In normal pregnancy there is a decrease in blood pressure in early pregnancy followed by a rise in late pregnancy. Hypertensive disorders of pregnancy (HDP), defined by high blood pressure in late pregnancy (after 20 weeks’ gestation), are associated with risk of adverse health outcomes for both the mother and offspring. Using a statistical model we can describe the change in blood pressure over time during pregnancy, and, for a woman in early pregnancy, we can use this model to predict her blood pressure measurements in late pregnancy given her observed measurements.

Modelling blood pressure measurements during pregnancy may be subject to two statistical complications. First, women who give birth prematurely are more likely to have high blood pressure compared to those women who have a full term pregnancy (a statistical complication known as informative dropout). Second, the blood pressure measurements are usually measured very frequently during pregnancy, such that a woman’s blood pressure measurements may be highly correlated (a statistical complication known as serial correlation or autocorrelation). Modelling the data using a method that ignores either or both of these complications may result in incorrect conclusions about the how blood pressure changes over time during pregnancy in a given population, and affect the accuracy of any predictions.

We propose a method that can account for informative dropout and autocorrelation. Also, a useful feature of the proposed method is that it can be used to: (1) describe the association between “time to birth” and blood pressure in very early pregnancy and change in blood pressure over time. (2) Estimate the average blood pressure measurement in very early pregnancy and the average change in blood pressure over time among women with a common gestation period. And, (3) provide future predicted blood measurements and a prediction of time to birth.