The relationship between early-life nutrition and long-term health outcomes has long fascinated researchers and the public alike, particularly given the alarming rise in childhood obesity rates worldwide. With nearly one in five children and adolescents considered overweight or obese (as of 2022, from the World Health Organization¹), understanding the early-life factors that might influence metabolic health has become increasingly crucial.

One potential factor may be a sort of “programming” in early life that predisposes a person to having a sweet tooth, leading them to seek out sugary and calorically dense foods throughout life. This could contribute to obesity and the slew of health issues that come along with excess weight. A recent study has garnered significant attention for its suggestion that sugar restriction during pregnancy and early childhood might protect against diabetes and hypertension later in life.² But before we jump to conclusions about sugar being the dietary villain, let’s examine what this study actually tells us – and more importantly, what it doesn’t.

The sweet opportunity for a natural experiment

The study leverages a unique historical circumstance: the end of sugar rationing in post-World War II Britain. The abrupt change between wartime rationing and post-war abundance essentially created a natural experiment in which researchers could compare the long-term effects of different levels of sugar exposure in infancy. Indeed, with the end of the ration, sugar consumption nearly doubled from 41 g/day to approximately 80 g/day. Thus, using a subset of data from the UK Biobank (n=60,183), authors Gracner et al. compared health outcomes between individuals who experienced sugar rationing during gestation and early life with those who didn’t.

The findings appear striking at first glance. Those exposed to sugar rationing (i.e., less sugar) during gestation and up to the first year of life showed a 25% lower risk of developing type 2 diabetes (T2D) (HR: 0.75, 95% CI 0.68-0.84) and 15% lower risk of hypertension (HR: 0.85, 95% CI 0.81-0.89) compared to those who never experienced sugar rationing. The association strengthened with increased exposure to rationing: individuals who experienced rationing both in utero and for up to two years after birth showed even greater risk reductions relative to those who did not experience rationing (T2D HR: 0.64, 95% CI 0.60-0.69; hypertension HR: 0.81, 95% CI 0.77-0.84).

Interestingly, exposure to less sugar exclusively during gestation (children conceived during sugar rationing, but born after rationing ended) was also associated with decreased risk of disease later in life. More specifically, they showed a 13% lower risk of developing type 2 diabetes (HR: 0.87, 95% CI 0.81-0.94) and 8% lower risk of hypertension (HR: 0.92, 95% CI: 0.88-0.97) compared to those who were never exposed to rationing. This group also experienced a decreased risk of being diagnosed with obesity (HR 0.69, %95 CI 0.60–0.80 for those exposed to rationing in utero plus up to 1 year of life). From this, it was concluded that early exposure to sugar rationing protected from T2D and hypertension later in life.

To strengthen their conclusion, the authors examined other diseases, such as type I diabetes, to gain insight into whether this was due to differences in time trends generally (i.e., more accessible healthcare or improved diagnostics), and saw no effect. Thus, it would seem the culprit must be sugar (or so they say).

A spoonful of skepticism

While these findings might seem compelling, we need to approach them with appropriate skepticism. As I’ve discussed extensively (refer to my studying studies series), observational epidemiology can be useful, but we must be careful not to stretch the conclusions beyond the study’s limitations. I probably sound like a broken record here, but correlation does not imply causation!

Let’s start with what the data do show clearly – (1) there was a spike in average sugar consumption per person when it was de-rationed, and (2) children conceived and born before the de-rationing had a decreased lifetime risk of being diagnosed with obesity, T2D, or hypertension compared to those never exposed to rationing. This is a useful insight, but this is about all that we can take away with any level of confidence. To conclude from these data that increased exposure to sugar in early life per se causes children to develop T2D and hypertension (as the mainstream media has done) is to ignore the numerous limitations of observational data.

First and foremost, when sugar consumption increased, overall caloric intake rose as well – there wasn’t a trade-off in which sugar replaced other calorie sources. The authors explored whether similar effects occurred when butter was de-rationed, finding no differences between children born after sugar was de-rationed but before butter was de-rationed compared to those born after de-rationing of both. However, sugar consumption accounted for 77% of the increase in caloric intake, making it the primary driver of the excess calories (~150 total kcal/day total increase from rationed period). In other words, increased fat consumption did not provide a great enough change in total caloric intake for the natural experiment to exist in the way it did for sugar. Even the animal studies cited as “ample causal evidence” of sugar’s harm in early life primarily examine high-sugar, high-fat, high-calorie, “obesogenic” diets rather than sugar specifically, or they test non-nutritive sweeteners and not sugar at all. This leaves an important question unanswered: is the issue with sugar itself, or with excess calories overall?

An additional concern is that the conclusions are all based on averages across entire populations rather than data on each individual. We can assume that not everyone increased sugar intake after de-rationing to the same extent, but we do not know if the children who increased sugar intake the most were the same individuals that developed hypertension or diabetes later in life. We only know that people in general tended to eat more sugar (and more calories), and then 50-60 years later people in general tended to have higher rates of T2D and hypertension.

Furthermore, the study can’t establish whether early sugar exposure led to sustained higher sugar intake throughout life – a key assumption in the authors’ hypothesis. We likewise have no data on other crucial lifestyle factors (or lack-thereof) that influence diabetes and hypertension risk beyond sugar intake, such as physical activity levels or overall dietary patterns. In short, while the study highlights a correlation between early sugar intake and later disease risk, it fails to establish causality or account for other important lifestyle factors that could influence these outcomes.

Spun out of control

Despite me harping on the limitations of this study, this doesn’t mean the question isn’t worth investigating. In fact, numerous animal studies have examined how maternal and early-life diet composition affects later metabolic outcomes using rigorous experimental designs. However, the leap from these controlled animal studies to human population-level recommendations requires careful consideration and stronger evidence than what this current study provides.

The bigger issue I have in this case is with the sensationalization of limited data. This is a common problem in science communication: the tendency to oversimplify complex findings into attention-grabbing headlines with no consideration of the study limitations. Many outlets have interpreted these results as evidence that sugar exposure in early life programs individuals for chronic disease. This interpretation goes far beyond what the data actually show.

The bottom line

Rather than viewing this study as definitive evidence for restricting sugar during pregnancy and early childhood, we should see it as an interesting historical analysis that raises questions worthy of further investigation through more rigorous research designs. The present study’s limitations don’t diminish the importance of maintaining a healthy diet during pregnancy and early childhood. However, we should not be making specific recommendations about sugar restriction based on these findings alone.

While the relationship between early-life nutrition and long-term health outcomes remains an important area of research, this particular study doesn’t provide the smoking gun against sugar that many headlines suggest. Instead of getting caught up in fears about specific nutrients, the focus should remain on established principles of healthy eating and lifestyle habits during pregnancy and early childhood, such as keeping overall calorie intake within recommended ranges. The key message here isn’t about sugar being uniquely harmful or beneficial during early life – it’s about the importance of maintaining perspective when interpreting observational studies and not letting provocative headlines drive health decisions. As we continue to investigate these important questions, we need to rely on more rigorous evidence before making sweeping dietary recommendations that might unnecessarily alarm parents and pregnant women.

For a list of all previous weekly emails, click here.