HUGS/Abrazos: A Community-Based Initiative Improved Perinatal Mental Health in an Urban Cohort

Abstract

Background:

Targeted programs to improve maternal mental health, particularly for those with social determinants of health, are increasingly critical since the onset of the COVID-19 pandemic, yet the impact of such programs are poorly understood.

Objective:

To evaluate the impact of a novel language-concordant community-based program on perinatal mental health.

Study Design:

We conducted a prospective cohort study of peripartum individuals referred to a new community-based intervention, Helping Us Grow Stronger (HUGS/Abrazos). Participants received up to four remote sessions from a cognitive behavioral therapy (CBT)-trained social worker, up to three resource navigation visits with a community health worker, a grocery gift card and care package. Before and after the program, participants completed validated survey instruments assessing mental health and social determinants of health.

Results:

178 participants were assessed after program completion, including 133 who were assessed pre- and post-program. The cohort identified as 62.9% Hispanic/Latinx with a mean age of 29.8 (SEM 0.46). There were high rates of food insecurity (111/178, 62.4%), experiences of discrimination (119/178, 66.9%), and SARS-CoV-2 infection (105/178, 59.0%). The program was associated with statistically significant improvement in Edinburgh Postnatal Depression (EPDS) scores (baseline 8.44 ± 0.55 vs 6.77 ± 0.51 post, p=0.0001, mean ± SEM) and Perceived Stress Scale (PSS) scores (baseline 15.2 ± 0.74 vs 14.0 ± 0.71 post, p=0.035). Participants with stressors including food insecurity and experience of discrimination had higher baseline depression, stress and anxiety scores. Those with experiences of discrimination, food insecurity, and SARS-CoV-2 infection in pregnancy were more likely to have improvement in mental health scores post-intervention.

Conclusion:

In this diverse urban cohort, a novel community-based intervention was associated with improvement in depressive symptoms, perceived stress, and anxiety, particularly for those with social determinants of health.

Introduction

Social determinants of health, including racism, food insecurity, insurance status, and education, have been implicated in perinatal mental health and obstetric outcomes.^1–5 These effects may be mediated by maternal immune activation, inflammation, and stress, causing adverse pregnancy outcomes and metabolic and neurodevelopmental outcomes in offspring.^6–8 The COVID-19 pandemic amplified these impacts, with more severe disease and inflammation in individuals with complex social determinants of health as well as worsening disparities in perinatal health during the pandemic.^9–17 As such, targeted interventions to improve perinatal mental health and address social determinants of health have become even more critical, with the potential to improve outcomes not only for pregnant individuals but also for their children.

Patient navigation programs that pair patients with advocates to access available resources have been shown to have high patient satisfaction and improve perinatal health outcomes.^18–21 However, the impact of multifactorial programs that combine resource navigation with behavioral health support have not been rigorously evaluated. Additionally, how such programs impact patients with significant social determinants of health in pregnancy during the COVID-19 pandemic has not yet been studied, nor has the impact of COVID-19-related inflammation on mental health indices before and after such programs.

This project aimed to evaluate the impact of a novel multi-component community-based intervention on perinatal mental health in a racially and ethnically diverse urban population during the COVID-19 pandemic. The primary objective was to evaluate the impact of the HUGS/Abrazos program on measures of maternal depression, anxiety, and stress during the COVID-19 pandemic. The secondary objectives were to measure the relationship between social determinants of health and response to the program, and to evaluate program satisfaction. We hypothesized that the intervention would be associated with improvement in mental health, particularly for those with social determinants of health needs and COVID-19 in pregnancy and the postpartum period.

Materials and Methods

Study Design

This was an open, single-arm proof-of-concept study of the efficacy of the Helping Us Grow Stronger (HUGS/Abrazos) Program in improving perinatal mental health and connecting participants to community resources. Participants referred to the program by their provider were approached by study staff via telephone and verbally consented for participation in a 30-minute series of validated survey instruments (described below) assessing mood, food insecurity, and experience of discrimination before and after program completion; those who had already initiated the HUGS/Abrazos program at the time of study initiation were assessed once after program completion. The cohort development is shown in Figure 1. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines were followed. The study was approved by our health system’s institutional review board.

Figure 1:

Cohort Development

HUGS/Abrazos Program

The HUGS/Abrazos Program is a community-based program developed to provide high-risk groups, including pregnant individuals, with behavioral health support and community resources targeted toward stressors exacerbated by the COVID-19 pandemic.²² The program and its implementation have previously been described.^22,23 Briefly, patients with behavioral health needs or social determinants of health (SDoH) are referred to the program by providers in Obstetrics and Gynecology, Pediatrics, Primary Care, and Family Medicine through an order in the electronic medical record. Pregnant and postpartum patients are eligible for referral if they screened positive for behavioral health or social determinants of health screens conducted at the initial prenatal, third trimester, and postpartum clinic visits. The referrals are triaged by the program, and participants are assigned to a community health worker (CHW) to navigate community resources targeting SDoH needs, a behavioral health worker (BH) to assist with mental health concerns, or both, depending on needs identified in the referral. Through the program, participants are eligible for up to three community health worker touchpoints and four behavioral health touchpoints, with referrals to ongoing care if needed. There are English- and Spanish-speaking community health workers and behavioral health providers available through the program, and participants who speak a different language are reached with the assistance of a certified hospital telephone interpreter. All participants receive direct relief with a care package including a $50 gift card for groceries, activity kits, language-concordant children’s books, and supplies (masks, hand sanitizer) to encourage safe practices during the pandemic.

Participants

Inclusion criteria were pregnant and postpartum (up to 1 year) individuals aged 18 or older who were referred to the HUGS/Abrazos Program between August 1, 2021, and August 30, 2022. Eligible participants were identified from referrals placed to the HUGS/Abrazos program in the medical record. Exclusion criteria were those individuals who declined HUGS/Abrazos services after referral, individuals who were more than one year postpartum, and those who lived outside of Massachusetts as the behavioral health workers were not licensed to provide services beyond our hospital network.

Surveys

Participants were asked to complete a set of validated surveys before and after completing the HUGS/Abrazos program. Participants could complete surveys either by phone with a study staff member, or via a link sent to the participant’s email address. Surveys were available to participants in English and Spanish if they completed them using the emailed link. There were certified bilingual study staff who conducted surveys by phone in English and Spanish, and for participants who spoke languages other than English and Spanish, a hospital telephone interpreter was used to conduct surveys. Baseline surveys were completed before the first HUGS/Abrazos touchpoint (either behavioral health or community health, and post-program surveys were administered to participants within one month of program completion. All participants completed the Edinburgh Postnatal Depression Scale (EPDS), the 10-item Perceived Stress Scale (PSS), the PROMIS Anxiety Scale, the 2-item Food Insecurity Screen , and the 19-item Major Experiences of Discrimination and Day to Day Unfair Treatment Scale.^24–29 Pregnant participants completed an additional measure of anxiety in order to assess anxiety specific to pregnancy and childbirth, the 10-item Pregnancy Related Anxiety Questionnaire (PRA).^30,31 All survey instruments were previously validated in pregnant and/or non-pregnant populations, and screen positive cutoffs were determined as recommended by the survey instruments themselves and/or prior peer-reviewed literature.^24–31 For the Major Experiences of Discrimination and Day to Day Unfair Treatment Scale, participants were considered to screen positive for discrimination if they endorsed at least one lifetime or everyday experience of discrimination. Participants were considered to screen positive for food insecurity if they reported that it was sometimes or often true that in the last 12 months “We worried whether our food would run out before we got money to buy more” or “The food we bought just didn’t last, and we didn’t have money to get more.”²⁸ After program completion, participants were asked three program satisfaction questions to assess this secondary outcome: “Overall, how has the HUGS program met your needs?”, “How helpful has the HUGS program been in connecting you to services, such as signing up for WIC, enrolling in SNAP, or participating in play groups?”, and “How likely is it that you would recommend the HUGS Program to a friend or family member who needed prenatal care or whose child needed pediatric care?”

Analysis

Survey data were collected and managed using REDCap electronic data capture tool. REDCap (Research Electronic Data Capture) is a secure, web-based application designed to support data capture for research studies.³² Demographic and pregnancy data as well as baseline and post-program survey scores were evaluated using descriptive statistics, including counts, percentages, means, and standard error of the means (SEM). Mean behavioral health survey scores (EPDS, PSS, PROMIS, and PRA) before and after program completion for pre-and post-program participants were compared using paired 2-tailed t-tests. To address the secondary objective of understanding the impact of social determinants of health on responsiveness to the program, change in mental health scores before and after program completion were evaluated in stratified subgroups defined by presence or absence of food insecurity, experience of discrimination, or COVID-19 infection in pregnancy or the postpartum period. Paired 2-tailed t-tests were again used to compare scores before and after program completion. The relationship between number of HUGS/Abrazos touchpoints and change in mental health scores was evaluated using Spearman’s rank correlation. Similarly, the relationship between time from COVID diagnosis and starting the HUGS/Abrazos program and change in mental health scores was evaluated using Spearman’s rank correlation. We estimated that 102 participants would yield 80% power to detect a 2.5-point change in perceived stress scores (corresponding to Cohen’s d ~ 0.3 or a small- to moderate-sized effect) with alpha=0.05 using a 2-tailed paired t-test; accounting for a possible 30% rate of attrition, this would require n=130 participants. Greater changes in PSS scores could be detected with even fewer patients. Score differences in this range have been associated with biologically relevant outcomes such as preterm birth in pregnant individuals.^33,34

Results

Participants

178 participants were assessed with validated instruments after completing the HUGS/Abrazos program, including 133 who were assessed both pre- and post-program. Participant characteristics are shown in Table 1. Most participants had public insurance (146/178, 82.0%), were born outside of the United States (92/178, 51.7%) and identified as Hispanic/Latinx (112/178, 62.9%). Overall, 111/178 (62.4%) screened positive for food insecurity in the past year, 105/178 (59.0%) had COVID-19 during pregnancy or postpartum, and 119/178 (66.9%) screened positive for everyday or lifetime experiences of discrimination. The majority of participants indicated satisfaction with the program, as 128/178 (71.9%) of participants reported that the program met their needs “Very Well” or “Extremely Well,” and when asked to rate how likely it is that they would recommend the program to others on a scale of 0–10, participants reported on average 9.23/10 (Table 2).

Table 1:

Participant Characteristics

	N=178
Age, mean (SD)	29.5 (6.1)
Race ethnicity
Hispanic or Latinx	94 (53.1%)
Non-Hispanic White	49 (27.7%)
Non-Hispanic Black	19 (10.7%)
Non-Hispanic Asian	3 (1.7%)
Non-Hispanic Other	6 (3.4%)
Not reported	6 (3.4%)
Language
English	99 (55.6%)
Spanish	65 (36.5%)
Portuguese	5 (2.8%)
Haitian Creole	2 (1.1%)
Urdu	1 (0.6%)
Arabic	5 (2.8%)
Not reported	1 (0.6%)
Insurance
Private	32 (18.0%)
Public	146 (82.0%)
Birth country
United States	44 (24.7%)
Other	68 (38.2)
Not Reported	66 (37.0%)
Education
More than high school	53 (29.8%)
High school graduate/GED	60 (33.7%)
Some high school or less	55 (30.9%)
Not reported	10 (5.6%)
Marital Status
Unmarried	111 (62.4%)
Married/living together	65 (36.5%)
Not reported	2 (1.1%)
Employment
Unemployed	105 (60.0%)
Employed	63 (35.3%)
Not reported	10 (5.6%)
Parity (median, IQR)	2 (1, 3)
Pregnant (vs postpartum at study entry)	47 (26.4%)
Ever had COVID	105 (59.0%)

Table 2:

HUGS Program Satisfaction

Question	Response, N (%) (N=178)
“Overall, how has the HUGS program met your needs?”
Not well at all (1)	3 (1.8%)
A little bit (2)	12 (7.2%)
Fairly well (3)	23 (13.9%)
Very well (4)	54 (32.5%)
Extremely well (5)	74 (44.6%)
Mean score	4.10 (1.02)
“How helpful has the HUGS program been in connecting you to services, such as signing up for WIC, enrolling in SNAP, or participating in play groups?”
Very unhelpful (1)	17 (10.3%)
Somewhat unhelpful (2)	8 (4.9%)
Somewhat helpful (3)	25 (15.2%)
Very helpful (4)	99 (60.0%)
Don’t know (5)	16 (9.7%)
“How likely is it that you would recommend the HUGS Program to a friend or family member who needed prenatal care or whose child needed pediatric care?”
Mean score (0 not likely at all, 10 extremely likely)	9.23 (1.93)

Improved Depression, Stress and Anxiety Scores After Completion of HUGS/Abrazos

Survey data before and after program completion for those with pre-and post-scores are shown in Figure 2. The program was associated with statistically significant improvement in EPDS scores (baseline 8.44 ± 0.55 vs 6.77 ± 0.51 post, p=0.0001, mean ±SEM) and PSS scores (baseline 15.2 ± 0.74 vs 14.0 ± 0.71 post, p=0.035), but not improvement in anxiety as measured by PROMIS scores (baseline 54.5± 1.04 vs 53.6±post, p=0.28) or PRA scores (baseline 18.54 ± 1.1 vs 17.88 ± 1.21 post, p=0.18) for pregnant participants. There was a significant reduction in those screened positive for depression on the EPDS as indicated by a score of 10 or above (42.1% baseline vs 31.2% post, p=0.003), and 46 participants (35%) had a clinically significant improvement in EPDS score, as indicated by a decrease in EPDS score of 4 or more points.^25,35 While PSS scores were significantly reduced post-program, a large proportion of participants reported moderate or high stress (a PSS score of 14 or above) both at baseline and post-program (54.1 baseline vs 49.6% post, p=0.46). There was not a significant improvement in the percent of participants who screened positive for anxiety as indicated by a PROMIS T-score of 55 or more (baseline 48% vs 45.9% post, p=0.6) or as indicated by a PRA score of 22 or more (baseline 36% vs 20% post, p=0.29). ^36,37

Figure 2:

Participant Mental Health Scores Pre- and Post-HUGS/Abrazos<>The HUGS/Abrazos program was associated with significant reductions in depression and perceived stress scores using validated instruments. The bar graphs depict the mean ±SEM scores on validated mental health survey instruments, including the Edinburgh Postnatal Depression Scale (EPDS), Perceived Stress Scale (PSS) and Pregnancy-Related Anxiety Scale (PRA) at baseline and and post-HUGS/Abrazos program completion. * indicates significant difference with p< 0.05; ** indicates significant difference with p<0.01.

Pregnant participants had higher pre-program EPDS (baseline 10.1 ± 1.02 for pregnant vs 7.7 ± 0.64 for postpartum), PSS (baseline 17.6 ± 1.4 for pregnant vs 14.1 ± 0.85 for postpartum), and PROMIS anxiety scores (baseline 58.5 ± 2.0 for pregnant vs 52.8 ± 1.1 for postpartum) compared to postpartum participants, but both pregnant and postpartum participants had significant improvements in EPDS, PSS, and PROMIS scores before and after program completion. As the number of HUGS touchpoints increased, improvement in EPDS scores (Spearman’s correlation coefficient r=0.22, p=0.01) and PROMIS anxiety scores (r=0.24, p<0.05) increased. There was no association between length of time from COVID diagnosis to time of program initiation and post-program score (EPDS Spearman’s r=−0.003, p=0.97; PSS r=−0.002, p=0.98; PROMIS r=−0.11, p=0.32) or change in score (EPDS r=−0.006, p=0.96; PSS r=0.04, p=0.75; PROMIS r=0.008, p=0.95) over the course of the program.

Impact of High-Risk Exposures on Depression, Stress and Anxiety Scores

To examine how high-risk exposures including food insecurity, SARS-CoV-2 infection in pregnancy, and experiences of racism were associated with maternal mental health scores and their experience with the program, we conducted pre-specified analyses stratified by the presence or absence of these exposures. Of the 133 participants who completed baseline and post-program surveys, 74 (55.6%) reported food insecurity, 78 (58.6%) screened positive for experience of discrimination, and 83 (62.4%) reported COVID-19 in pregnancy. Stratified survey results are shown in Figure 3.

Figure 3:

Scores Stratified by Food Insecurity, COVID-19 Status, and Experience of Discrimination

Life stressors such as food insecurity and experiences of discrimination were associated with higher baseline depression, stress and anxiety scores, and more improvement in scores post-program, as indicated by the top brackets. The bar graphs depict the mean ±SEM scores on validated mental health survey instruments, including the Edinburgh Postnatal Depression Scale (EPDS), Perceived Stress Scale (PSS) and PROMIS Anxiety Scale at baseline and post-HUGS/Abrazos program completion for individuals with and without food insecurity, COVID19, and discrimination. * indicates significant difference with p< 0.05; ** indicates significant difference with p<0.01, † indicates p>0.05 but <0.1.

Participants with food insecurity had significantly higher baseline EPDS (p<0.001), PSS (p<0.001) and PROMIS scores (p<0.001) compared to those without food insecurity. Those with food insecurity demonstrated significant improvement in EPDS (p=<0.001) and PSS (p=0.03) post-program, with improvement in PROMIS not meeting threshold for statistical significance (p=0.06). In contrast, those without food insecurity demonstrated significant improvement in depressive symptoms (p<0.005) but not improvement in other mental health scores (p=0.4 for PSS an p=0.6 for PROMIS, Figure 3).

Participants who screened positive for lifetime or daily experience of discrimination similarly had significantly higher baseline depression (p<0.001), perceived stress (p<0.001), and anxiety scores (p<0.001) than those who screened negative. Those who screened positive for major experiences of discrimination had significant improvement in EPDS scores (p<0.001) and PSS scores (p=0.02), whereas those who screened negative for experience of discrimination had significant improvement in EPDS scores (p=0.02) but not in PSS (p=0.57), or PROMIS scores (p=0.88) (Figure 3).

Participants with COVID-19 in pregnancy demonstrated significant improvement in mental health scores post-program, including improved EPDS (p<0.001), PSS (p<0.005), and PROMIS scores (p=0.04). In contrast, those without COVID-19 did not show improvement in mental health scores post-program (p=0.14 for EPDS, p=0.98 for PSS, and p=0.24 for PROMIS) (Figure 3). Thus, in this cohort, participants with high-risk exposures in pregnancy such as food insecurity, COVID-19 in pregnancy, and experiences of racism were more likely to have higher baseline stress, perceived stress and anxiety, and were more likely to demonstrate improvement in these indices after completing the program, compared to participants without such exposures.

Comment

Principal Findings

This study demonstrates that a multicomponent community-based program incorporating patient navigation, behavioral health care, and emergency relief for pregnant individuals was associated with improvement in depressive symptoms and perceived stress, and improvement in anxiety symptoms for a subset of participants. The improvement in depression scores was most robust, and was consistent across all primary and secondary (stratified) analyses with a significant reduction in those screening positive for depression post-program. The significant reduction in proportion of participants screening positive for depression post-program is likely the most clinically relevant finding. Those with significant life stressors including food insecurity, COVID-19 in pregnancy, and daily or lifetime experience of discrimination showed the greatest improvement in mental health scores after program completion. The study cohort was notable for high rates of food insecurity (62.4% versus 10% in the general population)³⁸ and experiences of racism and discrimination. These high rates of SDoH needs may reflect a typical urban underserved obstetric population and have not been previously described. There was high program satisfaction, with most participants reporting that the program met their needs and was helpful in connecting them with important resources for navigating the peripartum period and early parenthood.

Results in the Context of What is Known

These findings build upon a nascent body of literature demonstrating high baseline rates of stress, depression, and anxiety in those with SDoH needs.^39–42 We show that similar to other patient navigation programs evaluated in non-pregnant populations,^18,20 the HUGS/Abrazos program yielded high patient satisfaction in a peripartum population. Our study adds to the literature by demonstrating that the program was associated with improvement in perinatal mental health, particularly for those who experience stressors.

Clinical Implications

Our results suggest that a multicomponent intervention such as HUGS/Abrazos may supplement clinical care in addressing mental health concerns, particularly for those with SDoH needs, in the perinatal period. Adding support that couples direct relief with behavioral health support and patient navigation may help to address growing disparities in mental health treatment and access.

Research Implications

As this was a proof-of-concept study, future research could randomize individuals to the intervention versus standard of care to further assess the impact of the program. Future studies could also examine whether the observed benefit reflects simply the recognition of behavioral health or social determinants of health and the offer of support, or if specific individual components of the program such as direct relief, behavioral health support, or a patient navigator are most helpful for those with identified needs. Additionally, future studies could evaluate the impact of such interventions on neonatal and obstetric outcomes and long-term health of the maternal-neonatal dyad.

Strengths and Limitations

Strengths of this study include its prospective assessment of a novel program. We included a racially and ethnically diverse patient population, which may increase its generalizability. This study has several limitations. As a proof-of-concept study conducted during the early phases of the pandemic, we enrolled participants in multiple settings, leading to a more heterogeneous (but likely more generalizable) treatment population. While we included stratified analyses to reduce the risk of confounding, the non-randomized controlled design increases the risk of confounding in our analyses. Not all participants in the program agreed to participate in surveys, and were unable to assess the characteristics of individuals who declined to participate in the program. It is possible that those who declined participation may have had different baseline mental health or social determinants of health than those who participated. As such, the extent to which our results generalize to other populations and treatment settings will require further study. An additional limitation is the inability to examine potential benefits in terms of perinatal outcomes and neonatal or childhood outcomes because we captured individuals at different timepoints in the antepartum and postpartum periods, and we did not assess neonatal or child outcomes.

Conclusions

In aggregate, this study suggests that a multifactorial program including direct relief, resource navigation, and behavioral health support improves perinatal stress and depression scores, particularly for those with stressors including food insecurity and discrimination, which have been amplified by the COVID-19 pandemic. Multifactorial interventions such as the HUGS/Abrazos program represent a promising approach for addressing the mental health effects of social determinants of health and health inequity in the perinatal period.

Condensation:

A community-based intervention was associated with improvement in depressive symptoms and perceived stress, with greater effects among those with stressors including food insecurity, discrimination, and COVID-19.

AJOG at a Glance:

A. Why was this study conducted?

1. To evaluate the impact of a novel language-concordant community-based program on perinatal mental health during the COVID-19 pandemic

2. To understand the impact of social determinants of health, including food insecurity and discrimination, on mental health before and after the program

3. To evaluate participant satisfaction with the program

B. What are the key findings?

1. The program was associated with improvement in depression, anxiety, and perceived stress as measured by validated survey instruments

2. There were high rates of food insecurity, discrimination, and COVID-19 infection in this urban cohort, and for those with these toxic stressors, the program was particularly beneficial

3. There was high program satisfaction

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

1. Community-based interventions such as HUGS/Abrazos that pair direct relief with behavioral health support may help alleviate growing disparities in mental health support in the peripartum period