Hans van den Heuvel

8 expect for their occupation as researchers in the obstetric department focusing on home monitoring in pregnancy. Each focus group was open for five days, and two questions were posted on Facebook daily to which all women were invited to comment. When needed, the moderator commented in response to help the discussion along. All questions that were posted in the Facebook groups can be found in Supplemental file 1. Ethical approval This study was exempted from approval of the Medical Research Ethics Committee of the University Medical Center in Utrecht (reference number 16-203), as the Committee confirmed that the Dutch Medical Research Involving Human Subjects Act (WMO) did not apply to this study. Sampling and Recruitment Two different groups of women were approached by phone at 6 weeks postpartum through purposive sampling. The first group consisted of women who had been admitted to our hospital for PPROM, FGR or preeclampsia, and who gave birth before the start of the telemonitoring pilot. The second group consisted of women with one of the same three complications, but who went home to receive telemonitoring during the pilot phase. Eligible candidates for the FGs had to be >18 years, with singleton pregnancy and good ability to understand Dutch language. Those women interested in participation received written information by mail, including an informed consent form, the schedule for the study and additional information about Facebook and privacy settings. Candidates were able to ask question about the study prior to their decision to participate. Data collection & Privacy After informed consent, we provided additional information on how to join the discussion in a private Facebook group. All comments were saved using codes for data analysis. When the research group agreed that saturation had been reached, recruitment was stopped. Afterwards all comments were manually removed by the moderator and the Facebook groups were shut down. Data analysis Each step of data analysis, using an iterative and inductive process, was performed independently by JH and CT. Questions and responses were processed manually into transcripts using open coding, assigned to text fragments. After this, the initial codes were combined as they functioned as subcategories within a broader theme. Both researchers discussed the codes and grouping together ensuring accuracy of interpretation. This resulted in four themes, with three of them divided into subcategories (See Figure 1): (1) experience with obstetric care, (2) feelings regarding pregnancy, (3) privacy and (4) impact on daily life, Representative quotes for the different themes were selected and translated into English. USER EXPERIENCES OF HOME-BASED TELEMONITORING IN PREGNANCY 129