The PREMOM I study was set up in January 2015 as a collaboration in Belgium between Hasselt University, Ziekenhuis Oost-Limburg (Genk), and 7 other hospitals (AZ Vesalius, Tongeren; Heilig Hart Ziekenhuis, Mol; JESSA, Hasselt; Maria Ziekenhuis Noord Limburg, Overpelt; Sint Franciskusziekenhuis, Heusden; S. Trudo, Sint Truiden; and Ziekenhuis Maas & Kempen, Maasmechelen). Women who participated in the PREMOM I study received obstetric surveillance through a BP monitor, an activity tracker, and a weight scale. They were asked to perform 2 BP measurements each day (morning and evening), to wear the activity tracker continuously, and to register their weight once a week in the app. This information was to be recorded until the moment of their delivery or until they were admitted to the hospital. The data collected using these devices were transferred to a web-based dashboard developed by the Mobile Health Unit (Limburg Clinical Research Center, Hasselt University; Ziekenhuis Oost-Limburg; Jessa). Predefined alarm signals were developed. Alarm signals were communicated to the responsible obstetrician so that treatment options could be agreed upon with the midwife before the pregnant woman was contacted. The therapeutic interventions were in line with local treatment procedures. The workflow is summarized in Figure 1.

In the past few years, our research team has published several articles related to the PREMOM I study that demonstrated the benefit of RM for pregnant women at elevated risk of developing PIH. Two articles made comparisons between women who received RM and women who had an increased risk of developing PIH but did not participate in the PREMOM I study (conventional care group and control group [CG]) based on a retrospective and observational design. In both studies, prenatal hospitalization to a prenatal ward (until the moment of delivery), diagnosis of preeclampsia, and number of inductions were reduced in the RM group compared with the CG. However, women in the RM group had significantly higher risks of developing gestational hypertension and a spontaneous start of the birth process compared with the CG. In the study conducted in 2015, the total number of neonatal hospitalizations in the neonatal intensive care unit was lower in the RM group than in the CG group; these findings were not confirmed in the study conducted in 2015-2016. In the later study, the total number of prenatal visits was lower in the RM group than in the CG group; this difference was not apparent in the earlier study [33-35].

The process of patient enrollment is shown in Figure 2. A total of 124 pregnant women from the PREMOM study were invited to participate in this study. Seven (5.65%) of them refused to participate because of a lack of interest. Of the remaining 117 pregnant women, 7 (5.65%) were hospitalized at the prenatal ward due to complications before they could fill out the questionnaires. In total, 110 pregnant women (response rate: 88.71%) completed the questionnaires at the start of their participation in the PREMOM project. At 32 weeks of GA, 75 of them (retention rate: 68.18%) also filled out the questionnaires.

For this study, an explorative and descriptive quantitative design was used. Pregnant women were given written information about the study when they received their BP monitor and activity tracker at the start of their RM program. The following day, all participants received an email containing a SurveyMonkey link. After logging in, the participants were asked to complete 3 questionnaires (see subsection Questionnaires) at 2 moments in time: the day following their enrollment in the RM program and at 32 weeks of GA. Demographic and obstetric characteristics of the participants were collected at the moment of enrollment in the PREMOM program and after delivery, through the hospital administration or billing records.

The Generalized Anxiety Disorder-7 scale (GAD-7) was used to detect anxiety disorders. The GAD-7 was developed within the classical theory as a brief, dimensional screening instrument that aims at identifying probable cases of anxiety disorder and also assessing symptom severity [36]. Simpson et al [37] stated that the GAD-7 represents a clinically meaningful instrument for screening in a perinatal population. The GAD-7 consists of 7 items to be rated on a balanced 4-point Likert scale ranging from 0 (“not at all”) to 3 (“almost every day”). In addition, the Patient Health Questionnaire-9 (PHQ-9) [38] was used to measure the severity of depression. The PHQ-9 was developed for use in primary care settings and has been extensively tested for validity among diverse populations [39]. Nine questions were asked to be scored on a 4-point Likert scale ranging from 0 (not at all) to 3 (almost every day). Finally, the Pain Catastrophizing Scale (PCS) is used to assess the severity of worry in the context of actual or anticipated pain. It conceptualizes catastrophizing as a multifaceted construct with 3 dimensions, each with a separately summed score: rumination (“I can’t stop thinking about how much it hurts”), magnification (“I worry that something serious may happen”), and helplessness (“There is nothing I can do to reduce the intensity of the pain”). The PCS consists of 13 items, which query the 3 subscales to be rated on a balanced 5-point Likert scale ranging from 1 (“not applicable at all”) to 5 (“fully applicable”) [40]. The scores are calculated for each subscale, and the total scores are used. This questionnaire has been adapted by the research team to refer to pregnancy-related questions. For all 3 questionnaires, higher scores indicate higher levels of feelings of depression, anxiety, and catastrophizing, respectively.

Data analysis was performed using R statistical software (version 3.2.2; R Foundation). No data imputation was performed for missing data. The study sample size was based on an a priori sample size calculation with a power of minimum 70 participants. The Shapiro-Wilk test was used to assess whether or not the data were normally distributed. Nonparametric tests were used in case where the normality assumption was violated. Skewed data are expressed as a median and IQR. Independent t tests (parametric) or Mann-Whitney U tests (nonparametric) were used for between-group comparisons. Both dimensional and categorical analyses were performed. For the PCS, catastrophizing was dichotomized as described by Olsson et al [41]. The 2 categories are labeled as “noncatastrophizing” (≤17 for the total sum) and “catastrophizing” (>17 for the total sum). A score of more than 38 is classified as “clinically relevant.” P values <.05 were considered statistically significant.

The study was approved by the Ethical Committee of the Ziekenhuis Oost-Limburg (15/093U - B371201526777) and all participants provided written informed consent. The study was also registered at ClinicalTrails.gov (NCT03246737).

In total, 75 participants completed all questionnaires at both time points. Patient mean age was 30 (28.0-32.5) years, and 28 (37.3%) of the women were pregnant for the first time. Table 1 provides detailed information about the demographics of the study participants.

The results (Table 2) showed that the levels of anxiety, depression, and several aspects of catastrophizing changed statistically over time. Pregnant women at 32 weeks of GA feel more anxiety and depression than women at the moment of enrollment (GAD-7 score=7, range 4-11 at 32 weeks of GA vs 5, range 3-8; P=.01 at the moment of enrollment; and PHQ-9 score=6, range 4-11 at 32 weeks of GA; and 4, range 2-7; P<.001 at the moment of enrollment). Only magnification was significantly lower at 32 weeks of GA compared to the moment of enrollment (3, range 1-5.5 at 32 weeks vs 4, range 3-7; P=.04 at the moment of enrollment).

In Table 3 the number of participants in each anxiety and depression category is shown. Based on the GAD-7, 10 (13.33%) pregnant women of the 75 were clinically anxious (score≥10) at the time of enrollment and 26 (34.67%) of the 75 at 32 weeks of GA. On the PHQ-9 questionnaire, 7 (9.33%) of the 75 pregnant women were clinically depressed (score≥10) at the time of enrollment and 22 (29.33%) of the 75 at 32 weeks of GA.

Based on the PCS, 25 (33.33%) pregnant women of the 75 crossed the cutoff value of more than 17 for the total sum described by Olsson et al [41] at the time of enrollment. Three (4%) of the 75 pregnant women scored more than 38 and are classified as clinically relevant. At 32 weeks of GA, 25 (33.33%) of the 75 pregnant women scored more than 17, while 4 (5.33%) of the 75 pregnant women had a higher score than 38 (Table 4).

First, in line with our expectations, scores on the GAD-7 and PHQ-9 were higher at 32 weeks of GA compared to the moment of enrollment. The scores on the subscales of the PCS were all lower at 32 weeks of GA compared with 14 weeks of GA. Regarding the primary end point (higher score for anxiety and depression at 32 weeks compared to 14 weeks), this hypothesis can be adapted. This study showed that pregnant women with PIH followed a normal pattern of increased pregnancy scores, as described by Biaggi et al [42]. The proportion of pregnant women identified by the GAD-7 with a risk for anxiety (GAD-7 score >10) was 13.3% (10/75) at enrollment and had risen to 35.6% (26/75) during 32 weeks of GA. This is higher than the reported 8.5% of pregnant women surveyed in the GAD-7 by Sutter-Dalay et al [28] and the 9.9% prevalence described by Melville et al [43]. Additionally, the scores on the PCS for pregnant women at risk of PIH are in line with those of women experiencing a normal pregnancy; for both women with and without a risk for the development of PIH, the risk for anxiety and depression is low at the onset of the pregnancy compared to the end of the pregnancy [44].

Regarding the second research aim of this study, catastrophizing in women included in this study decreased significantly for the subscale “magnification,” but was relatively stable for the subscales “rumination” and “helplessness,” and for the categorical approach. This is in line with the reported stability of catastrophizing over time during a normative pregnancy [40]. However, the study of Ollson et al [41] reported that levels of catastrophizing changed (both increased and decreased) over time in a minority of participants. So, instead of the stable PCS score for the subscale “magnification” for normal pregnancy, pregnant women with a PIH have a significantly reduced score on this subscale.

The peripartum period has long been known to be associated with increased levels of stress and anxiety, which can be exacerbated by PIH. This study shows that pregnant women at risk for PIH have higher scores on the GAD-7 and PHQ-9 at 32 weeks of GA, compared to the moment of enrollment. The subscale scores of the PCS were all lower at 32 weeks of GA compared with 14 weeks of GA. Further research is recommended to explore if the GAD-7, PHQ-9, and PCS scores of these pregnant women with PIH at the moment of enrollment and at 32 weeks of GA are significantly lower or higher compared to women with a no- or low-risk pregnancy. The knowledge that follows from this research can help implement strategies to manage pregnant women at risk for PIH with their feelings of stress and anxiety.