New research reveals that larger cities see lower rates of both obesity and impulsivity, uncovering how lifestyle, education, and mental healthcare shape healthier urban populations.
Study: Investigating the link between impulsivity and obesity through urban scaling laws. Image Credit: Markus Mainka / Shutterstock
In a recent article published in the journal PLOS Complex Systems, researchers explored the link between impulsivity and obesity across 915 cities in the United States. Their findings indicate that obesity and impulsivity, measured by the prevalence of attention deficit hyperactivity disorder (ADHD), were less common in larger cities. ADHD appeared to influence obesity levels, with lifestyle acting as a moderating factor.
Background
Obesity is a growing global health crisis, particularly in the U.S., where its prevalence is expected to rise significantly by 2030. While various factors like behavior, genetics, and environment contribute to obesity, impulsivity, defined as acting without foresight, has emerged as a key psychological factor.
Though impulsivity can be adaptive in some contexts, excessive impulsivity is linked to poor food choices and weight gain. ADHD, a clinical form of impulsivity, has shown consistent associations with obesity in epidemiological, genetic, and pharmacological studies across diverse populations, including Dutch and Korean children.
However, most research overlooks how environmental features, particularly those of urban settings, may influence this link. Obesogenic environments vary across cities, including limited access to physical activity, healthy food, and social support.
Urban science, which studies how city features scale with population size, offers tools to explore this complexity. Urban scaling laws reveal how health outcomes, like obesity and mental disorders, change with city size. The study hypothesizes that lower ADHD prevalence in larger cities could stem from increased genetic diversity or better mental health access, though these explanations remain speculative.
About the Study
In this study, researchers applied a novel causal inference method to understand how ADHD and urban features influence obesity across American cities. The study also analyzes individual-level data from over 19,000 children to ensure robustness.
The study used both individual-level and city-level datasets to explore how factors like physical activity, obesity, ADHD, food insecurity, education, and mental healthcare access relate to urban population size and each other.
City-level data included physical inactivity, adult obesity, mental health service access, college education, and food insecurity. These data were grouped into 915 U.S. micropolitan and metropolitan areas.
Individual-level data included health and demographic data on one randomly selected child (aged 10–17) per household. Variables included body mass index (BMI) category, physical activity (days/week), ADHD severity, household food insufficiency, mental health service utilization, and caretaker education level. The final dataset included over 19,000 children after cleaning.
Urban scaling laws were modeled using ordinary least squares (OLS) regression on log-transformed data, with heteroskedasticity-consistent standard errors. The Gini index (adapted for negative values) measured within-state inequality in health and social indicators.
Causal relationships among variables were inferred using the Peter-Clark algorithm, which identifies associations suggesting causal links by testing conditional independence. While useful, this method assumes no hidden variables or feedback loops, which may not always hold. The study avoided combining individual and city-level information in causal models due to differences in data type, age groups, and missing location information.
Findings
At the city level (915 American cities), urban scaling analysis revealed that ADHD in children, adult obesity, and physical inactivity all scaled sublinearly with population size, indicating lower per-capita prevalence in larger cities.
In contrast, mental health service access and college education scaled superlinearly, being more common in larger cities, while food insecurity scaled linearly. Notably, smaller cities exhibited up to 30% higher probabilities of physical inactivity compared to larger ones.
Using scale-adjusted metropolitan indicators, the team applied a causal discovery algorithm to uncover key associations: physical inactivity led to increased obesity, and ADHD prevalence was associated with higher physical inactivity and food insecurity.
Mental health provider availability reduced physical inactivity, while college education was associated with better mental health access and less food insecurity. These links are correlational but align with known biological pathways, such as brain circuits regulating impulse control and dopamine-related genes.
At the individual level (data from over 19,000 children), the patterns mirrored those found in cities. ADHD severity correlated with less physical activity and greater BMI, suggesting both direct (e.g., poor dietary choices) and indirect (e.g., reduced exercise) pathways between ADHD and obesity.
Additionally, researchers noted the protective nature of adult education in households, being linked to better access to mental healthcare, lower food insufficiency, and healthier BMI in children, though possibly also lower physical activity time.
Conclusions
This study shows that overall well-being increases with city size: obesity, food insecurity, ADHD, and inactivity decrease in larger cities, while college education and mental healthcare access increase.
Causal analysis suggests ADHD leads to obesity through reduced physical activity. College education and food security indirectly reduce obesity by encouraging more physical activity.
Individual-level information supports these city-level patterns, highlighting ADHD and impulse control in obesity, with potential biological links involving brain function (e.g., anterior cingulate cortex) and genetic factors like dopaminergic signaling.
The study’s strength lies in combining large-scale city data with individual-level insights. Limitations include assumptions in the causal algorithm, potential hidden variables, mismatched age ranges, and inability to link individuals to specific cities. The focus on U.S. data also limits generalizability, though international studies hint at broader relevance.
Still, findings suggest targeted policies promoting physical activity and education may help reduce obesity, especially in smaller or underserved communities.