This is the time of year when lots of people establish their goals for the year, and for many this will be achieving weight loss. One of the most (theoretically….) important questions in nutritional research is “what is the best diet for weight loss?”. A quick search on PubMed finds around 6100 clinical trials done on nutrition and weight loss. And what is clear from this data………… is that there is no single best diet or approach.
The differences between the diets or approaches tested have typically been so small as to have essentially very little meaning. For example, in a trial in people who weigh 100kg and one diet gets 9kg weight loss on average and the other 10kg, does it really matter?
Then you look at the proportion of people in that dietary trial who manage to stick to the dietary interventions longer than 3 or 6 months (eg, people might start super low carb, but by the end of the study they are typically back or close to their normal carb intake) – and it’s clear that people struggle to stick to all sorts of dietary changes. And if you can’t stick to it, then it’s not “the best” diet, is it?
I’ve written about some of the physiological adaptations to weight loss which make weight maintenance so hard in a previous post. However, I also think that the amount of behaviours you have to change (essentially permanently) in order to sustain weight loss is also a huge barrier. And I think this is why the best approach to weight loss always will be necessarily individual and personalized to an enormous extent.
Losing weight can require changing A LOT of [deeply engrained] behaviours [permanently]. (This is hard).
Consider if we were to run a trial on the effect of a medication on a bacterial infection. This would likely be a short-term intervention of a week or so and whatever agent is used to get rid of the infection can be withdrawn once the infection is cleared. And the intervention itself might require a single, simple(?) behaviour change: a person might need to swallow a pill with water twice a day for 5 days. Most people would be able to stick to that. So you’d get a definitive answer after a few trials (not 6000+).
Conversely, weight loss - and most importantly its maintenance - requires a person to change their behaviour long-term. And is it’s not just one behaviour. In the infection example above, a person needs to change their behaviour sufficiently to allow the entry of an antibacterial substance into the body – a pill, washed down with water. Done. For weight loss people need to adapt sufficient behaviours such that they 1) consume fewer calories than they did previously* and/or 2) do more physical activity** .
And just think about the number of behaviours which can alter calorie intake or physical activity…
Food/calorie intake can be influenced by the type of food eaten, method of cooking, sleeping habits, whether a person is socialising or not, who they’re socialising with, whether they’re watching TV, what they’re watching on TV, whether they’re out or at home. Wider factors can also influence food intake indirectly: a person’s environment: tempting sweets or cakes in their office, what options are available in the work canteen, stress, medications, boredom…..
Then you consider the number of things that can influence each of these and this could be the job that you do, your commute, your income, your relationships, where you live, what supermarkets and other food outlets surround you in your daily life and more.
Likewise, physical activity can be influenced by the job you do, your commute, your relationships, where you live, whether you've slept, whether you're stressed, whether you have other comorbidities that might influence physical activity, your income etc.
Because each of these direct and indirect factors vary so immensely between individuals, and because individuals themselves have different preferences of what's important to them, each individual’s approach to weight loss is going to be very different.
Individuals are REALLY different
You might have heard the axiom “You can't outrun a bad diet”. And the trials that we have that demonstrate physical activity compared to diet isn't particularly effective at weight loss. But does this mean that exercise is not useful for weight loss FOR ANYONE? No. Trials work tell you the “average” result across a sample of people. But, for [the admittedly small number of people..] who really enjoy physical activity, or potentially really get into an activity such as running, or a particular sport, they will likely find that they do indeed lose weight, and significantly, overtime and are able to keep it off.
Does it mean you would necessarily recommend it as a good approach? The answer is - from a clinical perspective - rarely. And this is largely because of the environments that most of us find ourselves in. Most people drive to work, sit all day at work, have long hours and other demands outside of work, such that participating in the large volume of physical activity that is required to lose weight just would not be feasible [or enjoyable!] for them.
Similarly, because people are so different in their tastes, values, and goals, some can and do stick to diets others might view as “too restrictive” in the long term. I hear this a lot about keto. Is it probably too restrictive (or perhaps unpalatable or unaffordable) for most people? Based on trial data and clinical experience, yes. But there are people who really enjoy this diet and have had great long-term success with it.
Based on the evidence we have, the most effective method [remember: on average] for significant, long-term weight loss is a very-low-calorie, meal replacement-type approaches. Perhaps this method well because it comprises short-term significant sacrifice which achieves 10-15kg weight loss in about 3-4 months. And 4 months is probably around the max that most people are able to maintain significant sacrifice. And then people can go back to their normal diets – though maintaining a caloric balance to prevent/limit weight regain. This enables quite a lot of flexibility and variability in the type of diet and lifestyle a person can choose long-term after achieving their weight loss goal.
BUT – when patients are offered this dietary approach in primary care, about 75% of people turn it down. (Because – repeat – people are different, have lots of different things going on in their lives, like different things and want different things).
So what can we do to help more people find a diet that works for them?
Self-experimentation and self-monitoring can be key for an individual in finding a way to eat and live that helps them lose weight. I always encourage patients to keep regular food, mood and activity diaries, and to monitor their weight and physical activity. The micro- and macro-level data can help them identify patterns of what works and doesn’t.
In addition, formalizing self-experimentation could be a good way of helping people in healthcare systems to find an effective weight loss method that works for them. At Oxford we are running the NewDAWN study which seeks to do just that. We might do a better job at supporting patients with obesity to achieve sustainable weight loss if we offer a range of weight loss programmes in a structured way, rather than just sending them to one programme and hoping it works out for them.
I will write more about this study in a future post but for now I hope this post clarifies the issue:
*It looks like dietary protein intake can increase energy expenditure but the effect size is very, very small.
** and, of course, if that exercise causes a significant increase in muscle mass this will increase overall energy expenditure.