A childhood shaped by sugar rationing was linked to mental health more than 50 years later

By Vijay Kumar MalesuReviewed by Susha Cheriyedath, M.Sc.Aug 23 2026

A postwar rationing experiment offered researchers a rare window into how sugar exposure during the earliest stages of life might leave traces in diet, brain structure, and mental health decades later.

Study: Dietary sugar exposure in early life and risk of adult mental health disorders: UK Biobank cohort study. Image Credit: Rohane Hamilton / Shutterstock

In a recent study accepted for publication in the journal Translational Psychiatry, a group of researchers examined associations between early-life sugar-rationing exposure and adult sweet-food preference, sugar intake, anxiety, depression, and brain-related traits.

Background

Could the amount of sugar encountered before birth and during the first two years of life shape mental health decades later?

Diet quality is linked to metabolic and mental health, while higher sugar intake and stronger preferences for sweet foods have been associated with depression and anxiety. Food liking can serve as a behavioral proxy for intake, as people with stronger preferences for sweet foods tend to consume more of them.

Depression and anxiety are multifactorial conditions influenced by biological, genetic, and environmental factors, with epigenetic mechanisms potentially contributing to depression risk. Sugar rationing during the first 1,000 days has previously been linked with chronic disease.

About the study

The analysis used data from the United Kingdom (UK) Biobank, focusing on births between October 1951 and March 1956. After exclusions, 46,448 participants remained. Exposure was categorized according to how long participants experienced sugar rationing, ranging from exposure in utero only to exposure throughout pregnancy and the first two years of life.

The groups were classified as R270 (in utero only), R425 (in utero to 6 months), R635 (in utero to 12 months), R817 (in utero to 18 months), and R1000 (in utero to 24 months). Two groups born after sugar rationing ended served as comparisons, with participants conceived after all food rationing had ended used as the main reference group.

Sweet-food preference and sugar intake were assessed using questionnaires and Oxford WebQ dietary recalls. Anxiety and depression were identified from linked health records using the International Classification of Diseases, 10th Revision (ICD-10). Type 1 diabetes and post-traumatic stress disorder (PTSD) were included as negative control conditions.

Survival models examined the relationship between early-life rationing exposure and later anxiety and depression while accounting for demographic, lifestyle, and health factors, including body mass index, smoking, alcohol intake, sleep, family mental health history, and inflammation.

Additional analyses tested the robustness of the findings, including approaches to account for missing data and sensitivity analyses examining later-life sugar intake.

Brain magnetic resonance imaging (MRI) data from 5,990 participants were analyzed to identify differences in gray matter volume across rationing groups, with statistical correction for multiple comparisons. Structural equation modeling (SEM) was used to explore relationships among rationing, sweetness preference, sugar intake, inflammation, mental health, and MRI traits in 2,960 participants.

Study results

During follow-up, 2,776 participants were diagnosed with anxiety and 3,147 with depression. Adult sugar consumption was similar across the rationing groups, averaging about 58-59 g/day of free sugar and 122-124 g/day of total sugar, with no significant differences.

Sweet-food liking scores were also broadly similar. On a 9-point scale, participants exposed to rationing throughout pregnancy and the first two years of life had an average score of 5.82, compared with 6.04 in the main reference group. This was the only significant group difference in overall liking for sweet foods. Greater sweet-liking was associated with higher sugar intake, whereas rationing status itself showed little overall association with adult sugar consumption or preference.

Anxiety affected about 5.5% to 6.3% of participants across the groups, while depression affected about 6.0% to 7.2%. Compared with participants conceived after all food rationing had ended, those exposed to sugar rationing for the three longest periods had significantly lower hazards of both anxiety and depression. The shorter-exposure groups showed no significant differences.

Type 1 diabetes and PTSD showed no significant relationships with rationing exposure, providing some reassurance that the mental health findings were not part of a broader pattern across unrelated conditions.

A small Irish comparison group of 154 participants showed no significant associations between birth timing and anxiety or depression, although its small size and wide confidence intervals limited the strength of this comparison. Analyses designed to account for missing data produced similar overall findings.

The associations were stronger in women, although the direction of the findings was generally consistent across sexes. After accounting for later-life sugar intake, the anxiety finding persisted in participants exposed to rationing throughout pregnancy and the first two years of life, whereas the depression findings weakened and were no longer statistically significant.

The researchers cautioned that this sensitivity analysis was based on a smaller dietary subsample with relatively sparse dietary data, limiting conclusions about whether adult sugar intake could help explain the findings.

Among several rationed groups, the median recorded age of anxiety diagnosis was around 61 to 62 years, compared with 55.6 years in the main reference group. For depression, participants in two of the longest-exposure groups had median recorded diagnosis ages of 58.4 to 59.1 years, compared with 54.6 years in the reference group.

However, the researchers cautioned that these figures should not be interpreted as true ages of disease onset. Earlier diagnoses may have been undercaptured because linked health records were most reliable from 1997 onward.

Brain imaging identified differences across rationing groups in 80 of 139 gray matter regions. More detailed comparisons highlighted 11 regions that differed from participants conceived immediately after sugar rationing ended, including the brainstem, occipital fusiform gyrus, and several cerebellar regions.

Structural equation modeling was then used to examine the relationships among rationing exposure, sweet-food preference, sugar intake, inflammation, brain structure, and mental health.

In this exploratory analysis, longer rationing exposure was directly associated with lower anxiety but not depression. Greater sweet-food liking was linked with higher sugar intake and also correlated with brain MRI traits.

The model did not support inflammation or brain MRI traits as mediators of the relationship between rationing and depression, although inflammation itself was positively associated with depression.

Conclusions

The researchers concluded that longer exposure to sugar rationing during the first 1,000 days of life was linked to lower hazards of anxiety and depression in adulthood, with the clearest findings among participants exposed for longer periods.

The anxiety finding remained after accounting for later-life sugar intake, whereas the depression finding weakened. The longest rationing exposure was also linked to a modestly lower sweet-food liking score, while rationing groups differed in gray matter volume.

Exploratory modeling supported a direct relationship between longer rationing exposure and lower anxiety but did not identify a similar direct relationship with depression or evidence that inflammation or brain MRI traits mediated the observed associations.

The findings were observational and do not establish that limiting sugar exposure in early life causes better mental health. Broader differences between historical birth cohorts, including socioeconomic, environmental, healthcare, and other factors, could not be ruled out.

The researchers called for further epigenetic, gene-environment, inflammatory, and neuroimaging studies to investigate how early nutritional environments might influence mental health later in life.