高龄流产孕妇预期性悲伤变化轨迹及核心影响因素决策树分析

Longitudinal trajectories of anticipatory grief in older abortion pregnant women and decision tree analysis of core influencing factors

目的:探讨高龄流产孕妇预期性悲伤变化轨迹,分析高龄流产孕妇预期性悲伤的影响因素,并构建相关决策树模型。方法:采用便利抽样法,选取2022年1月—10月在我院门诊手术治疗的106例高龄流产孕妇为研究对象。采用一般资料调查表调查高龄流产孕妇一般资料,采用预期性悲伤量表调查高龄流产孕妇门诊术前、术后1周、术后2周及术后1个月时预期性悲伤水平。以潜变量增长混合模型分析高龄流产孕妇预期性悲伤变化轨迹,采用Logistic回归分析高龄流产孕妇预期性悲伤的影响因素,构建高龄流产孕妇预期性悲伤的决策树模型。结果:识别出2条高龄流产孕妇预期性悲伤变化轨迹,其中下降缓慢组高龄流产孕妇(51例)预期性悲伤评分相对较高,且术后1周、术后2周及术后1个月各时段预期性悲伤评分下降缓慢;下降快速组高龄流产孕妇(55例)预期性悲伤评分开始处于中等水平,但术后1周、术后2周及术后1个月各时段预期性悲伤评分下降较明显。Logistic回归分析结果显示,文化程度、居住地、家庭人均月收入、主要照顾者及术后并发症是高龄流产孕妇预期性悲伤的影响因素(P<0.05)。构建的高龄流产孕妇预期性悲伤决策树模型选择了文化程度、主要照顾者、居住地、家庭人均月收入及术后并发症5个临床特征作为模型节点,决策树模型和Logistic回归模型的受试者工作特征曲线下面积均为0.838[95%CI(0.756,0.919)]。结论:高龄流产孕妇预期性悲伤呈下降缓慢和下降快速2种趋势,文化程度、居住地、家庭人均月收入、主要照顾者及术后并发症是其影响因素。本研究构建的决策树模型能够较为准确地预测高龄流产孕妇预期性悲伤。

Objective:To explore longitudinal trajectories of anticipatory grief in elderly abortion pregnant women,analyze influencing factors of anticipatory grief in elderly abortion pregnant women,and construct related decision tree model.Methods:A total of 106 elderly abortion pregnant women who were treated in outpatient department of our hospital from January to October 2022 were selected as the research subjects by convenience sampling method.A general information survey scale was used to investigate general information of old abortion pregnant wamen,and premonitory sadness scale was used to investigate anticipatory grief level of elderly abortion pregnant women at preoperative,1 week after surgery,2 weeks after surgery and 1 month after surgery.The longitudinal trajectories of anticipatory grief in elderly abortion pregnant women was analyzed by latent growth mixture modeling.Influencing factors of anticipatory grief in elderly abortion pregnant women was analysed by Logistic regression.Construct related decision tree model.Decision tree model of anticipatory grief in elderly abortion pregnant women was constructed.Results:Two trajectories of anticipatory grief in elderly abortion pregnant women were identified.Among them,the slow decline group(51 cases)had relatively high score of anticipatory grief in elderly abortion pregnant women,and the decrease in score of anticipatory grief was slow at 1 week after surgery,2 weeks after surgery and 1 month after surgery.The score of anticipatory grief in elderly abortion pregnant women in fast decline group(55 cases)started at a moderate level,but there was a significant decrease in score of anticipatory grief at 1 week after surgery,2 weeks after surgery and 1 month after surgery.Logistic regression analysis showed that education level,place of residence,per capita monthly income of the family,primary caregivers,and postoperative complications were influencing factors of anticipatory grief in elderly abortion pregnant women(P<0.05).The decision tree model of anticipatory grief in elderly abortion pregnant women selected five clinical characteristics as the model nodes,included education level,primary caregivers,place of residence,per capita monthly income of the family and postoperative complications.Area under curve of receiver operator characteristic both decision tree model and Logistic regression model was 0.838(95%CI 0.756⁃0.919).Conclusions:The anticipatory grief in elderly abortion pregnant women shows two trends with slow decline and fast decline.Education level,place of residence,per capita monthly income of the family,primary caregivers,and postoperative complications were influencing factors of anticipatory grief in elderly abortion pregnant women.The decision tree model constructed in this study can more accurately predict anticipatory grief in elderly abortion pregnant women.