中国儿童保健杂志 ›› 2025, Vol. 33 ›› Issue (5): 478-483.DOI: 10.11852/zgetbjzz2024-0855

• 科研论著 • 上一篇    下一篇

学龄前儿童发育性协调障碍的影响因素及列线图预测模型的构建

王菲1, 金华2, 吴婷1, 沈梦晴1, 王云1   

  1. 1.苏州市第九人民医院儿童保健科,江苏 苏州 215200;
    2.苏州市妇幼保健院儿童保健科
  • 收稿日期:2024-08-05 修回日期:2024-11-25 发布日期:2025-04-30 出版日期:2025-05-10
  • 通讯作者: 吴婷,E-mail:wutjane@163.com
  • 作者简介:王菲(1985—),女,本科学历,副主任医师,主要研究方向为儿童营养和儿童生长发育。
  • 基金资助:
    江苏省妇幼健康科研项目(F201636)

Risk factors and nomogram prediction model of developmental coordination disorder in preschool children

WANG Fei1, JIN Hua2, WU Ting1, SHEN Mengqing1, WANG Yun1   

  1. 1. Department of Child Healthcare, Suzhou Ninth People's Hospital, Suzhou, Jiangsu 215200, China;
    2. Department of Child Healthcare, Suzhou Maternal and Child Healthcare Hospital
  • Received:2024-08-05 Revised:2024-11-25 Online:2025-05-10 Published:2025-04-30
  • Contact: WU Ting, E-mail: wutjane@163.com

摘要: 目的 调查苏州市吴江区学龄前儿童发育性协调障碍(DCD)现状,并分析其影响因素,建立列线图预测模型及验证。方法 2023年9月—2024年1月采用随机抽样法选取苏州市吴江区341例学龄前儿童作为建模集;2024年2—4月采用随机抽样法选取苏州市吴江区145例学龄前儿童作为验证集。采用儿童运动协调能力评估量表第2版(MABC-2)对其进行筛查;采用儿童发育性协调障碍问卷(DCDQ)对其进行问卷调查。比较不同性别和年龄儿童MABC-2各维度、DCDQ得分情况;单因素及多因素Logistic回归分析影响学龄前儿童发生DCD的因素;构建预测学龄前儿童发生发育性协调障碍列线图模型,绘制受试者工作特征(ROC)曲线和校准曲线分别评估列线图模型内部(建模集)及外部(验证集)验证的区分度和一致性。结果 女童手灵巧度得分高于男童,目标抓握得分低于男童(t=2.811、5.384,P<0.05)。女童、6岁儿童运动控制能力、精细运动/书写能力、一般协调性得分分别高于男童(t=2.860、5.775、3.801)、5岁儿童(t=4.372、3.099、2.679,P<0.05)。建模集中的341例学龄前儿童DCD检出31人,检出率为9.09%。儿童性别为男性(OR=2.072)、父亲文化程度高中及以下(OR=2.176)、母亲文化程度高中及以下(OR=2.088)、家庭年收入<5万元(OR=2.551)、住房面积<75m2(OR=5.138)是儿童DCD发生的独立危险因素(P<0.05)。列线图模型中影响评分的最重要因素是住房面积(99.8分),其次是家庭收入(59.2分)、父亲文化程度(48.8分)、母亲文化程度(45.5分)、性别(44.0分)。预测学龄前儿童发生DCD的内部验证(建模集)列线图模型ROC曲线下面积为0.789(95%CI:0.691~0.878),区分度较优;且Hosmer-Lemeshow拟合优度检验χ2值为4.202(P=0.838),一致性较好。列线图模型预测外部验证(验证集)学龄前儿童发生DCD的ROC曲线下面积为0.737(95%CI:0.676~0.853),区分度较优;且Hosmer-Lemeshow拟合优度检验χ2值为6.782(P=0.624),一致性较好。结论 苏州市吴江区学龄前儿童DCD检出率较高,其中儿童性别、父母亲文化程度、家庭年收入以及住房面积均对儿童DCD的发生有影响,以此构建的列线图模型具有良好的区分度与一致性,能直观预测儿童发育性协调障碍的发生风险。

关键词: 发育性协调障碍, 列线图, 预测模型, 学龄前儿童

Abstract: Objective To investigate the current status of developmental coordination disorder (DCD) among preschool children in Wujiang District, Suzhou City, to analyze its influencing factors, and to establish and validate a nomogram prediction model. Methods From September 2023 to January 2024, 341 preschool children in Wujiang District, Suzhou City, were randomly selected as the modeling set. From February to April 2024, 145 preschool children were randomly selected as the validation set. The Movement Assessment Battery for Children-2 (MABC-2) was used for screening, and the Developmental Coordination Disorder Questionnaire (DCDQ) was administered for surveys. The scores of MABC-2 dimensions and DCDQ were compared across different genders and age groups. Univariate and multivariate Logistic regression analyses were conducted to identify factors influencing DCD in preschool children. A nomogram prediction model was constructed, and receiver operating characteristic (ROC) curves and calibration curves were plotted to evaluate the discrimination and consistency of the model in internal (modeling set) and external (validation set) validation. Results Girls scored higher in manual dexterity but lower in aiming and catching compared to boys (t=2.811, 5.384, P<0.05). Girls and 6-year-old children scored higher in motor control, fine motor/writing abilities, and general coordination compared to boys (t=2.860, 5.775, 3.801) and 5-year-old children (t=4.372, 3.099, 2.679, P<0.05). Among the 341 preschool children in the modeling set, 31 cases of DCD were detected, yielding a detection rate of 9.09%. Independent risk factors for DCD included male gender (OR=2.072), father's education level of high school or below (OR=2.176), mother's education level of high school or below (OR=2.088), family annual income of <50 000 yuan (OR=2.551), and housing area of <75 m2 (OR=5.138, P<0.05). In the nomogram model, housing area was the most influential factor (99.8 points), followed by family income (59.2 points), father's education level (48.8 points), mother's education level (45.5 points), and gender (44.0 points). The area under the ROC curve (AUC) for internal validation (modeling set) was 0.789 (95%CI: 0.691 - 0.878), indicating good discrimination. The Hosmer-Lemeshow goodness-of-fit test yielded a value of 4.202 (P=0.838), indicating good consistency. For external validation (validation set), the AUC was 0.737 (95%CI: 0.676 - 0.853), with good discrimination, and the Hosmer-Lemeshow goodness-of-fit test yielded a value of 6.782 (P=0.624), indicating good consistency. Conclusions The detection rate of DCD among preschool children in Wujiang District, Suzhou City, is relatively high. Factors such as gender, parental education level, family annual income, and housing area significantly influence the occurrence of DCD. The constructed nomogram model demonstrates good discrimination and consistency, providing an intuitive tool for predicting the risk of developmental coordination disorder in children.

Key words: developmental coordination disorder, nomogram, prediction model, preschool children

中图分类号: