Authors: Martin Becker, Jennifer Dai, Alan L. Chang, Dorien Fayaerts, Ina A. Stelzer, Miao Zhang, Neal G. Ravindra, Sayane Shome, Eloise Berson, Thanaphong Phongpreecha, Yuqi Tan, Lei Xue, Melan Thuraiappah, Samson Mataraso, Davide De Francesco, Yeasul Kim, Geetha Saarunya, Seyedeh Neelufar Payrovnaziri, Camilo Espinosa Bernal, Jonathan A. Mayo, Cecele C. Quaintance, Ana Laborde, Lucy King, Ivana Marić, Firdaus Dhabhar, Ian H. Gotlib, Ronald J. Wong, Martin S. Angst, Gary M. Shaw, David K. Stevenson, Brice Gaudilliere, Nima Aghaeepour
Summary: Adverse pregnancy outcomes (APOs) such as preterm birth, gestational diabetes, or preeclampsia are associated with severe short and long-term consequences for both mother and child health. However, therapeutic options to prevent APOs are currently lacking, emphasizing the critical need to find actionable targets for assessing risk and preventing development. Psychosocial and stress-related factors (PSFs), assessed by an extensive questionnaire covering, e.g., lifestyle, social support, and health concerns, are potentially modifiable factors and accessible targets for interventions that are associated with APOs. Because APOs are relatively infrequent in population-level datasets and are therefore challenging to model, multi-task machine learning is an ideal tool for exploiting their interconnectedness and building on joint combinatorial outcomes to increase predictive power. We found strong associations between PSFs and APOs, including particularly associated categories such as life stress, health stress, and perceived pregnancy risks. Additionally, PSFs were found to be related to immune system characteristics, reinforcing previously hypothesized links between immune status and stress. Elucidating the connections between stress, APOs, and immune characteristics will facilitate the implementation of individualized, integrative models of pregnancy in clinical decision making, and the modifiable nature of stressors may promote the development of accessible interventions, with success tracked through immune characteristics.
Data and code availability:
The data questionnaire (Becker et al. 2021) and the immune system data (Aghaeepour et al. 2017, Ghaemi et al. 2019) has been previously published. For reproducing the results in this paper, we make the data as well the code publicly available. It can be download here:
Stress questions (except for birth “complications” and “defects”) are from the DQAQ-SPF Questionnaire, copyright (2020) Firdaus S. Dhabhar and the University of Miami, jointly with Stanford University. All rights reserved. This questionnaire may not be used fully or partially, reproduced, displayed, modified, or distributed without the express prior written permission from Dr. Dhabhar (firstname.lastname@example.org).