Predictive factors for live birth with IVF treatment: A retrospective study of 8450 treatments with single fresh embryo transfers
Independent thesis Advanced level (degree of Master (One Year)), 10 credits / 15 HE creditsStudent thesis
Background: Infertility is a relatively common problem among people worldwide. It is estimated that 1 in every 6-7 couples is facing infertility. If the couple tries to conceive for over a year without success then they have infertility problems. The most common treatment option for infertility is in-vitro fertilization (IVF), i.e. when the eggs are fertilized outside of the human body.
Objective: The purpose of the current study is to construct a prediction model for IVF success with single fresh embryo transfers. The outcome variable is live birth.
Material and methods: The data are collected at Carl von Linné Clinic from 1999 to 2014. A total of 8450 treatments from about 5700 patients are included in the analysis. The statistical method used for the analysis is Generalized Estimating Equations, as this method accounts for dependency in the data.
Results: In the model found in this study, eight predictors significantly affect the outcome of IVF treatment: embryo score, success rates after previous IVF treatments, ovarian response, female age, infertility cause, the thickness of the endometrium, presence of hydrosalpinx or fibroid and female height. Many of these factors confirm previous findings. Female height, however, has not been shown to be predictive for IVF success previously.
The model's performance is assessed by validation. The area under the ROC curve indicates moderate discriminative power (C-statistic = 0.68). The calibration, assessed by the HosmerLemeshow test, indicates very good correspondence between predicted and observed live birth rates.
Place, publisher, year, edition, pages
2015. , 39 p.
Infertility, IVF, prediction model, embryo, calibration
Probability Theory and Statistics
IdentifiersURN: urn:nbn:se:uu:diva-256777OAI: oai:DiVA.org:uu-256777DiVA: diva2:826979
Carl von Linné Clinic, Uppsala Clinical Research Center
Subject / course
Master Programme in Statistics