Background and objectives: While midlife obesity is consistently associated with cognitive decline in later life, there is limited understanding of how weight variations from early life to midlife affect cognitive decline. We investigated the association between early- to midlife weight trajectories and mid- to late-life cognitive decline.

Methods: This is a longitudinal cohort study that used data from 3 waves (2008-2019) of the Brazilian Longitudinal Study of Adult Health, a multicenter cohort study that enrolled active and retired public servants aged 35+ years from public universities in Brazil. Participants with a history of stroke, missing cognitive data at baseline, and with incomplete body shape data were excluded from the analyses. Self-reported body shapes from ages 5 to 40 using the Stunkard Figure Rating Scale were categorized as underweight, normal, overweight, and obese. Sequence analysis and hierarchical clustering identified weight trajectories. Global cognition Z-scores were derived from memory (immediate recall, delayed recall, and recognition of a word list), phonemic and semantic verbal fluency, and Trail Making Test part B (TMT-B). Linear mixed models adjusted for sociodemographic, clinical, and lifestyle covariates investigated associations between clusters of weight trajectories and global cognition Z-scores. Inverse probability of attrition weighting was used to account for attrition bias.

Results: Among 11,361 participants (mean age: 51.5 ± 8.6, 55% women, 42.4% Black/mixed race), "normal to overweight," "underweight to normal," and "stable overweight" trajectories exhibited faster global cognitive decline than "stable normal" trajectory (β = -0.024; 95% CI -0.043 to -0.005; p = 0.015; β = -0.026; 95% CI -0.040 to -0.012; p < 0.001; β = -0.034; 95% CI -0.066 to -0.001; p = 0.043, respectively), representing 4.6-6.5 excess years of cognitive aging over a median follow-up of 8 years. Cognitive decline associated with weight trajectories was driven mainly by declines in memory and TMT-B performance. Associations were observed only in Black/mixed races and women when stratified.

Discussion: Weight gain and stable overweight trajectories from early life to midlife were associated with faster cognitive decline than stable normal weight trajectories. Weight management during early life may mitigate cognitive decline. Study limitations include reliance on self-reported body shape data, potential recall bias, and residual confounding from unmeasured early-life factors.

背景与目的： 尽管中年时期的肥胖一直与晚年认知功能下降有关，但对从童年到中年期间体重变化如何影响认知功能下降的理解有限。我们研究了早期至中期的体重轨迹与中期至晚期的认知功能下降之间的关联。

方法： 这是一项使用巴西成年人健康纵向研究（2008-2019年间三波数据）进行的纵向队列研究，该研究在巴西公共大学中招募了35岁及以上的现役和退休公务员。排除了有中风史、基线认知数据缺失以及身体形态数据不完整的参与者。使用斯通卡德体型评分量表（Stunkard Figure Rating Scale），自报从5岁至40岁的体型被分类为体重过轻、正常、超重和肥胖。序列分析和分层聚类识别了体重轨迹。全局认知Z分数根据记忆（即时回忆、延迟回忆以及单词列表再认）、语音流畅性和语义流畅性及迷宫追踪测试B部分（TMT-B）得出。线性混合模型调整了社会人口学、临床和生活方式协变量，以调查体重轨迹群组与全局认知Z分数之间的关联。使用脱落倾向加权来解决流失偏差。

结果： 在11,361名参与者（平均年龄：51.5±8.6岁；女性占55%；黑人/混血占42.4%）中，“正常至超重”、“体重过轻至正常”和“稳定超重”轨迹比“稳定正常”轨迹表现出更快的全局认知下降（β=-0.024；95% CI -0.043到-0.005；p=0.015；β=-0.026；95% CI -0.040到-0.012；p

讨论： 从童年到中年时期的体重增加或稳定超重轨迹比稳定正常体重轨迹更快地导致认知功能下降。在早期生命阶段管理体重可能有助于减缓认知功能的下降。该研究的局限性包括依赖自报的身体形态数据、潜在的记忆偏差以及来自未测量的早年生活因素的残余混杂。