Breastfeeding Linked to Better Cognitive Outcomes After Socioeconomic Status Adjustment: JAMA

In this cross-sectional study, the generalized linear model was used in conjunction with survey weighting to analyze cognitive performance measured by age-standardized $z$ scores. Poor performance is considered a score that is equal to or lower than the 15th percentile on standard tests. The distinctive characteristic of this study is the socio-economic pattern in which the subjects were involved, as prolonged breastfeeding was observed in families who had lower incomes and lower parental education.

Key findings:

• Among the 5,436 subjects (mean age, 12.03 years; 52.5% male), 86.2% (4,686 subjects) were breastfed for greater than six months, while 13.8% (750 subjects) were breastfed for six months or shorter periods.
• Adolescents from lower socioeconomic backgrounds had an increased likelihood of being breastfed for more prolonged durations (26.2%) compared to those from higher SES (17.7%).
• Likewise, extended breastfeeding was more prevalent in adolescents whose mothers had lower educational attainments (31.2%) than higher (24.1%) as well as fathers who had lesser levels of education (17.3% vs. 13.2%).
• After adjusting for potential confounders, a positive but statistically insignificant relationship between breastfeeding duration and both math (β=0.03; 95% CI, -0.06 to 0.13) and word recognition (β=-0.02; 95% CI, -0.13 to 0.09) test results.
• However, the beneficial effects of breastfeeding were statistically significant once the model controlled for SES.
• Adolescents who were breastfed for more than six months had higher z scores for mathematics (β = 0.14; 95% CI, 0.05-0.22) and word recognition (β = 0.12; 95% CI, 0.02-0.21).
• Moreover, the protective effects of breastfeeding against low cognitive function were very considerable.
• Moreover, the protective role of breastfeeding regarding poor cognitive function was considerable.
• It was noted that the greater the duration of breastfeeding, the reduced probability of poor mathematical ability (OR, 0.65; 95% CI, 0.48-0.88) and poor word recognition (OR, 0.64; 95% CI, 0.47-0.87).

The present research involving young Chinese subjects demonstrates that prolonged breastfeeding for over six months positively impacts math skills and word recognition skills, assuming socio-economic status has been controlled for. It can be seen that the breastfeeding effect is a tangible reality and continues to manifest even when individuals reach their teenage years, regardless of whether breastfeeding is more prevalent among poorer people.

Reference:

Tang X, Qiu Y, Qin Z, et al. Breastfeeding Duration and Cognitive Performance Among Youths. JAMA Netw Open. 2026;9(4):e268725. doi:10.1001/jamanetworkopen.2026.8725