Reduce your calories. Eat more vegetables. Limit soft drinks and junk foods. For years, even decades, this has been the advice for those wanting a healthy body weight, lower blood pressure and better markers of metabolic health. Most weight-loss advice has focused on either what to eat (and what to avoid), or how much to eat. Think of dietary pyramids produced by government agencies, calories on food packaging and meals, and typical nutritional advice.

It’s all true, to a certain extent: it’s obviously better to eat a healthier, nutritionally balanced diet, and yes, lower body weight is broadly linked to reducing calories. But this type of approach can be hard to sustain. Even as a personal trainer who knows what I “should” be eating according to government dietary advice and has heard too much about calorie deficits, I take a slightly different approach to food. I think we need to bring nuance and a balanced approach to food and what we eat.

Sometimes I just want a piece of chocolate cake, although I know a fruit salad would be a nutritionally better choice. Also, counting calories and classifying foods as “good” or “bad” can not only be boring and time-consuming, it can also tip into disordered eating such as orthorexia, described by the British Dietetic Association as a “pathological fixation on healthy eating”. Plus, feeling hungry is hard to sustain over months and years, compared with a limited number of days.

So I was thrilled to read a new study that should have made headlines around the world. This meta-analysis investigates another dietary question: does when you eat make a difference? Researchers analysed 41 randomised controlled trial studies to look at the impact of early or mid-time restricted eating on a number of body weight and blood markers. Around 2,200 participants (42% women) with a mean age range of 19-69 years old were tracked for 4-48 weeks, depending on the study. Early-time restricted eating meant the last meal was eaten before 5pm, while mid-time was a meal ending between 5-7pm and late-time was ending a meal after 7pm.

The findings complicate the “calories in-calories out” narrative. Eating earlier in the evening (finishing either before 5pm or before 7pm) was associated with significant improvements in body weight, BMI, body fat percentage, waist circumference, blood pressure and metabolic markers (molecules in your blood like glucose, insulin and triglycerides, which can give a picture of how your metabolism is working).

Could this be due to eating fewer calories by those who ate earlier in the evening? Not exactly – several of the trials showed that even without eating less overall, eating within a certain window of time during the day resulted in better markers of metabolic health such as levels of fasting blood glucose, fasting insulin and triglycerides. Basically, an eating pattern earlier before bedtime has independent positive effects.

The researchers offer their explanation. First, the body manages sugar better earlier in the day, so the same meal causes a higher blood sugar spike at night compared with the morning. Studies have shown that the body releases the most insulin – the hormone that helps move sugar from blood into cells – from 12-6pm, and the lowest during sleep. Quite simply, because of how our hormones are secreted, we have a better handle over food intake in the morning and afternoon, compared with the evening and night.

Depending on how you read the findings, it might make your life easier or harder. Harder practically and socially: who can actually eat dinner before 5pm, or even 7pm, given how modern life is structured and wanting to eat together with family or friends? Late shift work makes this pretty much impossible, as can work and social commitments. If you really want that piece of cake, or croissant, it’s better to have it earlier in the day. I’m not saying have chocolate for breakfast, but I am saying that if you’re going to eat chocolate tomorrow, maybe have it before 5pm when your body’s better primed to deal with it.

• Prof Devi Sridhar is chair of global public health at the University of Edinburgh, and the author of How Not to Die (Too Soon)