Constructing a nomogram prediction model for treatment compliance in infertile patients with mycoplasma infection
10.3760/cma.j.cn101441-20241015-00375
- VernacularTitle:构建支原体感染的不孕患者治疗依从性的列线图预测模型
- Author:
Hong LEI
1
;
Jingyi ZHANG
;
Yun HUANG
;
Jian HUANG
Author Information
1. 江西省赣州市妇幼保健院检验科，赣州 341000
- Publication Type:Journal Article
- Keywords:
Mycoplasma infection;
Infertility;
Compliance;
Nomogram prediction model
- From:
Chinese Journal of Reproduction and Contraception
2025;45(7):709-714
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To analyze the relevant factors affecting treatment compliance in infertile patients with mycoplasma infection, construct a nomogram prediction model and validate it.Methods:In this case-control study, we retrospectively analyzed the clinical data of 400 infertile patients who attended the Department of Reproduction of Ganzhou Maternal and Child Health Hospital, Jiangxi Province from April 2022 to March 2024. They were divided into modeling group ( n=280) and validation group ( n=120) according to the 7∶3 principle with random number method, and further divided into complete compliance subgroup and unstable compliance subgroup based on different levels of compliance. Multiple logistic regression model was used to analyze the risk factors affecting treatment adherence in infertile patients with mycoplasma infection, variance inflation factor was used to assess the collinearity of the independent variables and to construct a columnar plot prediction model, and the validation and predictive efficacy of the columnar plot model were assessed by calibration curves and decision curves, respectively. The validation and predictive efficacy of the column chart model were evaluated by calibration curve and decision curve respectively. Results:In the modeling group, there was no significant statistical difference between the fully compliant group and the unstable compliant group in terms of body mass index, place of residence, occupation, settlement form, and attitude of medical staff (all P>0.05). Age in the unstable adherence subgroup [(33.22±3.20) years], proportion of junior high school and below in education [36.81% (67/182)], proportion of average monthly disposable amount ≤2 000 yuan [36.81% (67/182)], proportion of poor disease awareness [62.09% (113/182)], and infertility related stress scale (IRSS) scores (45.60±17.14) were higher than those in the full adherence subgroup [(31.36±3.24) years, P<0.001; 19.39% (19/98), P=0.003; 16.33% (16/98), P<0.001; 28.57% (28/98), P<0.001; 37.64±16.69, P<0.001]. The receiver operating characteristc curve analysis results showed that the optimal cutoff values for age and IRSS score were 32 years old and 42 points, respectively. The results of logistic multiple regression model showed that: age ( OR=1.453, 95% CI: 1.260-1.794, P=0.002), education level ( OR=1.386, 95% CI: 1.185-1.564, P<0.001), average monthly disposable amount ( OR=1.081, 95% CI: 1.002-1.246, P=0.023), disease awareness level ( OR=1.827, 95% CI: 1.390-2.359, P<0.001), and average monthly disposable amount ( OR=1.081, 95% CI: 1.002-1.246, P=0.023), disease awareness ( OR=1.827, 95% CI:1.390-2.359, P<0.001), and IRSS score ( OR=1.590, 95% CI: 1.255-1.902, P=0.006) were independent risk factors affecting treatment adherence in infertile patients with mycoplasma infection. The modeling group C-index was 0.872 (0.816-0.927), the intercept (-0.004) and slope (0.995) of the calibration curve were close to the desirable values of 0 and 1, the Brier score was 0.017, and thresholds between 0.37 and 0.96 provided a net clinical benefit. The C-index of the test model set was 0.816 (0.736-0.896), the intercept (-0.007) and slope (1.015) of the calibration curve were close to the ideal values of 0 and 1, the Brier score was 0.013, and the threshold value provided a net clinical benefit when the threshold value ranged from 0.21 to 0.98, and the H-L goodness-of-fit test showed that χ 2=7.36, P=0.499, and the constructed predictive model for the column line graphs had good discrimination and consistency. Conclusion:Age, educational level, monthly average disposable income, disease awareness, and IRSS score are the relevant factors affecting treatment compliance in infertile patients with mycoplasma infection. The nomogram prediction model constructed in this study based on those relevant factors has a good predictive value for treatment compliance in infertile patients with mycoplasma infection, and can provide certain reference for clinical intervention measures.
