Nurse-guided Mobile Health Care Program to Reduce Emotional Distress Experienced by Parents of Infants Prenatally Diagnosed with Critical Congenital Heart Disease: A Pilot Study

Abstract

Background:

Following prenatal diagnosis of critical congenital heart disease (CCHD), parents encounter emotional distress while facing caregiving challenges. Supportive psycho-educational interventions using mobile health (mHealth) can make care more accessible.

Objectives:

We tested a novel nurse-guided mHealth care program, Preparing Heart and Mind^™ (PHM^™), with the objectives of examining feasibility and estimating the effect of the intervention on parents’ emotional distress.

Methods:

This pilot study design randomized participants using a 2:1 intervention to control ratio. Analysis involved description of retention, and intervention attendance and engagement, and adjusted linear mixed models to estimate group differences in depressive (CES-D), anxiety (STAI-S), and traumatic stress (IES-r) symptoms.

Results:

The sample included 55 parents (n=38 PHM^™ group, n=17 control). Complete retention of 37 (67%) parents included 29 (76%) in the PHM^™ group and 8 (47%) control. Most attrition was due to infant death (7 parents), transplant referral (2 parents), or postnatal diagnostic ineligibility (4 parents). For the PHM^™ group, ≥96% of parents attended pre- and postnatal sessions and most (65%) messaged with the nurse. mHealth engagement was highest prenatally, with handling uncertainty the most viewed topic (average 94% pages viewed). In linear mixed models analyses, the PHM^™ group had on average 4.84 points lower depression (95% CI: −10.68–1.04), 6.56 points lower anxiety (−14.04–0.92), and 6.28 points lower trauma (−14.44–1.88) scores by study end.

Conclusion:

Findings suggest that a nurse-guided mHealth approach is feasible and may contribute to a clinically important reduction in parents’ emotional distress.

1. Introduction

Congenital heart disease is the most common major congenital anomaly worldwide, and approximately 40,000 infants are born with congenital heart disease each year in the United States [1,2]. Among these, 25% are diagnosed with critical congenital heart disease (CCHD), which requires surgical intervention soon after birth followed by ongoing medical management [1]. Recent population-based studies have shown that survival rates associated with congenital heart disease have greatly improved for infants born in developed countries [3–6]. For infants with CCHD, however, problems with growth and development remain [7–9].

CCHD is most often prenatally diagnosed during a routine ultrasound scan between 18–22 weeks of pregnancy, and current efforts are focused on increasing detection rates in community practices [10–13]. Following an early diagnosis, most parents proceed with pregnancy and experience prolonged uncertainty [14–17]. Heightened risk for emotional distress that is clinically concerning and often enduring has been recognized for these parents [18,19], with some struggling with fear and having difficulty finding hope in the face of uncertain outcomes [15]. To be prepared to care for a medically-vulnerable child, parents have described seeking out information from their healthcare providers about the diagnosis and about what to expect, valuing vetted online resources before the birth [20]. Although recent practice recommendations have underscored the necessity of providing timely, sensitively-tailored patient counseling, parents’ needs for emotional support and structured diagnostic information are not consistently met [21–23]. Additionally, there is evidence showing that gaps in prenatal care coordination related to a lack of communication among healthcare providers was associated with racial and economic disparities and poorer outcomes for infants with CCHD [24]. These communication-related gaps might also exacerbate inconsistencies in prenatal counseling and in addressing parents’ needs more broadly.

Parents of infants with CCHD continue to face numerous challenges postnatally. Challenges related to changes in providers across the care continuum, communication with healthcare providers and with family members, and the transition to parenting an infant with a life-threatening condition have been well described [16,25–30]. Parents often experience emotional distress at the time of infant hospitalization and/or cardiac surgery, with symptoms of depression and anxiety linked to parents’ perceived inability to protect or comfort their hospitalized infants [31]. Parents’ accounts have also underscored the psychological toll of early caregiving, often including infant feeding challenges and efforts to achieve crucial growth and developmental goals [32,33]. Left unaddressed, emotional distress could impede parents’ involvement in infant healthcare, agency in the caregiving role, and ability to address feeding challenges.

Mental health-related interventions to support parents of infants with CCHD have typically been delivered in person during the postnatal time [34–38], with a review of these intervention studies demonstrating high risk for bias [39]. Leveraging technology to expand access and reduce costs for healthcare has become common [40,41]; however, validated telehealth and mobile health (mHealth) interventions to support parents of infants with CCHD are scarce. One telehealth intervention intended to monitor families following infant cardiac surgery was trialed, with no significant improvements in parent mental health or infant growth outcomes [42]. More recently, the co-design process for an mHealth app for parents with a prenatal CCHD diagnosis was reported [43]. The feasibility and acceptability of the app-based telehealth component of this design was supported [44]. While there is potential in using technology to improve care [45], a need remains for evidence-based, theoretically grounded interventions that can support parents’ caregiving and infant outcomes.

Our work has involved a multi-staged approach underpinned by guided participation theory [46,47] to develop the Preparing Heart and Mind^™ (PHM^™) care program, a psycho-educational intervention designed to reduce emotional distress and support caregiving for parents with a fetal CCHD diagnosis. In this pilot randomized controlled trial, we aimed to: (1) evaluate feasibility of the study design through participant retention and of the intervention through attendance/engagement, and (2) examine differences in parents’ symptoms of emotional distress between the PHM^™ and control groups over time.

2. Methods

2.1. Study design

The pilot study was a randomized control trial with a mixed methods design and multiple measurement time points conducted from October 2020 - February 2023 (Clinicaltrials.gov: NCT05282368). This report is focused on analysis of quantitative data on feasibility and emotional distress outcomes.

2.2. Study sample

The study sample was recruited from maternal-fetal medicine clinics affiliated with two quaternary care centers located in the upper midwestern and southern regions of the United States. Potential participants were identified as eligible, and invited to participate in the study in the clinic setting or via MyChart (i.e., secure online healthcare portal). Potentially eligible participants could also self-refer using an online screening survey found on informational hand-outs in the clinic or through two university-based websites. The study nurse then contacted those interested, verified eligibility, discussed the study, and answered questions.

Eligibility criteria for study enrollment included being pregnant and between 24–34 weeks of pregnancy or a caregiving partner; ≥18 years old; able to read, write, and speak English; and having access and ability to use a smartphone, tablet, or computer in a private location. Participation of caregiving partners identified by the pregnant person as someone who would be involved in pregnancy, birth, and infant care (e.g., spouse, partner, or close family member) was encouraged, but not required. A caregiving partner could not participate without the pregnant partner. Eligibility also included having a primary fetal diagnosis of a major structural anomaly of the heart (i.e., CCHD) requiring surgical and/or medical intervention within the first year of life. Other anomalies or genetic and/or chromosomal conditions could be present. Exclusion criteria included pregnancy termination or a diagnosis highly likely to result in fetal or infant demise shortly after birth. Institutional Review Board approvals were obtained for the conduct of this study (XXX#:00005489; XXX#:H-46440), and participants provided written informed consent to participate.

2.3. Procedures

Following consent, participants were randomized to the intervention or control group with a random digit table created by a research team member who was not involved in any other study procedures. Pregnant person/caregiving partner dyads were randomized to the same group, if applicable. A 2:1 allocation ratio was used for assigning participants to intervention and control groups. This decision represents a compromise between maximizing precision for descriptive analyses of the PHM^™ intervention group, and maximizing power for exploratory analysis to estimate differences between the intervention and control groups. Figure 1 shows a flow diagram that specifies the recruitment, enrollment, and allocation process.

Figure 1.

Flow diagram

Although consent could have been completed as early as 24 weeks of pregnancy, data collection began between 27 – 34 weeks of pregnancy. PHM^™ group participants were assigned one of two nurse guides who were research team members and specially trained to conduct the intervention. Another research team member was dedicated to corresponding only with control group participants using scripted messages regarding progression through the study primarily focused on data collection. Figure 2 illustrates participants’ study experiences.

Figure 2.

Participant experience

2.4. Nurse-guided mHealth care program intervention

This nurse-guided mHealth care program has been previously described [48]. The PHM^™ care program was designed on a secure, cloud-based GetWell patient engagement web and app platform. The platform provided a parent-facing app and a nurse-monitored interactive clinician dashboard. Intervention sessions took place remotely through video conferencing (i.e., Zoom), often during evenings and weekends. The same nurse guide met with a pregnant/birthing person or dyad before and after infant birth. No sessions were conducted with the caregiving partner alone. During the first session, parents were provided access to the PHM^™ care program, with each pregnant person or caregiving partner provided their own, individual log-in for continuous access. The nurse guide used semi-structured session protocols to invite discussion about parents’ concerns, facilitate a nursing assessment, and provide tailored support for caregiving incorporating the PHM^™ app (e.g., mental health/wellbeing, condition-specific information, dyad relationship, social support) [48]. The PHM^™ app included 15 care program topics (e.g., Handling Uncertainty, Preparing for Hospitalization, Feeding Your Baby), and was organized consistently to include two key sections of Helpful to Know and Parent Stories. Helpful to Know provided evidence-based information, videos, and links to resources. Parent Stories included varied psychosocial issues using written vignettes and interactive response options to questions about an issue as well as strengths or areas of concern represented in the vignette. Also available were 17 condition-specific topics (e.g., Hypoplastic Left Heart Syndrome), offering a brief overview of anatomy and physiology with links to more comprehensive information that could be assigned to tailor the care program (for further description, see [46]). Nine prenatal topics and relevant condition-specific topics were made available to participants during the first session with the nurse guide. By design, six postnatal topics were opened proximate to infant birth. The PHM^™ app also sent regularly scheduled check-ins prompting participants’ responses to determine support needs, and offered app-based SMS text messaging for nurse-participant communication. Pregnant persons and caregiving partners engaged in check-ins and nurse messaging on an individual basis, not as dyads.

2.5. Measures

Participants completed up to five surveys online (See Fig. 2; 3 prenatal, 2 postnatal). Participants completed surveys with an assigned code used for de-identification and linking survey data across time points. Each survey was administered a minimum of 2 weeks apart, and within a week before an intervention session.

2.5.1. Participant characteristics

Demographics and health information

The first survey included items to collect demographic and health information. Several demographic variables were transformed due to response distribution. Response options for race included (select all) “American Indian/Alaskan Native/First Nations”, “Asian”, “Black”, “Middle Eastern”, “Native Hawaiian/Pacific Islander”, “White”, and “Other”; these were collapsed to “Asian, Black, or Indigenous” and “White.” Response options for highest level of education achieved were collapsed into “high school/vocational school”, “associate’s degree”, “bachelor’s degree”, and “graduate degree”. Response options for annual household income were collapsed to “$30,000–$49,999”, “$50,000–$89,999”; “$90,000–$149,999”, and “>$150,000”. The infant’s cardiac diagnosis was self-reported by the parent and verified by the pediatric cardiologist. These diagnoses were collapsed to single ventricle and biventricular physiology.

The Severity of Medical Illness (SMI) is an index to rate the complexity of infant health status after birth [49]. Ratings were completed by two pediatric cardiologists (XXX, XXX). Using an infant’s first echocardiogram, the baseline index rating is determined on a continuum from 1 (insignificant disorder) to 5 (severe disorder requiring complex palliation or transplant). Modifier points accounting for other relevant conditions (e.g., chromosomal abnormalities, extracardiac anomalies) can extend the baseline to an upper limit of a total index rating of 11, representing highest severity.

2.5.2. Feasibility

For the feasibility of the study design, data collection was used as an indicator of participant retention. In regard to feasibility of the intervention, we considered session attendance and engagement in the PHM^™ care program app for those in the intervention group.

2.5.3. Emotional distress outcomes

The Center for Epidemiological Studies-Depression (CES-D) is a 20-item self-report scale designed to measure depressive symptoms in the general population [50]. Response options correspond to symptom frequency within the past week on a 4-point scale, ranging from 0 (Rarely or None of the Time/<1 Day) to 3 (Most or All of the Time/5–7 Days). Scores range from 0 to 60, with scores ≥16 indicating clinical significance. An α = 0.92 supported its use for this sample.

The State-Trait Anxiety Inventory State subscale (STAI-S) is a 20-item self-report measure of anxiety related to a specific situation (e.g., fetal/infant health condition) [51,52]. Response options are on a 4-point scale with 1 (not at all) to 4 (very much so). Scores range from 20 to 80, with scores ≥40 considered clinically concerning. An α = 0.96 supported the use of this subscale for this sample.

The Impact of Event-revised (IES-r) is a 22-item self-report measure that elicits cognitive and behavioral responses regarding a traumatic life event (e.g., fetal/infant health condition), organized by three domains: avoidance, hyperarousal, intrusion [53,54]. Response options are on a 5-point scale from 0 (Not at all) to 4 (Extremely). Scores range from 0–88. A total score indicates a continuum of risk with cut-off scores of clinical risk (24–32) and high risk (≥33) for post-traumatic stress. A global α = 0.92 supported its use for this sample.

2.6. Data analysis

All analyses were conducted using R (version 4.2.2) [55]. Participant characteristics were summarized using n (%) or median (25^th%–75^th%). We calculated standardized mean differences (SMD) between study groups for baseline characteristics, with an absolute SMD ≥ 0.10 considered indicative of potential imbalances [56].

2.6.1. Feasibility

To evaluate feasibility of the study design, we used descriptive statistics to examine retention of study participants. Retention was also examined in the context of condition severity, including SMI ratings and single ventricle physiology. For participants in the PHM^™ group, we evaluated feasibility of the intervention by determining the proportion of intervention session attendance and reasons for missed sessions. For engagement, we used the clinician dashboard to quantify the proportion of participant-initiated views of care program content and their app-based messaging with the nurse.

2.6.2. Primary outcomes

We compared the primary outcomes of emotional distress with depression, anxiety, and trauma symptoms between study groups. Initial data exploration included visual examination of individual trajectories and linear trends. We also visually explored potential subgroup differences in intervention effect based on fetal cardiac diagnosis (i.e., biventricular versus single ventricle).

We conducted an exploratory estimation of the effect of the intervention on the primary study outcomes by fitting linear mixed models using the lme4 and lmerTest packages with an intention-to-treat approach [57,58]. Linear mixed models for repeated measures allow for missing data over study time points [59], which is advantageous when studies are conducted during the unpredictable context of pregnancy with a critically ill infant. All available data were added to the models, including data from participants who were withdrawn from the study or were lost to follow up. Models included a fixed effect for the interaction between study group and time, with additional covariates for adjustment selected to balance differences between the intervention and control groups (e.g., participant education level) and/or for potential to impact the outcomes of interest (i.e., study site, infant cardiac diagnosis). We explored models including subject-specific random intercepts, models with random slopes for subject-specific trajectories over time, and models including random slopes for both individual subjects and for clusters of birthing persons/caregiving partners. We compared models using likelihood ratio tests, with p<0.005 considered an indication of model improvement. Models with random slopes for participant trajectories, but not for birthing person/caregiving partner clusters, demonstrated the best fit. Adding a fixed effect to adjust for birthing person (y/n) did not improve the models or change the results. Model diagnostics included examination of normality of residuals via QQ plots, normality of the random effects, residuals vs. predicted values, and potential for influential points via Cook’s distance.

Considering potential differences between the prenatal and postnatal times (e.g., continuous access to healthcare teams while inpatient after birth; varied infant clinical course) and lower postnatal app usage, we fitted additional models that were limited to the prenatal time points. We created linear mixed models with a random subject-specific intercept and the same fixed effects, with likelihood ratio tests again used to explore the addition of random slopes for subjects and birthing person/caregiving partner clusters. For the prenatal time, models with a subject-specific random intercept demonstrated the best fit. The type I error probability for all model analyses was set at 0.05.

3. Results

3.1. Participants

A total of 55 participants were parents randomized to the intervention (n=38) or control (n=17) groups, including 32 (58%) pregnant/birthing persons who identified as mothers and 22 (40%) caregiving partners who identified as either a father or parent partner. Sample characteristics by group can be seen in Table 1. These parents reported a median age of 33 years old, and 13 (24%) self-identified as Asian, Black, or Indigenous. The majority (n=31, 56%) of parents had a previous child. For 19 (35%) parents, the primary fetal cardiac diagnosis involved single ventricle physiology, with a higher prevalence of single ventricle diagnoses in the control group (47% vs. 29%, SMD=0.38). Of the 33 infants born to these parents, 9 (27%) were female. The SMI indicated marked to severe ratings, with PHM^™ group infants indexed with a mean 4.64 (SD=1.48) and control group infants indexed with a mean of 5.06 (SD=1.57), and an SMD=0.27.

Table 1.

Sample characteristics at baseline (N=55)

	PHM group (n = 38) n (%) or median (25th, 75th)	Control group (n = 17) n (%) or median (25th, 75th)	SMD
Pregnant person			0.02
Yes	22 (58)	10 (59)
No	16 (42)	7 (41)
Age (years)	33.0 (32.0, 36.0)	33.0 (29.0, 39.0)	0.09
Race			0.44
Asian, Black, or Indigenous.	11 (29)	2 (12)
White	27 (71)	15 (88)
Hispanic/Latino/a ethnicity			0.08
Yes	3 (8)	1 (6)
No	35 (92)	16 (94)
Education			0.51
HS/Votech	4 (11)	3 (18)
Associate’s degree	7 (18)	1 (6)
Bachelor’s degree	22 (58)	12 (71)
Grad degree	5 (13)	S 1 (6)
Income			0.36
$30,000-$49,999	2 (5)	0 (0)
$50,000-$89,999	6 (16)	2 (12)
$90,000-$149,999	14 (37)	7 (41)
>$150,000	16 (42)	8 (47)
Insurance type			0.12
Commercial	32 (84)	15 (88)
Government aid	6 (16)	2 (12)
Relationship status			0.67
Married	28 (74)	15 (88)
Partnered	10 (26)	1 (6)
Single (Not Partnered)	0 (0)	1 (6)
First baby			0.27
Yes	15 (39)	9 (53)
No	23 (61)	8 (47)
Fetal cardiac diagnosis			0.38
BiV J	27 (71)	9 (53)
SV	11 (29)	8 (47)
Site			0.05
1	26 (68)	12 (71)
2	12 (32)	5 (29)
CES-D score	9 (5, 14)	14 (10, 23)	−0.60
STAI-S score	38 (33, 48)	45 (37, 51)	−0.40
IES-r score	14 (8, 22)	19 (13, 33)	−0.29
IES-r intrusion score	7.0 (4.0, 10.0)	10.0 (5.0, 15.0)	−0.31
IES-r avoidance score	5.5 (1.3, 9.0)	7.0 (5.0, 10.0)	−0.29
IES-r hyperarousal score	3.0 (0.0, 5.0)	4.0 (2.0, 5.0)	−0.20

Abbreviations: BiV = biventricular; CES-D = Center for Epidemiologic Studies Depression Scale; HS = high school; IES-r = Impact of Event Scale – Revised; SMD = standardized mean difference; STAI-S = State-Trait Anxiety Inventory – State subscale; SV = single ventricle; Votech = Vocational or technical school

3.2. Feasibility

Regarding feasibility in terms of study design, a total of 37 (67%) parents were retained and completed data collection at all study time points, including 29 (76%) in the PHM^™ group and 8 (47%) in the control group. Following randomization, 4 of 5 parents who changed their minds were fathers, including only one who withdrew from the PHM^™ group. A total of 7 parents were withdrawn from the study due to infant death (n=3 infants) or referral for transplant (n=1 infant). While the control group included a higher proportion of participants experiencing infant death or transplant (35% vs. 5%, SMD=0.81), SMI ratings reflected a small difference in severity of infant illness. An additional 4 parents were withdrawn due to diagnostic ineligibility verified after birth (n=3 infants). Completion of the study after the first postnatal survey was disrupted by mental health-related needs for one dyad and lost to follow up for one other parent.

Considering feasibility of the intervention, out of 38 parents in the intervention group, 6 parents (2 dyads, 2 pregnant persons) attended at least one prenatal session, but became ineligible before the postnatal session because of death, transplant, or diagnosis change (as described above). Of the remaining 32 parents (14 dyads; 4 pregnant/birthing persons), 31 (97%) attended both pre- and postnatal sessions with the nurse. A total of 4 participants’ (2 dyads) infants were born earlier than expected; these parents could only attend 1 prenatal session. For the remaining 28 parents (12 dyads and 4 pregnant/birthing persons) 27 (96%) attended all 3 sessions with the nurse. Each session lasted between 47–63 minutes.

The PHM^™ care program app was initiated by all parents (n=37) who attended the first session. Parents tended to engage more frequently with prenatal content (i.e., topics 1–9), with the introductory topic (99%), the topic about handling uncertainty (94%) and the topic about orienting (81%) viewed fully by most parents. Postnatal topics of working through emotions after birth (20%), handling uncertainty after birth (17%), and leaving the hospital (10%) were least commonly viewed (See Supplementary Fig. 1 for average proportion of each topic viewed).

A total of 24 (65%) parents in the PHM^™ group engaged in messaging with the nurse by sending any message. Of these parents, 8 (33%) initiated a message. Pregnant/birthing persons were more likely to use the messaging feature, with 16 (73%) sending any message, and 6 (27%) initiating a message. Eight (53%) fathers sent any message, and 2 (13%) initiated a message with the nurse.

3.3. Primary outcomes

3.3.1. Emotional distress

As can be seen in Table 1, the control group reported higher levels of depression, anxiety, and trauma symptoms at baseline. The prevalence of CES-D scores ≥16, indicating risk for clinical depression, was 31% overall (intervention, n=9, 24%; control, n=8, 47%); the prevalence of STAI-S scores >40, indicating probable clinical levels of anxiety, was 53% overall (intervention, n=18, 47%; control, n=11, 65%); and the prevalence of IES-r scores >32, indicating probable diagnosis of PTSD, was 16% (intervention, n=4, 11%; control; n=5, 29%).

3.3.2. Intervention effect

Figure 3 visualizes individual trajectories of parent-reported symptoms of depression, anxiety, and trauma over the five study time points, along with linear trends for each outcome, compared by group. The control group exhibited greater variability and higher symptom levels at all time points. For both groups, symptoms appeared to decrease postnatally.

Figure 3.

Data visualizations of individual trajectories and linear trends for outcomes compared by group across study time points

The results of the linear mixed model analysis can be found in Table 2. Considering each participant’s baseline score and individual trajectory and adjusting for study site, participant education level, and infant diagnosis, we estimate that the PHM^™ intervention resulted, on average, in 4.84 point lower depressive score by the end of the study time points (i.e., 1.21 lower CES-D score at each time point; 95% CI: 2.67 lower–0.26 higher). Similarly, the PHM^™ group reported 6.56 points lower anxiety score on average (i.e., 1.64 lower STAI-S score per time point; −3.51–0.23) and a 6.28 average lower trauma score (i.e., 1.57 lower IES-r score per time point; −3.61–0.47). The results of the subgroup analysis only including prenatal time points were similar in magnitude and direction (see Table 3) with a slightly more pronounced effect on symptoms of anxiety.

Table 2.

Linear mixed models to estimate the effect of being in the intervention group over time on symptoms of depression, anxiety and trauma

	β	SE	95% CI	p value
Model 1.a CES-D score
Intervention group:Time point	−1.21	0.75	(−2.67–0.26)	0.107
Model 2. STAI-S score
Intervention group:Time point	−1.64	0.96	(−3.51–0.23)	0.087
Model 3. IES-r score
Intervention group:Time point	−1.57	1.04	(−3.61–0.47)	0.132

^aAll models adjusted for fixed effects including study site, participant education level, and infant cardiac diagnosis (biventricular or single ventricle), with a random slope for participant-specific trajectories Abbrevations: CES-D = Center for Epidemiologic Studies Depression Scale; CI = confidence interval; IES-r = Impact of Event Scale – Revised; SE = standard error; STAI-S = State-Trait Anxiety Inventory – State subscale

Table 3.

Linear mixed models to estimate the effect of being in the intervention group on symptoms of depression, anxiety and trauma during the prenatal time

	β	SE	95% CI	p value
Model 1.a CES-D score
Intervention group:Time point	−1.21	1.00	(−3.18–0.75)	0.227
Model 2. STAI-S score
Intervention group:Time point	−2.93	1.28	(−5.44– −0.42)	0.022
Model 3. IES-r score
Intervention group:Time point	−1.81	1.44	(−4.63–1.02)	0.210

^aAll models adjusted for fixed effects including study site, participant education level, and infant cardiac diagnosis (biventricular or single ventricle), with a random slope for participant-specific trajectories Abbrevations: CES-D = Center for Epidemiologic Studies Depression Scale; CI = confidence interval; IES-r = Impact of Event Scale - Revised; SE = standard error; STAI-S = State-Trait Anxiety Inventory - State subscale

A visual exploration of pilot data illustrated potential differences in intervention effect based on infant cardiac diagnosis (See Supplemental Fig. 2). Participants with a fetal diagnosis of single ventricle physiology appeared to have slightly higher levels of depression, anxiety, and trauma symptoms at baseline in both groups; however, participants in the PHM^™ group appeared to experience reduced symptoms over time while symptoms in the control group remained elevated.

4. Discussion and Future Direction

This pilot study of a nurse-guided mHealth care program demonstrated feasibility and potentially reduced emotional distress for parents in the PHM^™ group compared to the control group. Feasibility reflected in participant retention favored the intervention group, however, attrition from both groups was mainly due to infant demise, transfer for transplant, or postnatal diagnostic ineligibility. Feasibility of the intervention was shown with ≥96% session attendance and most parents (65%) using app-based messaging with the nurse. Results also demonstrated a clinically meaningful reduction in emotional distress (i.e., depressive, anxiety, and traumatic stress symptoms) for the PHM^™ group. To our knowledge, this is the first report on the impact of a psycho-educational mHealth intervention on parents’ emotional distress before and after the birth of an infant with CCHD.

The intervention group had generally high retention, attendance, and engagement, suggesting that the PHM^™ care program was valuable to these parents. Our approach addressed some limitations of previous in-person interventions [34,36–39]. The nurse guide could engage parents and tailor the intervention through remote sessions combined with regularly scheduled check-ins and messaging. It is possible app topics discussed and recommended by the nurse guide further promoted parents’ engagement in the PHM^™ care program during the prenatal time. Sharing some similarities to this study was a recently piloted app-based telehealth intervention that included 3–12 prenatal sessions for parents after a fetal CCHD diagnosis with a clinical psychologist focusing therapeutic discussions on emotional experiences and preparing for stressors [44]. Feasibility of that intervention was demonstrated with 27 (77%) mothers and 9 (60%) fathers completing at least 3 prenatal telehealth sessions. Our study showed all eligible parents except for one caregiving partner attended available intervention sessions. Of the 37 parents in our study who attended prenatal sessions, 81% participated as dyads, which was substantially more than the 48% of dyads in the telehealth study. While parents attended 1–2 prenatal sessions, having the option to initiate messaging along with receiving frequent personalized messages from the nurse guide was important for ongoing communication. This app-based messaging allowed parents to ask questions, share new information and concerns, and maintain a relational partnership with the nurse guide.

Our nurse-guided PHM^™ care program was designed to reduce parents’ emotional distress through consistent psychosocial support across the perinatal time with anticipatory condition-specific information (e.g., about the CCHD diagnosis, surgical/medical treatment, course of illness, recovery, and feeding). The provision and timing of this type of support and information were identified gaps in the telehealth study [44]. Other efforts to enhance nurse-parent partnerships have been successful in increasing mothers’ involvement in care, psychological connection to their infants, and parental self-efficacy while reducing anxiety during their infants’ recovery from cardiac surgery [36,60]. Additionally, several studies on nurse-parent partnerships have supported parents’ development of caregiving competencies in the NICU setting [61]. These findings illustrate nurses’ potential to provide much-needed care to parents after a prenatal CCHD diagnosis and across the perinatal time. Having interventionists with advanced practice degrees (e.g., nurse practitioner or clinical psychologist) could be appropriate to meet some parents’ needs. However, the availability of these clinicians to address parent mental health is limited in the maternal-fetal care context, and the additional cost to integrate these services would require further study. Our nurse guides were able to combine therapeutic communication skills and direct care experience for families of children diagnosed with CCHD with their specialized interventionist training to partner with and support parents in this intervention. In this relational, nurse-parent partnership context, parents could view the nurse guide as a navigator and ally, thus facilitating screening, appropriate referrals, and uptake for mental health services, which aligns with several policy statements focused on improving perinatal care in the United States [45,62,63].

At baseline, parents in our sample reported prenatal symptoms of depression, anxiety, and traumatic stress that were elevated compared to norms [64–66], and our findings align with other studies of parents in this context [67–69]. Studies measuring emotional distress within 3 months post-birth that explicitly included parents who had a prenatal diagnosis reported postnatal scores for anxiety, depression, and traumatic stress that were similar to our baseline scores [70–73]. Another study found lower levels of preoperative maternal depression and anxiety symptoms compared to our sample; however, those infants appeared to have less complex cardiac conditions [74]. Lastly, one study reported higher parental traumatic stress scores following the Norwood surgery compared to all study time points for our full sample, and for the subgroup whose infants were diagnosed with single ventricle physiology [75].

Importantly, we found that, on average, scores on all emotional distress measures were reduced for PHM^™ group parents compared to the control group across study time points. It should be noted that PHM^™ group emotional distress scores were significantly lower at baseline, compared to the control group; therefore, it is promising that we observed point estimate reductions in intervention group symptoms despite more limited room for improvement. We also observed an overall decline in emotional distress postnatally in both groups. This decline might have reflected that parents’ uncertainties were reduced following infant birth, upon completion of surgery/medical interventions, and/or after a plan of care was established. Furthermore, postnatal data were not collected from parents after infant demise (disproportionally in the control group), suggesting that some participants with particularly complicated infant diagnoses were represented in the prenatal, but not the postnatal data. Further visualization exploring outcomes suggested that the care program may have benefited PHM^™ group parents of infants with single ventricle physiology in particular. Considering that single ventricle CHD is considered the most severe form of the disease, extending mHealth care to these parents could be important. These families appear to be especially vulnerable and may be more responsive to psycho-educational support. One large study capturing birthing persons’ mental health longitudinally from the prenatal to postnatal time frame showed a strikingly worse trajectory for emotional distress with significantly higher depression and anxiety scores for birthing persons of infants with more severe CCHD, (e.g., single ventricle CHD) [73,76,77].

Finally, extending technology-based care has the potential for reducing socioeconomic and racial inequities. Recent findings point to inequities related to prenatal care coordination and pre-operative morbidities, including worse infant feeding outcomes [24]. While telehealth monitoring and apps have been designed for tracking growth and development for infants with CCHD [42,78–80], it is possible that employing mHealth to broaden access to adequate prenatal preparation for infant feeding might positively impact infant and parent outcomes [48]. Improving care specifically to support human milk feeding and direct breastfeeding over the perinatal time is especially needed [81,82].

4.1. Strengths and limitations

Strengths of this study include piloting an mHealth-based intervention to extend evidence-based supportive care following a prenatal diagnosis with repeated measures at geographically diverse sites. Pilot study findings, however, must be considered in the context of the following limitations. Although our results indicate potentially clinically meaningful reduction in emotional distress outcomes, the final sample size was underpowered to detect statistical significance. Our recruitment approach did not allow us to describe the sociodemographic or clinical representativeness of this sample within the broader CCHD population at our study sites. Additionally, self-selection bias may have influenced results, as parents who participated could have been more emotionally stable or had access to more resources (e.g., partner support, income, education) compared to those who did not participate. While we explored analytical adjustment for pregnant person/caregiving partner status, it is possible that the PHM^™ intervention could have different effects on different members of the parental dyad, and this is an important consideration for future study. Finally, although recruitment took place in two healthcare centers, regional and institutional care practices could differ and our results may not be generalizable to all settings.

4.2. Future direction

Study findings are informing further development of the PHM^™ care program and preparations for larger-scale testing. We are continuing to develop the PHM^™ care program to meet parents’ needs for support and information. While the nurse guides were able to learn about and address many issues of parental concern [48], screening and referral to mental health services (e.g., nurse practitioner or psychologist on the healthcare team or within the clinic/medical center) could be a key integration for the PHM^™ care program. In terms of addressing parents’ caregiving needs, we identified a clear gap in human milk feeding and direct breastfeeding content in the piloted care program. Content tailored for single ventricle diagnoses will also be considered, given a recent report underscoring that anxiety and traumatic stress are of notable concern for parents post-Norwood [75]. Lastly, all care program content has been translated, and future iterations will be available in English and Spanish.

5. Conclusion

In this pilot study of the PHM^™ care program, we demonstrated intervention feasibility and potential improvements in emotional distress for parents following a fetal diagnosis of CCHD. Using mHealth with a nurse-guided patient engagement approach holds promise for extending care to this vulnerable population of parents with psychosocial risks following a CCHD diagnosis [19,45]. Future directions include expanding care program content on infant feeding and optimizing patient engagement strategies to ensure visibility and tailoring of content for parents in English and Spanish. Further testing with a larger, more diverse sample is needed to determine the effectiveness of the PHM^™ care program. The long-term goal is to integrate a feasible and efficacious mHealth care program with patient engagement strategies into clinical settings to reduce parents’ emotional distress and increase caregiving capacities to improve infant and family outcomes.

Supplementary Material

1

Supplementary Figure 1. Average proportion of each topic viewed by parents

2

Supplementary Figure 2. Visual exploration of potential differences in intervention effect based on infant diagnosis

Highlights.

• Parents need support and information after a prenatal CCHD diagnosis

• Emotional distress can endure for parents as they face many caregiving challenges

• Parents’ engagement with the PHM^™ care program demonstrated feasibility

• PHM^™ is a novel approach with potential to reduce parents’ emotional distress

Abbreviations:

CCHD

critical congenital heart disease

mHealth

mobile health

PHM^™

Preparing Heart and Mind^™ care program, a psycho-educational intervention

SMI

Severity of Medical Illness

CES-D

Center for Epidemiological Studies-Depression

STAI-S

State-Trait Anxiety Inventory-State subscale

IES-r

Impact of Event-revised