Researchers analyzed two generations of Danish conscripts and discovered that while obesity rates soared, the inverse relationship between BMI and intelligence remained stable, challenging assumptions about shifting cognitive trends.
Study: Intelligence and obesity during the obesity epidemic. Image Credit: DenisProduction.com / Shutterstock
In a recent study published in the journal Scientific Reports, researchers investigated whether the increased prevalence of overweight and obesity during the obesity epidemic has altered the relationship between body mass index (BMI) and intelligence.
Numerous studies over the past 50 years have uncovered an inverse association between higher BMI and intelligence, viz., lower intelligence among obese individuals or higher obesity prevalence in people with low intelligence. Many countries have recorded a considerable increase in obesity prevalence, referred to as the obesity epidemic, during this period. This raises concerns about whether the relationship between intelligence, BMI, and obesity has altered during this period.
Prior Danish studies involving conscripts born before 1960 have revealed a J-shaped, inverse association between BMI and intelligence, with lower intelligence at higher BMI. One study also observed that this pattern remained stable over time, despite the rising obesity prevalence. The researchers noted that this may suggest that the inverse relationship between BMI and intelligence was historically more pronounced in a smaller subset of individuals with a particular type of obesity. However, the overall relationship has not significantly changed in recent decades.
About the study
In the present study, researchers evaluated whether the increase in obesity/overweight prevalence has impacted the relationship between BMI and intelligence in two Danish cohorts. The cohorts included males born between 1939 and 1959 (early cohort) and 1983 and 2001 (late cohort) in Denmark. All males appeared before a conscription board for cognitive and physical examinations around age 18 but before age 27. The examination comprised assessments of weight, height, and intelligence.
Height and weight were measured at the examination, and BMI was calculated. Intelligence was assessed using the Børge Priens Prøve test, a standardized cognitive test that has remained unchanged since 1957, which had four subtests: letter matrices, number series, geometric figures, and verbal analogies. The number of correct answers was aggregated as the intelligence test score (ITS). The researchers used logistic and multiple linear regression models, including non-linear terms for statistical analyses, with birth year, height, and geographic district as covariates.
Findings
The early cohort comprised data of over 728,000 males who appeared before the conscription board between 1957 and 1984. Likewise, the late cohort included over 514,000 conscripts examined between 2006 and 2019. Of these, only 258,882 and 162,250 subjects from the early and late cohorts, respectively, were included in the analyses.
The early cohort participants had a lower BMI than the late cohort subjects. The late cohort showed a substantially higher obesity prevalence at 6.7% than the early cohort (0.8%). Similarly, the early cohort had a lower prevalence of overweight at 7.9% than the late cohort (21.3%). The late cohort had a slightly higher mean ITS (41.1) than the early cohort (39.4), though the overall intelligence distribution differed between cohorts.
Further, obesity prevalence decreased across ITS quartiles in both cohorts, albeit higher in the late cohort compared to the early cohort across quartiles. The logistic model showed a consistent and monotonic, inverse association between obesity and ITS. Importantly, there were no statistically significant differences in this association between cohorts, indicating that the relationship has remained stable despite the obesity epidemic.
Each cohort had only minor differences in average BMI across ITS quartiles. The highest BMI was among participants in the lowest ITS quartiles. Notably, the mean ITS was the highest in participants with normal weight, followed by underweight individuals. The study confirmed a J-shaped, inverse association between ITS and BMI, where intelligence scores peaked at a BMI of approximately 20 kg/m² before declining. The decline in intelligence scores continued until plateauing around a BMI of 38 kg/m², though this decline was somewhat steeper in the early cohort.
Conclusions
Together, the prevalence of obesity showed a consistent and monotonic inverse association with intelligence, with no significant differences between the two cohorts. Besides, the analyses on the prevalence of overweight yielded comparable results. Despite the rising prevalence of obesity and overweight, the study found that the fundamental inverse relationship between BMI and intelligence has persisted.
The researchers emphasized that the use of advanced statistical models, including non-linear regression, confirmed the robustness of these findings. They suggested that investigating the mechanisms behind the persistence of this association could help better understand the underlying factors driving the obesity epidemic. This could inform public health strategies aimed at mitigating the impact of obesity while considering its complex relationship with cognitive function.