Hans van den Heuvel

6 Strengths and limitations Strength of this study is the direct data extraction of healthcare consumption from electronic patient files of our study center. These data allowed a factual, real life comparison of direct healthcare costs between groups. For the costs associated with use of the digital health platform, a precise calculation was made based on acquisition of blood pressure monitors, subscription fees of the dashboard and time spent by obstetric care professionals in the telemonitoring team for daily monitoring of abnormal values. Addition of these costs to the direct healthcare costs of the telemonitoring participants allowed a complete interpretation of differences between groups. We compared groups of women with risk factors for (development of) preeclampsia at intake, and therefore we were able to compute costs of full antenatal care up to delivery. The results of the cost analysis therefore reflect clinical practice of antenatal follow-up in this risk group, which aids the applicability and generalizability for similar health care settings. Finally, a major strength of this study is the addition of societal costs of travelling and productivity loss due to complete follow-up to the analysis. These results extend the overview of costs associated with care for pregnant women at risk of hypertensive complications. There are several limitations to the study. The retrospective nature of the control groupmight have caused substantial selection bias. The two groups were similar regarding to all baseline characteristics, however more women with pre-existent kidney disease, and less women with a history of preeclampsia were included in the SAFE@HOME group. Furthermore, women with arm circumference over 42 were excluded for participation in this group. These differences may have influenced the results of pregnancy outcomes and therefore healthcare consumption and costs. In general, cost analyses are made with calculations based on assumptions of costs for Dutch healthcare, which could hamper extrapolation of results to other countries or settings. Our analysis was restricted to antenatal care costs. As there were no significant differences in perinatal outcomes between groups, especially with regards to mode of delivery and neonatal outcome, it is legitimate to provide an overview of costs of antenatal care only. 17 Interpretation and further research The use of mobile-health technology to assist antenatal care has been suggested before for its advantages regarding access to care, enhanced satisfaction and reduction of health care consumption. 4,20,21 However, to implement eHealth-enhanced strategies in antenatal care, a deliberate approach is needed before widespread implementation. Alongside evaluation of effects of telemonitoring onperinatal outcome and patient satisfaction, economic evaluations are needed to determine the added value of digital health strategies. Cost analysis as performed in this study is of interest to both health care providers, pregnant women and other stakeholders in the process of decision-making in the future of healthcare. For decisions on funding and adoption, governments on national, European and global levels SAFE@HOME: COST ANALYSIS 95