An interactive childbirth education platform to improve pregnancy-related anxiety: A randomized trial

Abstract

Background:

Perinatal anxiety is one of the most common conditions during pregnancy and is associated with adverse maternal and neonatal outcomes. Interventions that are focused on childbirth education and health literacy have been shown to help decrease pregnancy-related anxiety. These programs, however, have limitations. Transportation, childcare, and work conflicts pose barriers to patients. Additionally, many of these programs have not been studied in a high-risk patient population, the patients most at risk for pregnancy-related anxiety. It is uncertain, whether an online childbirth education course can help to improve outcomes in a high-risk patient population.

Objective:

To compare an interactive online platform for childbirth education (Birthly), to usual prenatal education on anxiety, emergency healthcare utilization, and delivery outcomes for high-risk pregnancies.

Study Design:

We performed a randomized trial comparing an interactive online childbirth education platform along with usual prenatal education (intervention) to usual prenatal education alone (usual care). Nulliparous, English-speaking patients with internet access and a high-risk pregnancy (medical or mental health disorders) were included. Patients in 2 urban clinics that serve under-resourced patients were enrolled at <20 weeks. The intervention included 3 interactive courses (prenatal bootcamp, breastfeeding, newborn care) and access to a clinician-moderated online community. Pregnancy-related Anxiety Scale questionnaires were administered at randomization and at 34-40 weeks. The primary outcome was 3^rd trimester Pregnancy-related Anxiety Scale score. Secondary outcomes included change in Pregnancy-related Anxiety Scale score, unscheduled emergency visits, delivery and postpartum outcomes. In order to demonstrate a 15% decrease in Pregnancy-related Anxiety Scale score 37 patients would be needed per group. Accounting for a 20% loss to follow-up rate, we planned to recruit 90 total patients or 45 per group.

Results:

90 patients were randomized with no differences in demographics or baseline Pregnancy-related Anxiety Scale scores. The majority of patients self-identified as Black and were publicly insured. More than 60% of patients (62.2%) in the intervention arm completed at least one Birthly course. Patients in the intervention arm had significantly lower third trimester Pregnancy-related Anxiety Scale scores (indicating lower anxiety) compared to usual care, 44.6 ± 7.3 vs. 53.9 ± 13.8, p< 0.01 (Table 2), with a decrease in score of 8.3 points (intervention) vs. 0.7 (usual care), p<0.01. Patients in the intervention arm also had fewer emergency visits (1[0-2] vs. 2[1-3], p=0.003). There were no differences in delivery outcomes. Patients in the intervention arm were more likely to breastfeed at delivery, though this was not different by the postpartum visit. Finally, patients who received the intervention were more likely to be satisfied with their childbirth education (94.6% vs. 64.9%, p< 0.01).

Conclusion:

An interactive online childbirth education platform can reduce pregnancy-related anxiety and emergency healthcare utilization while improving satisfaction in a high-risk patient population.

Condensation:

An interactive online childbirth education platform can reduce pregnancy-related anxiety and emergency healthcare utilization in a high-risk patient population.

Introduction

Perinatal anxiety is one of the most common mental health conditions during pregnancy, affecting about 20% of all patients in the peripartum period.¹ Perinatal anxiety has significant implications for both pregnant patients and their children. Perinatal anxiety is associated with adverse maternal and neonatal outcomes including a 1.5-1.7 increased odd of preterm birth, a 1.5 increased risk of low birthweight, and long-term developmental and mental health sequelae in children.^1,2 Pregnancy-related anxiety, defined as a multidimensional anxiety related to childbirth and concerns about fetal health, is one specific anxiety domain that is exceptionally common during pregnancy; recent studies suggest that 55-70% of nulliparous women exhibit at least mild pregnancy-related anxiety by the third trimester with 5-15% experiencing severe symptoms.^3,4

In-person interventions focused on childbirth preparation and improving health literacy have been shown to decrease pregnancy-related anxiety while improving maternal and neonatal delivery outcomes.^5,6 These in-person programs, however, have limitations, as transportation, childcare, and work pose barriers to patients.⁷ Internet-based prenatal education is an increasingly popular option to help offset these burdens. One such program is Birthly, a platform for interactive online childbirth education. Serving as an adjunct to prenatal care, Birthly gives patients access to education about birthing, breastfeeding, and newborn care and allows patients to engage in a virtual community. Though Birthly has been implemented in many clinical practices across the United States, its impact on pregnancy-related anxiety has not been studied.

Additionally, there have been few studies that explore the implementation of prenatal education programs in a high-risk patient population. Patients with medical and mental health comorbidities are often excluded from such studies, but are at the highest risk for pregnancy-related anxiety.⁸ There has also been increasing attention to the relationship between social determinants of health, including anti-Black racism, socioeconomic status, and social support, on patient anxiety as well as adverse pregnancy outcomes.^9,10 There is public health urgency to determining if prenatal education interventions may help to improve outcomes for this vulnerable patient population.

We sought to compare Birthly plus usual prenatal education to usual prenatal education alone on pregnancy-related anxiety, emergency healthcare utilization, and delivery outcomes for high-risk pregnancies in an underserved patient population.

Materials and Methods

We performed a randomized controlled trial comparing Birthly plus usual prenatal education versus usual prenatal education alone from April 2021 to April 2022. This study was approved by the Institutional Review Board at the University of Pennsylvania and was registered on ClinicalTrials.gov (NCT04783480) and written consent was obtained for participants prior to enrollment. CONSORT reporting guidelines were followed.¹¹

Patients were approached from April 2021 to November 2021 and were followed until their postpartum visit; the final patient completed follow-up in April 2022. Patients were recruited from two urban clinics affiliated with our health system. The majority of patients served in these clinics are under-resourced and publically insured. Patients were included if they met all of the following criteria: aged ≥18 years, <20 weeks gestation, nulliparous, English-speaking, reported internet access either through their phone or home computer, and were determined to have a high-risk pregnancy. High-risk pregnancy was defined as at least one pre-existing maternal medical or mental health condition. Maternal medical conditions eligible for inclusion were chronic hypertension, diabetes, at least moderate persistent asthma, autoimmune conditions (Lupus or antiphospholipid antibody syndrome), a history of venous thromboembolism, thyroid disease, history of bariatric surgery, and body mass index (BMI) ≥40. Maternal mental health conditions eligible for inclusion were a documented history of major depressive disorder, generalized anxiety disorder, post-traumatic stress disorder, bipolar disorder, and substance use disorder. Patients were excluded if they had a multiple gestation, were enrolled in group prenatal care, or had an indication for a planned cesarean delivery that was identified at the index prenatal visit.

Once the patient was enrolled, randomization was performed via computer-generated block randomization in a 1:1 scheme to (1) Birthly + usual prenatal education or (2) usual prenatal education alone. The patient was emailed their allocation; those randomized to Birthly were provided the website, instructions for scheduling classes, and a unique code to use at the time of scheduling which allowed the investigators to track patients course enrollment and attendance. While Birthly is a fee-for-service program, patients randomized to Birthly in the study received the courses at no cost. Courses range from $50-$150 and can be accessed at this website: https://www.mybirthly.com/.

The Birthly intervention included three live, interactive courses, which included “Prenatal Education Bootcamp”, “Breastfeeding 101”, and “Newborn Care.” Two additional courses “Early Pregnancy Preparations” and “Coping and Comfort” were offered as supplements. Courses were taught by prenatal education experts including lactation consultants, nurses, doulas, and certified childbirth educators. Patients also had access to a clinician-moderated online community and complementary articles and blog posts written by the childbirth educators. Patients allocated to Birthly plus usual prenatal education were encouraged to take all three courses before 36 weeks gestation and were sent monthly reminder e-mails if they had not registered for the courses.

Both arms received usual prenatal education. While this was not standardized for the study, usual prenatal education in our health system generally consists of counseling during routine prenatal care visits and meetings with a nursing care coordinator at the initial prenatal visit, as well as each trimester. These meetings focus on anticipatory guidance for pregnancy milestones, signs and symptoms of labor, and fetal kick counts. Clinicians caring for patients were not informed of the patients’ allocation in this study.

The primary outcome was third-trimester pregnancy-related anxiety, assessed using the Pregnancy-related Anxiety Scale (PrAS) questionnaire. The PrAS is a validated instrument, which includes a 32-item self-report scale that is designed to assess pregnancy anxiety with a specific focus on the fear of childbirth.¹² The scale is divided into 8 domains: childbirth concerns, body image concerns, attitudes towards childbirth, worry about self, acceptance of pregnancy, attitudes towards medical staff, avoidance, and baby concerns.¹³ It is based in a Likert scale [1-4] and takes approximately 10 minutes to complete. Scores range from 32-128 with a higher score indicating greater pregnancy-related anxiety. Previous validation research on the PrAS has demonstrated that this questionnaire specifically captures pregnancy-related anxiety, with minimal contribution of anxiety and depression to the score.¹³ The questionnaire was emailed to patients to complete at the time of randomization and again at 34-40 weeks. All patients were provided a $25 gift card for participation at the completion of the post-intervention survey.

Planned secondary outcomes included the change in Pregnancy-related Anxiety Scale score between the randomization and third trimester, unscheduled emergency healthcare utilization (defined as unplanned visits to the emergency department or labor floor triage during pregnancy after randomization which did not result in admission), as well as number of phone calls and/or electronic messages to the practice triage nurse during that time frame. Finally, we collected data on maternal and neonatal delivery outcomes. Secondary maternal outcomes included gestational age at delivery, mode of delivery, and maternal length of stay. Neonatal outcomes included birthweight, neonatal ICU (NICU) admission, and neonatal length of stay. Patient satisfaction with prenatal education was also assessed at the time of the 34-40 week PrAS score, using a Likert scale [1-5]. Postpartum data was also collected including breastfeeding at delivery discharge, contraceptive uptake, attendance at the postpartum visit (defined as a visit with a maternal health clinician within 2-12 weeks postpartum), maternal readmission, Edinburgh Postpartum Depression Scale (EPDS) score, and feeding method at the postpartum visit. Finally, we assessed if patients accessed any additional prenatal education resources and their satisfaction with their prenatal education as part of the third trimester survey.

In order to determine our expected sample size, we assessed existing Pregnancy-related Anxiety Scale score data. While no studies have demonstrated a cut-off for pregnancy related anxiety, in a large population study of 600 patients in Australia, the mean PrAS score for pregnant women was 64.22 (range 32 to 128) with a standard deviation of 15.22.¹⁴ In order to have 80% power to demonstrate a 15% decrease in PrAS score from 64 to 55 with a SD of 15.22, assuming an alpha of 5%, 37 patients would be needed per group. This change in PrAS score was thought to be clinically significant. Accounting for a 20% loss to follow-up rate, we planned to recruit 90 total patients or 45 per group.

Fisher exact tests and chi-square tests were used for categorical variables and t-tests or Wilcoxon rank sum tests were used for continuous variables, where appropriate. Analysis was performed as intention-to-treat. Statistical analyses were performed with Stata 15 (StataCorp, College Station, TX) and statistical significance was set at p<0.05.

Results

From April to November 2021, a total of 242 eligible patients were eligible for enrollment. 174 patients were successfully contacted of whom 90 (51.7%) provided informed consent and underwent randomization. A total of 45 patients were randomized to Birthly plus usual prenatal education (intervention) and 45 to usual prenatal education alone (usual care) (Figure 1).

Figure 1:

Study Participants

Baseline demographics were similar between groups (Table 1). The majority of patients self-identified as Black and were publicly insured. Maternal mental health disorder was the most common high-risk condition in both groups with more than 60% of the population having either major depressive disorder, generalized anxiety disorder, or post-traumatic stress disorder. PrAS scores were similar at the time of study enrollment. More than 60% of patients (62.2%) in the intervention arm completed at least one Birthly course with a mean of 1.7 (SD±1.6) courses completed per participant.

Table 1:

Baseline characteristics

Demographics	Birthly + usual prenatal education (n=45)	Usual prenatal education alone (n=45)
Maternal age (yrs)*	23.7 [19, 27]	23.7 [20, 29]
Black race	35 (77.8%)	31 (68.9%)
Non-Hispanic Ethnicity	43 (95.6%)	41 (91.1%)
Insurance status
Private	10 (22.2%)	14 (31.1%)
Medicaid	35 (77.8%)	31 (68.9%)
Gestational age at randomization (wks)	13.7 ± 2.5	14.5 ± 2.3
BMI at first prenatal visit*	26.6 [22.3, 31.9]	26.6 [22.0, 37.6]
Baseline PRAS score	53.2 ± 1.6	55.0 ± 1.7
High risk condition
Chronic hypertension	3 (6.7%)	8 (17.8%)
Diabetes	6 (13.3%)	3 (6.7%)
Mental health disorder†	27 (60.0%)	30 (66.7%)
Asthma	14 (31.1%)	13 (28.9%)
Other§	12 (26.7%)	13 (28.9%)

All data presented as n (%) vs. mean +/− SD unless specified

^*Median [IQR]

^†Includes major depressive disorder, generalized anxiety disorder, post-traumatic stress disorder

^§Includes at least moderate persistent asthma, autoimmune conditions (Lupus or antiphospholipid antibody syndrome), a history of venous thromboembolism, thyroid disease, history of bariatric surgery, and BMI ≥40

Patients in the intervention arm had significantly lower third trimester PrAS scores (indicating lower anxiety) compared to usual care, 44.6 ± 7.3 vs. 53.9 ± 13.8, p<0.01 (Table 2), with a decrease in PrAS score of 8.3 points (intervention) vs. 0.7 (usual care), p<0.01. Patients in the intervention arm also had fewer emergency visits (1[0-2] vs. 2[1-3], p<0.01). There were no differences in the number of triage telephone calls or messages during pregnancy between groups. Differences in scores for individual questions of the third trimester PrAS questionnaire are shown in Supplemental Table I. Patients who received Birthly had a significantly lower mean score in the domains of body image concerns, worry about self and generalized anxiety, concerns about the baby, and improved attitudes towards delivery providers. The results were similar when this analysis was performed examining the groups as-treated. Those patients who attended at least one Birthly course had a significant decrease in third trimester PrAS score (9.8 ± 2.4 vs. 1.5 ± 1.6, p<0.01) and significantly fewer emergency visits (1.5 [1-3] vs. 1 [0-2], p<0.01) compared to those that did not attend any classes.

Table 2:

Outcomes for intervention vs. usual care

Outcomes†	Birthly + usual prenatal education (n=42)	Usual prenatal education alone (n=41)	P-value
Post-intervention PrAS scores	44.6 ± 7.3	53.9 ± 13.8	<0.01
ΔPrAS (pre intervention - post intervention)	−8.3 ± 1.9	−0.7 ± 2.0	<0.01
Unscheduled emergency visits (unplanned ED or triage visits) after randomization*	1 [0, 2]	2 [1, 3]	0.003
Triage calls or messages during pregnancy	1.5 [1,3]	2 [0,4]	0.46
Delivery outcomes
Number of prenatal visits	9.1 ± 2.6	8.7 ± 2.5	0.40
GA at delivery*	39.0 [37, 39]	38.0 [37, 40]	0.88
Mode of delivery			0.73
Vaginal	27 (64.3%)	29 (70.7%)
Operative vaginal	1 (2.4%)	0 (0.0%)
C-section	14 (33.3%)	12 (29.3%)
Maternal length of stay*	2 [2, 3]	2 [2, 2.5]	0.99
Neonatal outcomes
Birthweight (grams)**	3050 [2730, 3230]	3012 [2770, 3180]	0.42
Neonatal ICU admission	9 (21.4%)	7 (17.5%)	0.83
Neonatal length of stay**	2 [2, 3]	2 [2, 3]	0.85
Postpartum outcomes
Breastfeeding at delivery discharge	32 (80.0%)	23 (56.1%)	0.03
Contraceptive uptake at delivery discharge	20 (47.6%)	17 (41.4%)	0.83
Postpartum unscheduled emergency visits	0 [0,0]	0 [0,0]	0.34
Postpartum triage calls or messages	0 [0,1]	0 [0,1]	0.62
Attended postpartum visit	32 (74.0%)	32 (78.0%)	0.70
EPDS score at postpartum visit*	4 [0, 8]	4 [1, 10]	0.55
Breastfeeding at postpartum visit§	10 (32.0%)	15 (47.0%)	0.31

All data reported as N (%) vs. mean +/− SD unless otherwise specified

^*Median [IQR]

^†Six patients enrolled lost to follow-up

^§Among those who attended the postpartum visit

There were no differences in delivery outcomes, including in gestational age at delivery or mode of delivery (Table 2). Patients in the intervention arm were more likely to breastfeed at delivery (80.0% vs. 56.1%, p=0.03), though this was no longer significant by the postpartum visit. There was also no difference in rate of postpartum visit attendance or EPDS scores at the postpartum visit (Table 2). There was no difference in the numbers patients who sought other prenatal education among patients in the Birthly arm compared to usual care (20.0% vs. 15.6%, p=0.57). Patients who received the intervention were more likely to be satisfied or very satisfied with their childbirth education (94.6% vs. 64.9%, p<0.01).

Comment

Principal findings

In this study, patients randomized to Birthly plus usual prenatal education had significantly reduced third trimester pregnancy-related anxiety compared to those patients who received usual prenatal education alone. The Birthly intervention specifically improved patient scores in the domains of childbirth concerns, body image concerns, worry about self and generalized anxiety, baby concerns, and attitudes towards delivery providers. Patients who received the Birthly intervention also had fewer unscheduled emergency visits, higher uptake of breastfeeding immediately postpartum, and had improved satisfaction with their prenatal education.

Results in the Context of What is Known

While there has been substantial research on the positive impact of in-person prenatal education on maternal pregnancy-related anxiety and pregnancy outcomes, less is known about the implementation of virtual courses. One study randomized 23 patients to an 8-session virtual childbirth education course and 21 patients to usual prenatal education.¹⁵ The authors found improved fear of childbirth and preparedness for childbirth among patients who received the intervention. Specifically, on the Fear of Birth scale, scores improved from 5.4 out of 10 to 4.4 out of 10 p=0.01, indicating decreased fears related to childbirth. Notably, however, the majority of patients in this study had high or moderate income and the study excluded patients with pre-existing mental health conditions. We aimed to study a high-risk, vulnerable patient population due to the higher risk of worsening anxiety during pregnancy. In fact, the majority of the study group had pre-existing depression, anxiety, or PTSD. The demonstrated benefit of Birthly compared to usual prenatal care in reducing pregnancy-related anxiety in this patient population lends even further strength to the value of the intervention and the generalizability of the results. Similarly, we chose to conduct our study in a diverse, underinsured and under-resourced patient population as these patients are at higher risk of adverse outcomes, including perinatal anxiety and depression.

Clinical Implications

Recent studies have suggested that pregnancy-related anxiety has only increased in the wake of the COVID-19 pandemic. Research from the United States and Italy has shown that patients who gave birth during the pandemic had significantly increased fears of childbirth during pregnancy, acute stress response during birth, and post-traumatic stress symptoms in the postpartum period.^16,17 In our own institution, a survey of multiparous postpartum patients during the pandemic found that 34.4% of patients reported increased postpartum anxiety compared to prior pregnancies.¹⁸ The implementation of an online platform for childbirth education is, therefore, particularly timely, as patients who are pregnant during the pandemic have increased mental health needs yet less access to in-person groups and services.

Additionally, while pregnancy-related anxiety has substantial implications for individual patients and their families, this also an impact on the healthcare system. Previous research has demonstrated that there is an association between psychosocial factors including pregnancy-related anxiety and high utilization of unscheduled emergency care.¹⁹ Our study found that the Birthly intervention significantly reduced the number of unscheduled emergency triage visits; this supports the idea that prenatal education can help to decrease low-value care.

Research Implications

This small, randomized trial provides initial data supporting the use of Birthly to reduce pregnancy-related anxiety and emergency healthcare utilization. Further work should be done to study the implementation and cost-effectiveness of this intervention in a larger population. While the cost of Birthly may be a barrier to individual patients, a cost-effectiveness analysis may lend evidence to insurer-based coverage of this service for patients.

Strengths and Limitations

There are several strengths to our study. This study was a randomized controlled trial of a diverse, high-risk patient population. A majority of the patients had a history of a mental health disorder, the population that is at highest risk for postpartum depression and adverse pregnancy outcomes. As Birthly was shown to improve pregnancy-related anxiety in our study population, this suggests that the intervention can benefit those with psychiatric comorbidities. Additionally, though patients and the authors were aware of the patients’ study allocation, prenatal care providers were blinded to the groups. Additionally, we were able to assess a variety of different outcomes and collect granular data, including the number of unscheduled emergency visits and triage phone calls and messages from the patients’ index prenatal visit to the postpartum visit.

Our study also has several limitations. All patients received their prenatal care at two of our institution’s clinical sites which follow similar practice patterns. There was still likely significant variation in the “usual prenatal education” that patients received. This does, however, suggest that our results are generalizable to other practices, which may provide different prenatal education as part of routine care. Additionally, we did not preclude patients from seeking additional resources or classes during their prenatal care. There was no difference, however, in the numbers patients who sought other prenatal education among patients in the Birthly arm compared to usual care. Furthermore, our planned sample size was based on a study using PRAS outside of the US, and with minimal data supporting what might be a clinically significant difference in score. In addition, in-depth understanding of the difference seen in emergency utilization was limited, as indication for unscheduled emergency visit was not collected. We were also unable to collect data on Birthly user engagement. While we were able to track course completion for patients who received Birthly, we were not able to assess engagement in the courses or other supplementary materials. Finally, we were not powered to detect differences in maternal or neonatal outcomes; a larger study may be helpful to ascertain if Birthly can impact delivery outcomes in addition to pregnancy-related anxiety.

Conclusion

An interactive online childbirth education platform can reduce pregnancy-related anxiety and emergency visits while improving satisfaction in a high-risk patient population. Implementation of Birthly has the potential to improve individual patient outcomes while helping to reduce high cost, low-value healthcare utilization.

AJOG at a Glance:

A. Why was this study conducted?

• This study aimed to determine whether an interactive online childbirth education platform could improve pregnancy-related anxiety, as measured by the Pregnancy-related Anxiety Scale questionnaire, compared to usual prenatal care

B. What are the key findings?

• Patients randomized to the childbirth education platform had significantly lower third trimester Pregnancy-related Anxiety Scale scores compared to those receiving usual care

• Patients randomized to the childbirth education platform had fewer emergency visits and were more likely to breastfeed at delivery than those who received usual care

• Patients randomized to the childbirth education platform were more likely to be satisfied with their childbirth education

C. What does this study add to what is known?

• Previous studies have found that in-person interventions can improve pregnancy-related anxiety. These studies have been limited primarily to low-risk patient populations.

• The results of this study demonstrate that an interactive online childbirth education platform improves pregnancy-related anxiety in high-risk patients.

• These findings are particularly timely as the COVID-19 pandemic has increased mental health needs amount pregnant patients while limiting access to in-person groups and services.