Adolescent Obesity and Cognitive Performance: A Nationwide study of 2.48 Million Israeli Adolescents - הכינוס השנתי של החברה הישראלית לפדיאטריה קלינית

מאיה שמחוני ¹ Estela Derazne ² Orit Pinhas-Hamiel ^2,3 Tali Cukierman-Yaffe ^2,4 Cole D. Bendor ¹ Aya Bardugo ¹ Arnon Afek ^2,5 Gilad Twig ^1,2,4

¹Department of Military Medicine, Hebrew University of Jerusalem and the Israel Defense Forces Medical Corps, Ramat Gan, Israel.
²Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel.
³Pediatric Endocrinology and Diabetes unit, Edmond and Lily Safra Children’s hospital, Sheba medical center, Tel Hasomer, Ramat Gan, Israel
⁴Division of Endocrinology, Diabetes and Metabolism, Sheba Medical Center, Tel Hashomer, Ramat Gan, Israel.
⁵Central Management, Sheba Medical Center, Tel Hashomer, Ramat Gan, Israel.

Aims: We assessed in a nationwide cross-sectional study the association between adolescent BMI and cognitive function.

Methods: Israeli born adolescents aged 16-19 years old, including 1,459,522 males and 1,027,953 females, were evaluated for military service between 1967-2018. Weight and height were measured to calculate BMI. Cognitive performance was assessed by a validated Intelligence-Quotient-equivalent test and was standardized into year- and sex-Z-score. For 445,385 persons the parental cognitive scores could be identified from early years of the study sample. Multinomial logistic regression models were applied.

Results: 29.4% of male adolescents with severe obesity gained a cognitive score below the 25th percentile, compared to 17.7% among their normal-weighted (50th-84th percentile) counterparts. There was a J-shaped relationship between BMI and odds ratios for low cognitive score, with a non-overlapping 95% CI between consequent groups: underweight, 1.45 (1.43-1.48); overweight, 1.13 (1.12-1.15); mild obesity, 1.36 (1.33-1.39); and severe obesity, 1.58 (1.52-1.64). Point estimates were overall consistent in unadjusted and multivariable models adjusted for socio-demographic confounders and persisted in sensitivity analyses that account for coexisting morbidities and parental cognitive scores at adolescence. Similar findings were observed in females. When analyzing the ORs for high cognitive performance, a mirror image to the findings above was observed. In both sexes examinees with abnormal BMI had higher ORs for lower-than-expected cognitive score, based on their parents data as adolescents, in manner that depended on obesity severity.

Conclusions: obesity is associated with an increased risk for lower cognitive performance and inability to fully achieve cognitive potential, independently of socio-demographic background.