Digital Health in Pregnancy Care: Potential and Application of Mobile Health in Research and Medical Practice

An estimated 4.5 million pregnancy-related deaths occur globally every year. The majority of deaths happen in lower and lower-middle income countries. In low- and high-income countries alike, the majority of pregnancy-related maternal deaths are preventable. Challenges also remain in high-income countries: Pregnancy care is not attended by all pregnant women, guidelines are not based on the latest evidence, healthcare spendings are an issue, and open questions such as on medication evidence exist. Digital health technology such as mobile health applications, wearable devices, and the use of big data and machine learning may be able to assist in solving some of these challenges. Although this potential has been recognized, missing links for the use in clinical and research practice exist. In light of these challenges, this work aims to assess how digital health can be best used in the field of pregnancy care. For this purpose, I address four objectives as part of this cumulative dissertation: The first objective is the assessment of the features, scientific guidance, commercialization, and user perception of smartphone pregnancy apps. As part of a systematic analysis of applications, we identified 49 distinct functionalities, out of which 6 were previously unreported [P1]. All analyzed apps were of commercial nature. The validation of the accuracy of heart rate recording of two wrist-worn wearables is the second objective [P2]. We showed that two previously unvalidated wearable devices were able to accurately perform heart rate recordings in pregnancy-related activities. The results of both works lead to the third objective, the development of a smartphone application that unifies a digital health concept for pregnancy care, including wearable device integration and medical trial functionality [P3]. This app was subsequently evaluated regarding usability and user perception, in line with user-centered design. The fourth objective is the assessment of the potential of big data in pregnancy care. For this purpose, a large data set consisting of 1,549,186 tracked symptoms from a smartphone pregnancy app symptom tracker was analyzed [P4]. The work revealed several previously disputed or unknown symptom trajectories. It showed that data of single-time app users is comparable to those using the app for longer time periods. Furthermore, it provided a new methodological approach to the analysis of symptom reports. The contributions are accompanied by a review of the current state of the art of digital health in pregnancy care. This includes the topics of digitalization in pregnancy care, pregnancy apps, health apps, digital maternity records, ubiquitous computing and smart sensing, effectiveness, and the use of big data and machine learning. In conclusion, digital health offers a high potential in pregnancy care. This especially applies to pregnancy research. Here, digital health can collect data in a resource-efficient manner, in unprecedented quantities, detail, and from real-world conditions. In the context of health apps, future works should investigate frameworks and concepts for user engagement theory. Regarding general digital health in pregnancy care, while feasibility has been shown, more evidence regarding the effectiveness with respect to costs and the improvement of maternal-fetal outcomes (such as preterm birth) is needed. The designated aims of digital health tools (e.g. apps) should be clearly communicated and evaluated (e.g., user education or treatment effectiveness). In lower and lower-middle income countries, digital health efforts must be implemented with care to ensure that limited resources are not diverted from more effective traditional measures.