radiation that caused the summer increase in food intake in men, or that there is some physiological reason that this did not happen in women.
“The results in mice and cell cultures and the other human studies do at least make it plausible that solar radiation and sex differences could be involved in causing the patterns, but they can’t rule out other possible explanations. It’s also plausible that men and women behave differently in relation to sunny weather, for social reasons, because of social norms and customs about how much skin men and women might expose to the sun, because of different jobs that they might do, or for a long list of other reasons, and that that might all have something to do with differences in calorie consumption. The patterns of calorie consumption in different seasons, and the difference between men and women, might be explained in part, or even completely, by differences in other factors like this. In many observational studies, the researchers carry out statistical adjustments to try to allow for other factors, but this appears not to have been done at all in this study. And anyway one has to ask whether similar patterns in calorie consumption would show up in a country like the UK, where (despite the current hot sunny weather over much of the country) we don’t usually have such warm sunny summers as in Israel. Even if the dietary patterns in Israel are indeed largely caused by solar radiation and differences in male and female hormones, would the same patterns show up here?
“What about the other studies in living humans reported here? There were two sets.
“A small number of people (13 men and 14 women) were asked to spend about 25 minutes in the sun around midday (on just one occasion). They were asked about their appetites, and blood samples were taken and investigated. The results are compatible with the findings from the mice and the cell cultures, that exposure to the sun changes level of the hormone ghrelin and changes appetite in men, but not in women. It’s a possibility that that is the explanation for the results, but it isn’t the only possibility, and again I don’t think it’s clear exactly how this relates to what would happen in the everyday human world. The paper also reports that 5 men and 5 women, after this sunny lunchtime, had blood samples examined for levels of blood serum proteins, and again the results are broadly compatible with the other findings. (It’s not made very clear whether these 10 people were or were not included in the sample of 27 that I’ve already mentioned.) We’re told that the study participants were volunteers from researchers in the university medical faculty where the lead researchers on this study were based, and that they were ‘recruited by convenience sampling’, which generally means that they are people that the researchers happened to ask to take part. So we don’t know from this report whether the subjects for this study knew about the research hypotheses of the overall research, and whether they might have behaved differently in the sun, or answered the questions, differently, on the basis of any expectations about what the results might be. (We don’t even know from this report whether the subjects for this part of the study were amongst the researchers for this piece of research.) We again don’t know whether there were socially-determined differences between the men and the women in how they exposed themselves to the sun (except that they were asked to wear shorts and sleeveless clothes). So again I think it’s not really possibly to say how much this part of the study tells us about the impact of the research findings on humans going about their normal activities. If this part of the study had found no difference between the responses of men and women, that would clearly have thrown doubt on the other findings, and that did not happen. But I still don’t think these particular human findings go very far to confirm the other findings.
“The final study in humans involved 32 people (14 men and 18 women) undergoing phototherapy treatment (treatment by exposure to light in this case specifically UVB) for various skin conditions. Again the participants were recruited by convenience sampling, so it’s not even clear that they are representative of people undergoing phototherapy for skin conditions, let alone of people in general. They were asked standard questions about appetite, before and one month after their UVB treatments began. The researchers report that there were significant differences between the two sets of responses on appetite for men, though not for women. That’s true – the results are in Table 1 in the paper though the headings do not make it clear that the first set of columns is for males and the second for females. But on the whole the differences are very small. The researchers report that males reported more variation in their daily appetite after the phototherapy than before, that they felt hungry more frequently, and that they ate less frequently, and that none of these differences were seen in women. It’s true that, for men, the average differences on these three measures are large enough so that they are probably not due to chance variability. (Though, to nitpick, that’s not quite true for the variation in daily appetite, using the conventional boundary for deciding what might plausibly be due to chance alone. Also the researchers in these calculations only took into account the possibility of a difference between the responses in one direction – in the jargon they used one-tailed tests – which isn’t usual in this context.). And for women, though there are differences on these three measures after the phototherapy than before, and though they (on average) go in the same direction as for the men on variation in appetite and on feeling hungry frequently, they are small enough so that they might possibly be due only to chance variation.
“However, these statistical results come from changes in very few participants. On the question about feeling hungry frequently and also on the question about eating frequently, 11 of the 14 men and 15 of the 18 women gave exactly the same response before and after the phototherapy. Rather fewer didn’t change their response on the question about variation in appetite (8 men and 11 women didn’t change). But, overall, I really don’t see these findings from the phototherapy patients as being convincing evidence of a real-world effect in humans. Also the finding on eating frequency was that the three men who gave a different response after the therapy said (on average) that they ate less frequently afterward than before (though they may have eaten more on each occasion), while the three women who changed their response after therapy said (on average) that they ate more frequently, though the result for women could easily be due just to chance. So this particular finding seems possibly to go in the opposite direction as you might expect from the mouse studies.”
‘Food-seeking behaviour is triggered by skin ultraviolet exposure in males’ by Shivang Parikh et al. was published in Nature Metabolism at 16:00 UK time on Monday 11 July 2022.
DOI: 10.1038/s42255-022-00587-9
Declared interests
Dr Duane Mellor: “I have no conflicts of interest in relation to this work.”
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.”