Hans van den Heuvel

Table 2. Set standard threshold values for alarms in the telemonitoring platform. All thresholds are adjustable for each individual patient. Parameter Threshold Alarm color in dashboard Blood pressure Systolic >140 mmHg Orange Systolic >150 mmHg Red Diastolic >90 mmHg Orange Diastolic >100 mmHg Red Increase (jump) > 20 mmHg Red Symptom checklist 1 or more present symptoms Red Outcome measurements Patient interaction and compliance was measured by registering the number of times patients sent their blood pressure and/or the checklist. The accuracy of the automatic alert system was evaluated by manual comparison of all entered values with the system thresholds for error positive or missing alerts. Clinical impact of the alert system could be assessed through the submitted combination of BP and concurrent presence or absence of preeclampsia symptoms. The patient satisfaction and usability of the app and platform was examined one week after the end of the study period. The online survey contained 8 statements with 5 answer options (varying from strongly agree to strongly disagree), 3 questions using a 10-point Likert scale and an open comment form. RESULTS Participants In June 2017 a total of 14 pregnant women were included, after counseling of 33 women. Baseline characteristics of the study population are represented in Table 3. Ten participants used an iOS device and downloaded the app to send their measurements (71%), four used an Android device and manually entered the data in the web-based portal. During the first 4 telemonitoring days, one participant sent in multiple measurements exceeding the set threshold (BP of >140/>90) at 14 weeks gestational age, despite baseline check of in- and exclusion criteria. Themeasurements were confirmed in the outpatient clinic and she was diagnosed with chronic hypertension. Because the telemonitoring platform was able to detect this development of disease, her pregnancy was followed up (including medication) outside our study protocol. Therefore, this participants’ data on interaction and experiences was excluded, leaving 13 out of 14 participants for final data analysis. CHAPTER 4 58