*4.5. Statistical Analysis*

To evaluate the main research question, analyses of variance were performed with period (anticipation, task, post-task) as the within-subjects factor, group (PE, UP, CO) as the between-subjects factor, and the respective cardiovascular variable as the dependent variable. Scores obtained at baseline were entered as covariates. That way, changes of cardiovascular variables produced by the stress manipulation were adjusted for baseline levels, ensuring that the analyzed residual scores were due to the acute stress, and not to individual di fferences in baseline levels. A further advantage of this approach, compared to simple change scores or the inclusion of baseline in the within-subjects factor, is that it controls for measurement error inherent in the use of repeated measures of the same kind [85–87]. If necessary, Greenhouse–Geisser corrections were used to adjust for nonsphericity of the variance–covariance matrices. In addition, the described statistical analyses were done within the group with a history of preeclampsia, with mild and severe as well as early, preterm, and term preeclampsia as the between-subjects factor. For graphical representation of the di fferent time courses, standardized residualized change scores were calculated by linear regressions using the baseline period to predict the variables during the following periods, respectively [88,89]. These scores best represent those constituting the statistical results in the analyses of variance; i.e., they are adjusted for di fferences in baseline levels. One-way analyses of variance was performed to explore potential di fferences between the three groups in age, BMI, chronic stress, depression, and hemodynamic levels in resting conditions (at baseline). Further supplementary analyses were done to explore potential di fferences between groups in day of gestation, baby weight and height (independent *t*-tests), and parity (Chi-square test).

**Author Contributions:** Conceptualization, H.K.L., I.P. (Ilona Papousek), and M.G.M; Methodology, H.K.L. and I.P. (Ilona Papousek); Software, H.K.L.; Validation, I.P. (Ilona Papousek), M.L. and M.G.M.; Formal analysis, H.K.L. and I.P. (Ilona Papousek); Investigation, K.S.-Z. and I.P. (Isabella Pfniß); Resources, H.K.L. and M.G.M.; Data curation, I.P. (Isabella Pfniß), K.S.-Z. and V.K.-K.; Writing—original draft preparation, H.K.L., I.P. (Ilona Papousek), M.L., M.C.-Z., V.K.-K., O.N., and M.G.M.; Writing—review and editing, H.K.L., I.P. (Ilona Papousek), M.L., O.N., and M.G.M.; Visualization, H.K.L. and I.P. (Ilona Papousek); Supervision, M.G.M., M.C.-Z., and M.L.; Project administration, H.K.L. and K.S.Z.; Funding acquisition, H.K.L., I.P. (Ilona Papousek), and M.G.M.

**Funding:** Supported by funds of the Oesterreichische Nationalbank (Austrian Central Bank, Anniversary Fund, project number: 16426).

**Acknowledgments:** We wish to thank Anja Nischelwitzer, Kathrin Hilgarter, Jakob Riedl, Martina Rokov, Claudia Gruber, and Daniel Varga for their help in data collection. Furthermore, we wish to thank Trent Haigh for proofreading.

**Conflicts of Interest:** The authors declare no conflict of interest.
