Related Experiment Videos
Prediction of small for gestational age by logistic regression in twins
Shi Wu Wen1, Hongzhuan Tan, Qiuying Yang
1School of Public Health, Central South University, Changsha, China. swwen@ohri.ca
Insights
A new prediction model for small for gestational age (SGA) in twin pregnancies uses early gestation data. This model can aid healthcare providers in managing high-risk twin pregnancies and improving perinatal outcomes.
Area of Science:
- Perinatal Medicine
- Maternal-Fetal Medicine
- Twin Gestation Research
Background:
- Small for gestational age (SGA) significantly impacts perinatal mortality and morbidity, with potential links to adult diseases.
- Early prediction of SGA is crucial for antenatal management and prevention, but current methods lack diagnostic accuracy and timely prediction intervals.
- Existing SGA prediction methods are unsatisfactory due to poor performance and short prediction-to-delivery intervals.
Purpose of the Study:
- To develop a predictive model for small for gestational age (SGA) specifically for twin pregnancies.
- To identify key variables obtainable in early gestation for SGA prediction in twins.
- To establish a reliable tool for early identification of twin pregnancies at risk for SGA.
Main Methods:
- Utilized a large US twin registry dataset (1995-1997) for model development and validation.
- Employed logistic regression analysis with twin pairs as the unit of analysis.
- Investigated two outcome measures: one or both twins SGA, and both twins SGA.
Main Results:
- A prediction model incorporating maternal race, education, marital status, parity, prenatal care initiation, and smoking, along with paternal race, was developed.
- The model achieved a sensitivity of 52.3%, specificity of 62.5%, and positive predictive value of 21.5% at a cutoff of 0.16.
- These results indicate potential for early SGA risk stratification in twin pregnancies.
Conclusions:
- A prediction model utilizing early gestation determinants can be valuable for managing twin pregnancies at high risk for SGA.
- The developed model offers a potential tool for improving perinatal care in twin gestations.
- Further research may refine this model for enhanced clinical utility in identifying SGA twins.
Background:
Small for gestational age (SGA) is one of the major determinants of perinatal mortality and morbidity, and may relate in adult diseases. Early prediction of SGA could be helpful for health care providers and public health workers in guiding antenatal management and prevention. The reported methods of SGA prediction are not satisfactory because the diagnostic performance is poor and the interval between prediction and delivery is too short.
Aims:
To establish a SGA prediction model for twin pregnancies based on variables obtainable in early gestation.
Methods:
We used a large twin registry United States data (1995-1997). The study subjects were randomly divided into two groups: group 1 to establish the prediction model by logistic regression and group 2 to validate the prediction model. SGA was defined as birth weight for gestational age z scores less than 10th percentiles. Pair of twin was the unit of analysis. Two sets of multiple logistic regression analyses with different outcome measures - one or both twins SGAs and both twins SGAs - were used to establish the prediction model.
Results:
The sensitivity, specificity, and positive predictive value were 52.3, 62.5, and 21.5%, respectively, at the cutoff value 0.16 in a SGA prediction model based on maternal race, education, marital status, parity, prenatal care visit initiation, cigarette smoking, and paternal race.
Conclusions:
A prediction model based on determinants that can be obtained at early gestation might be useful in the management of pregnancies with high risk of SGA in twins.
Related Concept Videos
01:52Regression Toward the Mean
01:11Residuals and Least-Squares Property
If the observed data point lies above the line, the residual is positive, and the line underestimates the actual data value for y. If the observed data point lies below the line, the residual is negative, and the line overestimates the actual data value for y.
The process of fitting the best-fit...
01:25Multiple Regression
Farmers can use multiple regression to determine the crop yield based on more than one factor, such as water availability, fertilizer, soil properties, etc. Here, the crop yield is the response or dependent variable as it depends on the other independent variables. The analysis requires the construction of a scatter plot...
01:18Polygenic Traits