7 July 2025
Sleep is one of the Six Pillars of Lifestyle Medicine, and for good reason. Both inadequate quantity and quality of sleep have deleterious effects on every organ and system in the body. From impeding the immune system’s ability to defend against infections and cancer, to diminishing glucose tolerance and insulin sensitivity, to ramping up appetite (especially for ‘comfort foods’ high in fat and sugar), to impairing cognitive function, poor sleep contributes substantially to just about every complaint that plagues modern humans… and that I see in my practice, every day. There are also significant societal and economic effects of our collective sleep problem, including increased risk of motor vehicle and workplace accidents, worse academic outcomes, and decreased workplace performance and productivity. In short, poor sleep has significant ramifications, and figuring out ways to help people sleep better should be a major priority of health professionals and the research community.
Back in March 2019, I wrote about a study that explored the effect of meal patterns with varying quantities of fat (and its subtypes), sugar, starch, fibre and total energy (kilojoules/calories), on sleep quality in healthy adults. The TL;DR is that on days when participants ate more fibre, they went to sleep faster and slept more deeply, whereas on days when they ate more saturated fat and refined carbohydrates, they spent less time in deep, restorative sleep and woke more often during the night.
The lead author of this study, Marie-Pierre St-Onge, has published a slew of studies investigating the impact that factors such as intake of specific nutrients, meal composition, diet quality, meal timing and calorie distribution across the day, have on not only sleep quantity and quality, but on human metabolism, cognitive function and disease risk.
St-Onge’s most recent study examined the effects of particular food groups and components on the sleep fragmentation index – an easily- and non-invasively-measured marker of sleep quality – in young adults. Sleep fragmentation index (SFI) is the ratio of the number of awakenings per night to the total sleep time in minutes. A lower SFI is indicative of less disrupted sleep (better sleep quality), while a higher SFI indicates more disrupted sleep (worse sleep quality).
For this study, dietary intake was tracked using the Automated Self-Administered 24-Hour (ASA24) Dietary Assessment Tool from the National Cancer Institute, with participants completing multiple days of 24-hour dietary records (a far more accurate procedure for assessing food intake than the food frequency intake questionnaires typically used in cohort studies). Wrist actigraphy – performed by fitting participants with a triaxial accelerometer worn on the nondominant wrist – was used to objectively monitor sleep; sleep diaries were used to validate these findings where required1.
After all the data were crunched, St-Onge and her team found that the dietary factor that best predicted the ability to sleep through the night with minimal disruption was… drumroll please… eating more fruits and vegetables.
OK, it’s not a particularly sexy finding, I’ll grant you that. I’m sure the study would have gotten more traction if it had revealed that the key to sleeping through the night was sipping a tea brewed from the flowers of a tree found only in the heart of the Amazon which blooms just once every five years, and that have to be harvested by a virgin under the light of the full moon in order to preserve their soporific potency. I can see the infomercial now. I’ll bet that slimy used car salesman world-renowned cardiologist, Steven Gundry, would be right into it. It would complement his exotic Tunisian olive oil schtick beautifully.
But back in the real world, given a) the high prevalence of poor sleep quality and b) the extraordinarily low consumption of fresh fruits and vegetables in the general population2, upping the fresh produce intake sure seems like a low-risk and potentially very high-yield intervention for health professionals to recommend to their underslept patients.
The other major findings of the study were that higher total carbohydrate intake (but not added sugar) also promoted less fragmented sleep and conversely, there were trends toward associations between higher intake of fibre and magnesium with less disrupted sleep, and higher intake of red and processed meat with more disrupted sleep, none of which reached statistical significance.
It’s sad that I have to point this out, but people don’t eat ‘carbohydrate’ or ‘magnesium’ or ‘fibre’; they eat food. Food is a complex nutritional matrix that exemplifies the maxim that the whole is more than the sum of its parts. For example, all plant foods contain carbohydrates that humans cannot digest, and which we collectively term ‘fibre’, but each plant food contains widely varying types of fibre, that in turn form the food supply for many different types of bacteria that comprise our gut microbiota. And each different plant food also contains its own unique mix of micronutrients (vitamins and minerals) and phytonutrients (such as carotenoids, polyphenols and isoflavones), that influence our biochemistry and metabolism in unique ways.
In other words, each individual natural food is a complex system. Dietary patterns are also complex systems, and so are human beings. Layer these three levels of complex systems over each other, and you have a degree of complexity that is unfathomable to the human mind! However, the most salient characteristic of a good scientist is the relentless drive to fathom the unfathomable… or die trying.
And so, St-Onge and her team put forward a number of hypotheses, all of them supported by evidence, to explain the primary finding, namely that higher fruit and vegetable intake during the day predicted less fragmented sleep that night:
- Higher dietary intake of carbohydrates – the primary macronutrient in most fruits and many vegetables – facilitates increased uptake of the amino acid tryptophan by the brain. The brain uses tryptophan to synthesise serotonin, which is subsequently converted to the sleep-promoting hormone, melatonin.
- Dietary carbohydrate intake leads to increased stimulation of glucose-sensing neurons in the hypothalamus, which in turn reduce the activity of neurons that release orexins (neurotransmitters that block REM sleep and keep you awake) and increase the activity of melatonin-releasing neurons.
- Polyphenols from fruits and vegetables influence gut microbiome-brain communications in ways that could be beneficial to sleep quality; observational studies have reported that higher polyphenol intakes are linked to better overall sleep quality.
- Many fruits (including tomatoes) contain melatonin, and consumption of them has been shown to raise melatonin levels; tart cherry juice in particular has been shown to increase circulating melatonin and improve sleep quality, especially in older adults.
There’s another potential explanation that the authors of the study don’t mention: food is a zero sum game. Every piece of fruit that you eat crowds out some other type of food that you could have been eating instead. If fruit replaces, for example, ice cream or cheesecake, you get a sleep-enhancing double-whammy: no sleep-disrupting sugar or saturated fat, along with a decent helping of sleep-enhancing polyphenols, melatonin, fibre and unrefined carbohydrate.
How much fruit and veg do you need to eat, to get better sleep?
The researchers calculated that adding five cups of fruits and vegetables per day to the diet of someone who ate no fresh produce, would be expected to reduce the sleep fragmentation index – remember, that’s the number of awakenings per night divided into the total sleep time in minutes – by 16 per cent. Would you notice a 16 per cent improvement in your SFI? That probably depends on how fragmented your sleep was in the first place; the higher the SFI (i.e. the more fragmented one’s sleep), the stronger the correlation with self-reported insomnia symptoms and daytime sleepiness.
It’s also important to point out that the average age of participants in this study was 28. Sleep fragmentation index is generally quite low in this age group, which makes it harder to identify any factors that increase it. I suspect that the effect would be more pronounced and noticeable in older adults, who experience progressively higher levels of sleep fragmentation as they age up. But even if you aren’t consciously aware of your night-time awakenings, the more you have of them and the longer they last, the more impaired your cognitive and emotional processing will be, the higher your pain sensitivity, and the more fatigued you’ll feel.
Now, five cups of fruit and veg might sound like a lot if you’re eating the typical Western diet that’s heavily slanted toward ultraprocessed foods and animal products… and even more so if you’ve jumped aboard the carnivore clown car. But personally, I eat more than that by lunchtime! In fact, in the interests of science, I measured my fruit and vegetable intake on the day I wrote this article, and it totalled just over nine and a half cups. Mmmm, I’m looking forward to a good night’s sleep tonight!
On a related note, many years ago I wrote about a study that drew on three large, representative, cross-sectional studies of random samples of adults in England, Scotland, and Wales – a total of over 80 000 UK citizens – which found that “happiness and mental health rise in an approximately dose-response way with the number of daily portions of fruit and vegetables.” The tipping point was seven to eight portions per day; once participants met or exceeded this level of intake, their scores on questionnaires assessing happiness, life satisfaction and mental wellbeing became noticeably elevated. A ‘portion’ equates to half a cup of either fruit or non-leafy vegetables, or one cup of leafy vegetables… so seven to eight portions per day is pretty darn close to that five cups of fruit and vegetables that St-Onge and her colleagues used as a reference point.
Are there any lifestyle interventions that reduce sleep fragmentation index?
Absolutely! Exercise – both aerobic and resistance training – is highly effective, and so is eliminating light exposure during sleep. Speaking of light, getting more blue-enriched light exposure during the morning (best achieved by going outside) decreases sleep fragmentation too.
On the other hand, alcohol consumption increases sleep fragmentation, and so does caffeine – although the effects are dose-dependent; a single cup of coffee consumed soon after waking has little to no impact on sleep, but the larger the amount you consume and the closer to bedtime you have it, the greater the impact on sleep quality.
One last thing. I expressed a generally negative assessment of the value of using ‘wearables’ to monitor and improve health status in Decoding the science of continuous glucose monitoring. But St-Onge’s study used wearable tech – wrist actigraphy – to assess sleep quality in this study. Might it be helpful to use wrist actigraphy (either via a doctor-prescribed device or a wearable with actigraphic capacity, such as a Fitbit) to assess the impact of dietary changes on your sleep quality?
A reader left the following comment on the aforementioned post, Decoding the science of continuous glucose monitoring:
I can see a potential value in short-term use of a wearable that incorporates sleep actigraphy, as a motivator to make changes in one’s daily habits. But not everyone is going to have as favourable a response as Phil, as I emphasised in my reply to his comment:
Personally, I’m going to skip the wrist actigraphy and just keep on eating my fruit and veg.
Are you confused by the scientific claims and counter-claims that you encounter through popular and social media? Would you like to learn how to read scientific research, assess its biases, and understand how it fits within the body of scientific literature? My EmpowerEd membership program is custom-made for you. Activate your free 1-month trial today!
- Wrist actigraphy is movement-based, so if you are lying still, but awake, in bed, the device might misinterpret that state as sleep. A sleep diary can help to clarify this confusion. ↩︎
- As I wrote in 99% of Aussie kids don’t eat enough vegetables – and their parents aren’t much better, among Australian men aged 19–50, 74.1% don’t meet the Australian Dietary Guidelines (ADG) recommendation for fruit which is a mere two servings, or 300 g of fruit per day, while 98.3% fall short of the vegetable intake recommendation, which is up to six servings per day, or 450 g. Among Australian women aged 19–50, 80% aren’t eating sufficient fruit and 94% don’t make the cut for vegetables (which is even lower than the target for males, at just 5 servings, or 375 g). ↩︎
