儿童早期体重指数轨迹与超重风险关联的前瞻性队列研究

doi:10.19723/j.issn.1671-167X.2024.03.003

摘要/Abstract

摘要：

目的: 基于前瞻性队列比较不同方法确定的儿童早期体重指数(body mass index, BMI)轨迹与超重风险的关联，识别在儿童早期生长发育重要窗口期具有较高肥胖风险的人群。方法: 共纳入北大通州出生队列的1 330名儿童，分别在儿童刚出生, 1、3、6、9、12、18、24月龄和3岁进行随访，根据其身长/身高和体质量计算BMI Z评分。应用潜类别增长混合模型(growth mixture modelling, GMM)和基于纵向数据的k-means聚类方法(k-means for longitudinal data, KML)分析儿童早期(从出生至24月龄)BMI轨迹分组，采用线性回归比较不同方法确定的儿童早期BMI轨迹和儿童3岁时BMI Z评分的关联，通过五折交叉验证的平均受试者工作特征曲线下面积，比较Logistic回归模型中不同方法确定的儿童早期BMI轨迹分组对儿童3岁超重风险(BMI Z评分>1)的预测性能。结果: 在纳入的研究对象中，用GMM确定的三分类轨迹分为低、中、高轨迹，分别占39.7%、54.1%、6.2%；用KML方法确定的二分类轨迹分为低轨迹和高轨迹，分别占50.3%和49.7%；用KML方法确定的三分类轨迹分为低、中、高轨迹，分别占31.1%、47.4%、21.5%。用不同方法确定的儿童早期BMI轨迹反映的生长模式存在一定差异。线性回归分析发现，在调整母亲民族、受教育水平、分娩方式、产次、分娩时年龄、分娩孕周、胎儿性别和1月龄母乳喂养等协变量后，用KML方法确定的三分类轨迹中的高轨迹组(表现为出生时BMI Z评分略高、婴儿期快速生长后持续稳定在高水平的生长模式)和儿童3岁BMI Z评分的关联最强。Logistic回归分析发现，KML三分类轨迹分组对儿童3岁超重风险具有最佳的预测效果。在额外调整儿童平衡膳食指数正端分、平均每天身体活动时间和视屏时间后，结果基本一致。结论: 采用不同方法识别具有不同变化特征的儿童早期BMI轨迹，发现KML方法确定的高轨迹组能更好地发现儿童早期具有较高超重风险的人群，为选择适宜方法开展儿童早期BMI轨迹相关研究提供了依据。

关键词: 儿童早期, BMI轨迹, 超重, 前瞻性队列

Abstract:

Objective: To compare the association between body mass index (BMI) trajectories determined by different methods and the risk of overweight in early childhood in a prospective cohort study, and to identify children with higher risk of obesity during critical growth windows of early childhood. Methods: A total of 1 330 children from Peking University Birth Cohort in Tongzhou (PKUBC-T) were included in this study. The children were followed up at birth, 1, 3, 6, 9, 12, 18, and 24 months and 3 years of age to obtain their height/length and weight data, and calculate BMI Z-score. Latent class growth mixture modeling (GMM) and longitudinal data-based k-means clustering algorithm (KML) were used to determine the grouping of early childhood BMI trajectories from birth to 24 mouths. Linear regression was used to compare the association between early childhood BMI trajectories determined by different methods and BMI Z-score at 3 years of age. The predictive performance of early childhood BMI trajectories determined by different methods in predicting the risk of overweight (BMI Z-score > 1) at 3 years was compared using the average area under the curve (AUC) of 5-fold cross-validation in Logistic regression models. Results: In the study population included in this research, the three-category trajectories determined using GMM were classified as low, medium, and high, accounting for 39.7%, 54.1%, and 6.2% of the participants, respectively. The two-category trajectories determined using the KML method were classified as low and high, representing 50. 3% and 49. 7% of the participants, respectively. The three-category trajectories determined using the KML method were classified as low, medium, and high, accounting for 31.1%, 47.4%, and 21.5% of the participants, respectively. There were certain differences in the growth patterns reflected by the early childhood BMI trajectories determined using different methods. Linear regression analysis found that after adjusting for maternal ethnicity, educational level, delivery mode, parity, maternal age at delivery, gestational week at delivery, children' s gender, and breastfeeding at 1 month of age, the association between the high trajectory group in the three-category trajectories determined by the KML method (manifested by a slightly higher BMI at birth, followed by rapid growth during infancy and a stable-high BMI until 24 months) and BMI Z-scores at 3 years was the strongest. Logistic regression analysis revealed that the three-category trajectory grouping determined by the KML method had the best predictive performance for the risk of overweight at 3 years. The results were basically consistent after additional adjustment for the high bound score of the child' s diet balanced index, average daily physical activity time, and screen time. Conclusion: This study used different methods to identify early childhood BMI trajectories with varying characteristics, and found that the high trajectory group determined by the KML method was better able to identify children with a higher risk of overweight in early childhood. This provides scientific evidence for selecting appropriate methods to define early childhood BMI trajectories.

Key words: Early childhood, BMI trajectory, Overweight, Prospective cohort

中图分类号:

R179

岳芷涵,韩娜,鲍筝,吕瑾莨,周天一,计岳龙,王辉,刘珏,王海俊. 儿童早期体重指数轨迹与超重风险关联的前瞻性队列研究[J]. 北京大学学报（医学版）, 2024, 56(3): 390-396.

Zhihan YUE,Na HAN,Zheng BAO,Jinlang LYU,Tianyi ZHOU,Yuelong JI,Hui WANG,Jue LIU,Haijun WANG. A prospective cohort study of association between early childhood body mass index trajectories and the risk of overweight[J]. Journal of Peking University (Health Sciences), 2024, 56(3): 390-396.

图/表 5

图1

图2

表1

表2

表3

参考文献 25

1	Barker DJ . The fetal and infant origins of adult disease[J]. BMJ, 1990, 301 (6761): 1111. doi: 10.1136/bmj.301.6761.1111
2	Geserick M , Vogel M , Gausche R , et al. Acceleration of BMI in early childhood and risk of sustained obesity[J]. N Engl J Med, 2018, 379 (14): 1303- 1312. doi: 10.1056/NEJMoa1803527
3	Yuan Y , Chu C , Zheng WL , et al. Body mass index trajectories in early life is predictive of cardiometabolic risk[J]. J Pediatr, 2020, 219, 31-37, e36. doi: 10.1016/j.jpeds.2019.12.060
4	Aris IM , Bernard JY , Chen LW , et al. Infant body mass index peak and early childhood cardio-metabolic risk markers in a multi-ethnic Asian birth cohort[J]. Int J Epidemiol, 2017, 46 (2): 513- 525.
5	den Dekker HT , Jaddoe VWV , Reiss IK , et al. Fetal and infant growth patterns and risk of lower lung function and asthma. The generation R study[J]. Am J Respir Crit Care Med, 2018, 197 (2): 183- 192. doi: 10.1164/rccm.201703-0631OC
6	Mattsson M , Maher GM , Boland F , et al. Group-based trajectory modelling for BMI trajectories in childhood: A systematic review[J]. Obes Rev, 2019, 20 (7): 998- 1015. doi: 10.1111/obr.12842
7	Cao T , Zhao J , Hong X , et al. Cord blood metabolome and BMI trajectory from birth to adolescence: A prospective birth cohort study on early life biomarkers of persistent obesity[J]. Metabolites, 2021, 11 (11): 739. doi: 10.3390/metabo11110739
8	Meir AY , Huang W , Cao T , et al. Umbilical cord DNA methylation is associated with body mass index trajectories from birth to adolescence[J]. EBioMedicine, 2023, 91, 104550. doi: 10.1016/j.ebiom.2023.104550
9	Jin C , Lin L , Han N , et al. Effects of dynamic change in fetuin-A levels from the first to the second trimester on insulin resistance and gestational diabetes mellitus: A nested case-control study[J]. BMJ Open Diabetes Res Care, 2020, 8 (1): e000802. doi: 10.1136/bmjdrc-2019-000802
10	World Health Organization. WHO child growth standards: Length/height-for-age, weight-for-age, weight-for-length, weight-for-height and body mass index-for-age: Methods and development [R]. Geneva: World Health Organization, 2006.
11	de Onis M , Lobstein T . Defining obesity risk status in the general childhood population: Which cut-offs should we use[J]. Int J Pediatr Obes, 2010, 5 (6): 458- 460. doi: 10.3109/17477161003615583
12	World Health Organization, the United Nations Children' s Fund (UNICEF). Indicators for assessing infant and young child feeding practices: Definitions and measurement methods [R]. Geneva: World Health Organization and the United Nations Children' s Fund (UNICEF), 2021.
13	房玥晖, 何宇纳, 李春丽. 基于中国学龄前儿童平衡膳食指数的2010—2012年中国学龄前儿童膳食质量评价[J]. 中华预防医学杂志, 2020, 54 (6): 662- 667.
14	Wen LM , van der Ploeg HP , Kite J , et al. A validation study of assessing physical activity and sedentary behavior in children aged 3 to 5 years[J]. Pediatr Exerc Sci, 2010, 22 (3): 408- 420. doi: 10.1123/pes.22.3.408
15	Proust-Lima C , Philipps V , Liquet B . Estimation of extended mixed models using latent classes and latent processes: The R package lcmm[J]. J Stat Softw, 2017, 78 (2): 1- 56.
16	Genolini C , Alacoque X , Sentenac M , et al. kml and kml3d: R packages to cluster longitudinal data[J]. J Stat Softw, 2015, 65 (4): 1- 34.
17	Stekhoven DJ , Bühlmann P . MissForest: Non-parametric missing value imputation for mixed-type data[J]. Bioinformatics, 2012, 28 (1): 112- 118. doi: 10.1093/bioinformatics/btr597
18	Hu Z , Tylavsky FA , Han JC , et al. Maternal metabolic factors during pregnancy predict early childhood growth trajectories and obesity risk: The CANDLE study[J]. Int J Obes (Lond), 2019, 43 (10): 1914- 1922. doi: 10.1038/s41366-019-0326-z
19	Montazeri P , Vrijheid M , Martinez D , et al. Maternal metabolic health parameters during pregnancy in relation to early childhood BMI trajectories[J]. Obesity (Silver Spring), 2018, 26 (3): 588- 596. doi: 10.1002/oby.22095
20	Michael N , Gupta V , Fogel A , et al. Longitudinal characterization of determinants associated with obesogenic growth patterns in early childhood[J]. Int J Epidemiol, 2023, 52 (2): 426- 439. doi: 10.1093/ije/dyac177
21	李春刚, 严双琴, 高国朋, 等. 母亲孕期饮食模式与儿童早期BMI变化轨迹关联的队列研究[J]. 中华流行病学杂志, 2023, 44 (11): 1769- 1775.
22	Xiong C , Chen K , Xu LL , et al. Associations of prenatal exposure to bisphenols with BMI growth trajectories in offspring within the first two years: Evidence from a birth cohort study in China[J]. World J Pediatr, 2023, 2023-11-29 (2024-02-01)
23	Nguena Nguefack HL , Pagé MG , Katz J , et al. Trajectory modelling techniques useful to epidemiological research: A comparative narrative review of approaches[J]. Clin Epidemiol, 2020, 12, 1205- 1222.
24	Den Teuling NGP , Pauws SC , van den Heuvel ER . A comparison of methods for clustering longitudinal data with slowly changing trends[J]. Commun Stat Simul Comput, 2023, 52 (3): 621- 648.
25	Huang W , Meir AY , Olapeju B , et al. Defining longitudinal trajectory of body mass index percentile and predicting childhood obesity: Methodologies and findings in the Boston birth cohort[J]. Precis Nutr, 2023, 2 (2): e00037.

Metrics

Viewed

Full text

157

HTML			PDF

Just accepted	Online first	Issue	Just accepted	Online first	Issue
0	0	23	0	0	134

From	Others	local

Times	65	92
Rate	41%	59%

Abstract

224

Just accepted	Online first	Issue

0	0	224

From	Others	local

Times	193	31
Rate	86%	14%

Cited

Web of Science	Crossref	ScienceDirect	Search for Citations in Google Scholar >>


This page requires you have already subscribed to WoS.

Shared

Discussed

Items	Overweight risk at 3 years	Non-overweight risk at 3 years	P
Participants in the study	172 (12.9)	1 158 (87.1)
BMI Z-score at 3 years	1.85±0.84	-0.42±0.84	<0.001
GMM (three-category trajectories)			<0.001
Class1	45 (8.5)	483 (91.5)
Class2	108 (15.0)	612 (85.0)
Class3	19 (23.2)	63 (76.8)
KML (two-category trajectories)			<0.001
Class1	16 (2.4)	653 (97.6)
Class2	156 (23.6)	505 (76.4)
KML (three-category trajectories)			<0.001
Class1	8 (1.9)	405 (98.1)
Class2	56 (8.9)	575 (91.1)
Class3	108 (37.8)	178 (62.2)
Maternal ethnicity (missing=6)			0.534
Han	160 (12.8)	1 089 (87.2)
Minority	12 (16.0)	63 (84.0)
Maternal educational level (missing=6)			0.037
Low (high school and below)	38 (13.5)	244 (86.5)
Middle (junior college)	75 (15.8)	401 (84.2)
High (university and above)	59 (10.4)	507 (89.6)
Parity			0.700
Primipara	100 (12.6)	695 (87.4)
Multipara	72 (13.5)	463 (86.5)
Caesarean section (missing=3)			0.396
No	87 (12.2)	628 (87.8)
Yes	85 (13.9)	527 (86.1)
Maternal age at delivery/years	30.20±3.78	30.41±3.97	0.512
Gestational age at birth/weeks	39.26±1.48	39.41±1.40	0.180
Children’s gender			0.806
Boys	88 (13.2)	577 (86.8)
Girls	84 (12.6)	581 (87.4)
Exclusive breastfeeding at the first month (missing=150)			0.035
No	49 (16.8)	242 (83.2)
Yes	105 (11.8)	784 (88.2)
Subset: participants with children’s questionnaires	73 (16.9)	360 (83.1)
High bound score of DBI-C at 3 years (missing=20)	13.76±9.29	12.26±8.64	0.197
Total physical activity at 3 years (hours per day)	2.13±1.64	2.76±2.43	0.037
Screen time at 3 years (hours per day)	0.87±1.13	0.61±0.79	0.019

Items	Crude model (n=1 330)	Adjusted model 1 (n=1 172)	Adjusted model 2 (subset: n=358)
GMM (three-category trajectories)
Class3 vs. Class1	0.557 (0.296, 0.818)	0.787 (0.477, 1.097)	1.333 (0.735, 1.931)
Class3 vs. Class2	0.313 (0.057, 0.570)	0.484 (0.188, 0.780)	1.052 (0.486, 1.617)
Class3 vs. Class1+Class2	0.416 (0.164, 0.668)	0.567 (0.271, 0.862)	1.122 (0.558, 1.686)
KML (two-category trajectories)
Class2 vs. Class1	1.204 (1.101, 1.307)	1.237 (1.124, 1.349)	1.447 (1.223, 1.671)
KML (three-category trajectories)
Class3 vs. Class1	1.774 (1.633, 1.916)	1.828 (1.673, 1.983)	1.958 (1.656, 2.260)
Class3 vs. Class2	1.004 (0.872, 1.135)	1.008 (0.867, 1.150)	1.116 (0.849, 1.383)
Class3 vs. Class1+Class2	1.309 (1.178, 1.439)	1.324 (1.183, 1.465)	1.418 (1.156, 1.680)

Five-fold CV AUC	Crude model (n=1 330)	Adjusted model 1 (n=1 172)	Adjusted model 2 (subset: n=358)
GMM (three-category trajectories)	0.590±0.033	0.573±0.082	0.638±0.066
KML (two-category trajectories)	0.735±0.022	0.771±0.031	0.777±0.066
KML (three-category trajectories)	0.783±0.015	0.799±0.057	0.810±0.045

[1]	陈敬,单蕊,肖伍才,张晓蕊,刘峥. 青春期和成年早期自制力与抑郁症状和超重肥胖共病风险的关联：基于全国调查的十年前瞻性队列研究[J]. 北京大学学报（医学版）, 2024, 56(3): 397-402.
[2]	吴一凡,玉应香,谢岚,张志达,常翠青. 不同体重指数青年男性的静息能量消耗特点及预测方程评价[J]. 北京大学学报（医学版）, 2024, 56(2): 247-252.
[3]	党佳佳,蔡珊,钟盼亮,王雅琪,刘云飞,师嫡,陈子玥,张依航,胡佩瑾,李晶,马军,宋逸. 室外夜间人工光暴露与中国9~18岁儿童青少年超重肥胖的关联[J]. 北京大学学报（医学版）, 2023, 55(3): 421-428.
[4]	刘杰,郭超. 正/负性情绪对中国老年人死亡风险影响的前瞻性队列研究[J]. 北京大学学报（医学版）, 2022, 54(2): 255-260.
[5]	董彦会，宋逸，董彬，邹志勇，王政和，杨招庚，王西婕，李艳辉，马军. 2014年中国7~18岁学生血压状况与营养状况的关联分析——基于中国儿童青少年血压评价标准[J]. 北京大学学报(医学版), 2018, 50(3): 422-428.
[6]	程兰，李钦，宋逸，马军，王海俊. 中国9~11岁小学生体育锻炼、静态行为和#br# 超重与肥胖的关系[J]. 北京大学学报(医学版), 2016, 48(3): 436-441.
[7]	谢高强1△, 于晖2, 陈敬洲2, 赵连成3△, 任福秀4, 史平4, 武阳丰1, 5, 6. 基因变异及心血管危险因素与单核细胞分泌白细胞介素6和10的关系[J]. 北京大学学报（医学版）, 2014, 46(4): 589-595.
[8]	付连国, 马军, 王海俊, 董彬, 王晶晶, 尚晓瑞, 张子龙, 刘芳宏, 蔡赐河, 文晗. 儿童青少年超重、肥胖对其身体机能影响的多水平分析[J]. 北京大学学报（医学版）, 2012, 44(3): 359-363.