Racial disparities in emergency mental healthcare utilization among birthing people with preterm infants

BACKGROUND: Birthing people of color are more likely to deliver low birthweight and preterm infants, populations at significant risk of morbidity and mortality. Birthing people of color are also at higher risk for mental health conditions and emergency mental healthcare utilization postpartum. Although this group has been identified as high risk in these contexts, it is not known whether racial and ethnic disparities exist in mental healthcare utilization among birthing people who have delivered preterm. OBJECTIVE: We sought to determine if racial and ethnic disparities exist in postpartum mental healthcare-associated emergency department visits or hospitalizations for birthing people with preterm infants in a large and diverse population. STUDY DESIGN: This population-based historic cohort study used a sample of Californian live-born infants born between 2011 and 2017 with linked birth certificates and emergency department visit and hospital admission records from the California Statewide Health Planning and Development database. The sample was restricted to preterm infants (<37 weeks’ gestation). Self-reported race and ethnicity groups included Hispanic, non-Hispanic Black, non-Hispanic Asian, non-Hispanic White, and non-Hispanic others. Mental health diagnoses were identified from the International Classification of Diseases Ninth and Tenth revision codes recorded in emergency department and hospital discharge records. Logistic regression analysis was used to estimate the association between mental health-related emergency department visits and rehospitalizations by race or ethnicity compared with non-Hispanic White birthing people and controlling for the following characteristics and health condition covariates: age, parity, previous preterm birth, body mass index, smoking, alcohol use, hypertension, diabetes, previous mental health diagnosis, and prenatal care. RESULTS: Of 204,539 birthing people who delivered preterm infants in California, 1982 visited the emergency department and 836 were hospitalized in the first year after preterm birth for a mental health-related illness. Black birthing people were more likely to have a mental health-related emergency department visit and hospitalization (risk ratio, 1.8; 95% confidence interval, 1.5–2.0 and risk ratio, 1.9; 95% confidence interval, 1.5–2.3, respectively) within the first postpartum year than White birthing people. Hispanic and Asian birthing people were less likely to have mental health-related emergency department visits (adjusted risk ratio, 0.7; 95% confidence interval, 0.7–0.8 and adjusted risk ratio, 0.2; 95% confidence interval, 0.2–0.3, respectively) and hospitalizations (adjusted risk ratio, 0.6; 95% confidence interval, 0.5–0.7 and adjusted risk ratio, 0.2; 95% confidence interval, 0.1–0.3, respectively). When controlling for birthing people with a previous mental health diagnosis and those without, the disparities remained the same. CONCLUSION: Racial and ethnic disparities exist in emergency mental healthcare escalation among birthing people who have delivered preterm infants. Our findings highlight a need for further investigation into disparate mental health conditions, exacerbations, access to care, and targeted hospital and legislative policies to prevent emergency mental healthcare escalation and reduce disparities.


Introduction
A ntenatal and postpartum mental health conditions such as anxiety and depression are prevalent, estimates range from 10% to 25% 1-3 and have important short-and long-term health consequences for birthing people (a term that recognizes that not all people giving birth identify as women) and their infants. [4][5][6] For example, depression has been associated with the development of adverse health outcomes and behaviors during pregnancy, including inadequate nutrition, weight gain, and substance use, and adverse perinatal outcomes like preterm birth (PTB) and low birthweight. 5,7,8 Postnatally, depression among birthing people has been associated with adverse child hood outcomes including impaired growth, behavior, and cognitive development. 9 Thus, the prevention, identification, and treatment of mental health conditions during and after pregnancy is critical for both birthing people and infant health. Several risk factors for developing postpartum mental health conditions have been identified, including lower socioeconomic status, history of domestic violence, history of mental health conditions, lack of partner or social support, pregnancy complications, pregnancy loss, and poor infant health. 3,[10][11][12] Birthing people with preterm infants are at higher risk for anxiety, depression, and stress than birthing people who give birth to term infants. 12 Birthing people have described unique stressors during neonatal intensive care unit (NICU) hospitalizations including barriers to routine parenting and bonding experiences like holding, protecting, and caring for their infant and have described subsequent posttraumatic stress disorder (PTSD)-like symptoms after NICU hospitalizations. 13 In addition, giving birth prematurely is a stressful life event that can exacerbate mental health conditions and has been shown to be associated with inpatient mental healthcare utilization postpartum including both emergency department (ED) visits and hospitalizations. 14 Several studies have shown that non-White birthing people and those with low income are at higher risk for depression and anxiety during and after pregnancy. 2,[15][16][17][18][19][20][21] Racial and ethnic disparities have also been described for both outpatient and inpatient postpartum mental healthcare utilization. 22,23 A study in New Jersey found that Black and Hispanic birthing people were less likely to receive outpatient mental healthcare treatment after delivery, had a longer time from identification to treatment initiation, were less likely to receive continued mental healthcare, and less likely to fill prescriptions for antidepressant medication. 22 Disparities also exist in potentially avoidable, costly, and morbid inpatient hospitalizations associated with mental health conditions. A study in California found that compared with White birthing people, Black birthing people were at increased risk for receiving postpartum hospital-based mental healthcare. 24 Racial and ethnic disparities associated with PTB are well described; Black birthing people are more likely to deliver both preterm and low birth-weight infants. 25 Disparities for preterm infants continue to persist after hospital discharge; Black infants are more likely to be readmitted and die in their first year of life. 26 Increasingly, structural racism, discrimination, systemic oppression, and social disadvantage are being recognized as sources of chronic stress and poor reproductive health outcomes for Black birthing people. [27][28][29][30] Although previous work has examined disparities in ED and hospital utilization after giving birth to both term and preterm infants, 24 we specifically chose to focus on preterm infants, because PTB and its complications have also been associated with poor mental health and increased mental healthcare utilization. 14,24 In addition, although previous work has focused more specifically on postpartum depression, 24 we elected in this study to examine all mental health conditions including but not limited to mood disorders (depression, bipolar disorder), anxiety related disorders (posttraumatic stress disorder, generalized anxiety disorder), and schizophrenia. Because birthing people of color are a unique population at high risk for PTB, poor infant outcomes, mental health conditions, and mental healthcare utilization, we hypothesized racial and ethnic disparities also exist in mental health-related ED visits and hospitalizations among a population of birthing people who delivered infants preterm. admission records obtained from the California Office of Statewide Health Planning and Development database. The database includes information on infant and birthing people demographics, health conditions, and healthcare utilization up to 1 year postpartum derived from ED and hospital discharge records and birth certificates. Discharge records include diagnosis codes based on the International Classification of Diseases, Ninth and Tenth Revision, Clinical Modification (ICD-9-CM, ICD-10-CM).

Sample
The sample was restricted to birthing people who delivered singleton, live-born preterm infants between 22 and 36 weeks' gestation without significant congenital anomalies.The obstetrical age was derived from the gestational age (GA) indicated in the birth certificate based on ultrasound or the last menstrual period.

Race and ethnicity
Self-reported birthing person race and ethnicity were organized into the following groups: non-Hispanic White (referred to as White), non-Hispanic Black (Black), Hispanic, and non-Hispanic Asian (Asian). "Other" race and ethnicity included American Indian or Alaska Native, Hawaiian or Pacific Islander, other race, >1 race, and those with not stated or unknown race and ethnicity because of sample size constraints among these groups.

Covariates
We chose covariates known to be associated with PTB and/or mental health conditions during pregnancy, including maternal age at term (<18 years, 18-34 years, >34 years), parity (multiparous or nulliparous), previous PTB, body mass index (BMI) (underweight, normal, overweight, obese, unknown), smoking and/or drug/alcohol abuse during pregnancy, gestational hypertension (HTN), gestational diabetes mellitus (DM), adequate prenatal care, 31 previous mental health diagnosis, GA, birthweight at GA, infant death, and payer for delivery. Covariate definitions and data sources are listed in Supplemental Table B.

Mental healthcare utilization
Data on previous mental health conditions and postpartum emergency mental healthcare utilization were obtained from linked hospital discharge records. To identify a birthing person's mental health diagnoses, relevant ICD-9 and ICD-10 codes were used, including but not limited to mood disorders like depression and bipolar disorder, stress-related disorders like anxiety, behavioral syndrome disorders caused by psychoactive substance use, schizophrenia disorders, and personality disorders (see Supplemental Appendix A). The 4 primary outcomes were mental health-related ED visits and rehospitalizations at 3 months and 1 year postpartum.

Analysis
We used a logistic regression to test the association of mental health-related ED visits and rehospitalizations with race or ethnicity compared with the referent White birthing person group, chosen as the reference group because of lower rates of mental health utilization in this population based on previous literature, and controlling for birthing people characteristics and health condition covariates previously described by using the following 3 models: model 1 was a crude, unadjusted model. Model 2 adjusted for birthing people characteristics and health condition covariates other than a previous mental health diagnosis. Model 3 adjusted for a previous mental health diagnosis in addition to the covariates in model 2. We stratified preterm infants according to GA groups of <32 weeks and 32 to 36 weeks, populations with significant differences in morbidity and mortality. Methods and protocols for the study were approved by the Committee for the Protection of Human Subjects of the Health and Human Services Agency of the State of California and the institutional review board of the University of California San Francisco.

Results
Of the 3,448,707 live births in California from 2011 to 2017, 204,539 were preterm singleton infants for whom linked birthing people and infant hospital and discharge records were available ( Figure 1). 30 This sample was 51% Hispanic, 22% non-Hispanic White, 7.3% non-Hispanic Black, 14% non-Hispanic Asian, and 5.5% non-Hispanic other race or ethnicity. Most birthing people were 18 to 34 years old, multiparous, and had adequate prenatal care. Overall, 9.9% of this sample had a previous mental health diagnosis. A total of 12.8% of birthing people delivered infants at <32 weeks GA and 87.2% delivered at 32 to 36 weeks GA. A variety of types of mental health-associated visits were represented within the sample, listed in descending order of frequency as follows: anxiety, mental and behavioral disorders caused by psychoactive substance use, depression, schizophrenia spectrum disorders, other behavioral disorders, and personality disorders (Table 1). Of the birthing people who delivered preterm infants in California, 1089 (0.5%) and 1982 (1.0%) visited the ED for a mental health-related visit within the first 3 months and 1 year after PTB, respectively. Regarding mental health-related hospitalizations, 365 (0.2%) and 836 (0.4%) birthing people were hospitalized in the first 3 months and 1 year after giving birth prematurely ( Table 2).
Despite representing only 7.3% of the population in this sample, Black birthing people represented 15.6% of the mental health ED visits in the first year after delivery and 18.1% of the mental health hospitalizations in the first year after delivery. Hispanic and Asian birthing people made up a smaller proportion of ED visits (45.9% and 3.7%, respectively) and hospitalizations (40.3% and 3.4%, respectively) than the representative sample proportion (51.1% and 14.2%, respectively). The proportion of White birthing people with reported ED visits (27.2%) and hospitalizations (29.7%) was more than their sample representation (22.1%) ( Table 2).
Hispanic birthing people were less likely to have a mental health-related ED visit in the first 3 months and 1 year after birth and less likely to have a mental health-related hospitalization in the first 3 months after birth than White birthing people but this likelihood lost statistical significance after controlling for a previous mental health diagnosis (model 2: aRR, 0.7; 95% CI, 0.6-0.9; aRR, 0.8; 95% CI, 0.7-0.9; aRR, 0.7; 95% CI, 0.5-0.9, respectively). Hispanic birthing people were also less likely to have a mental health-related hospitalization 1 year postpartum, which persisted after controlling for covariates (model 3: aRR, 0.8; 95% CI, 0.7-1.0) ( Table 2).
When stratified by GA groups of <32 and 32 to 36 weeks, disparities were narrowed in the <32 weeks group and remained in 32 to 36 weeks group (Supplemental Table 1 and Supplemental Table 2).

Discussion
Our findings add to literature describing racial and ethnic disparities in postpartum mental healthcare utilization by highlighting the disparities after preterm delivery. In our study, Black birthing people who have delivered preterm infants are at higher risk for both mental health-related ED visits and rehospitalizations than White birthing people in California. These findings persisted despite controlling for factors known to be associated with mental health conditions including previous mental and physical health conditions and social factors.
A previous study by Chan et al 24 in which the same database of Californian infants was used also demonstrated racial and ethnic disparities in hospital-based care for postpartum depression. We thought it prudent to expand on this work by investigating whether disparities also exist for birthing people of color who have delivered infants prematurely, because PTB and its complications also have been associated with poor mental health and increased mental healthcare utilization. 14,24 The Chan et al 24 study was restricted to birthing people with postpartum depression, whereas our study considered other important mental health conditions, which require a different approach to diagnosis and treatment, faced by birthing people like anxiety, bipolar disorder, schizophrenia, behavioral conditions, personality disorders, and not otherwise specified mental health conditions for a more complete understanding of mental health and severe mental healthcare utilization in this population. In addition, we controlled for infant characteristics like degree of prematurity (<32 vs 32-36 weeks' gestation) and infant birthweight among preterm infants because prematurity and birthweight have been identified previously as risk factors for mental health conditions. 12 Families with infants requiring prolonged hospitalization to treat complications of PTB face unique and magnified challenges when compared with those with an uncomplicated postpartum course. PTB may exacerbate postpartum mental health symptoms and existing disparities in postpartum mental healthcare in several ways. [12][13][14] For example, preterm infants are at high risk for major preterm comorbidities, unexpected complications requiring procedures and/or surgeries, compromised neurodevelopment impairment, technology dependence and disability, and mortality. 25 Long hospitalizations and complications of PTB can contribute to financial hardships for families.
Our study also found that when compared with White birthing people, Hispanic and Asian birthing people had a lower risk for mental health-related healthcare utilization after controlling for several covariables, which could reflect underutilization of needed mental health resources or a lesser burden and exacerbation of mental health conditions. Our study cannot make this differentiation and further investigation is necessary. We remain concerned that minority populations may have underdiagnosed and undertreated mental health conditions owing to language barriers, disparate healthcare access, and cultural differences as indicated by previous literature. [22][23][24] When stratifying by a GA of <32 weeks, racial disparities narrowed for all groups except for mental health-related hospitalizations for Black birthing people 1 year postpartum, in which case the risk worsened. Disparities persisted in the larger 32-to 36-week GA group. Although there were fewer significant findings in the analysis of infants born <32 weeks, the point estimates were similar to the combined analysis for all preterm infants. We suspect that part of the reason the results became insignificant in this group is that sample sizes for healthcare events became very small. These findings are consistent with previous literature indicating PTB as a risk factor for healthcare utilization 14 and suggest the need for mental health support for families with preterm infants of all GA.
Although our study does not provide insight into potential interventions to reduce these disparities, we felt it prudent to provide evidence in literature to direct future considerations for investigation and action. Hospital-level interventions that facilitate family-centered care may prevent mental health exacerbations, because improved parent participation during NICU hospitalization and early bonding opportunities are associated with reduced mental health symptoms. [32][33][34][35] Hospital system changes to include free parking, free hospital transportation, and food vouchers for families could alleviate barriers to visitation that structural racism and an unequal distribution of resources, goods, and services secondary to present and historic racism may exacerbate for families of color. 36,37 Reducing the financial burden by guaranteed and extended maternity leave for hospitalized family members, similar to Canadian policies, could mitigate stressors. [37][38][39] healthcare access. 40,41 Routine screening for early intervention of postpartum depression is recommended, however, a recent study found that birthing people of color are less likely to be screened. 42 Because outpatient referrals often do not lead to sustained outpatient therapy, [43][44][45] initiating counseling by mental health providers at the time of birth has been shown to have beneficial results and addresses potential barriers to care such as time availability, stigma, and childcare issues. Recognizing the birth hospitalization as an early intervention access point may lead to improved access, early diagnosis, and treatment.
The American College of Obstetricians and Gynecologists has released a statement that advocates for the optimization of postpartum care with ongoing comprehensive care after birth, including supporting psychological well-being, via regular visits with obstetrical providers. Optimization of postpartum care will require legislative policies that improve the access and quality of community-based mental healthcare and advocate for expanding health insurance coverage postpartum, and interventions that promote social support and reduce financial burden may reduce the impact of PTB on families. 37,38,45 Limitations Although this dataset is large and mental health conditions in pregnancy are common, mental health-related ED visits and hospitalizations are relatively rare events when compared with other reasons for postpartum healthcare utilization, limiting our power to detect significant differences. In addition, because this is a retrospective observational study, we can only deduce association between variables, not causality. Disparate healthcare utilization can reflect disparate access to preventative, outpatient, and/or emergency care or disparate exposures and exacerbations of mental health conditions. Our study is unable to distinguish between the 2 scenarios. This is an important point to distinguish because it impacts where resources would better be spent to reduce these disparities. Important, yet unavailable data included mental health screening rates, records on access to and participation in outpatient mental healthcare, medication access and use, and stressors that may impact mental health including racism, discrimination, food security, social support, major life events, and those lost to follow-up. Similarly, although we suspect that the observed disparate access to resources by race in our dataset is impacted by the historic and present structural racism contributing to unequal access to resources and opportunities for families of color that influences healthcare and outcomes, 30 we did not have a structural racism metric. Lastly, detailed subcategories to reflect the heterogeneous nature of racial, ethnic, and ancestral groups were not available.

Conclusion
Racial disparities exist in emergency postpartum mental healthcare utilization for birthing people with preterm infants, with Black birthing people at highest risk for mental healthrelated hospitalization and ED visits. To achieve health equity for all birthing people, governments and health systems must critically examine the factors that exacerbate this disparity and implement multifaceted interventions.

Why was this study conducted?
Birthing people of color experience higher rates of postpartum mental health conditions, preterm birth, poor infant outcomes, and mental healthcare utilization. Preterm birthassociated hospitalizations are an additional contributor to poor mental health during the postpartum period. It is not known whether racial and ethnic disparities exist in emergency mental healthcare utilization by birthing people who deliver preterm infants.

Key findings
In this large and diverse population-based cohort, racial and ethnic disparities existed in mental healthcare utilization after preterm birth, with Black birthing people more likely to have a mental health-related emergency department visit or hospitalization 1 year postpartum (relative risk [RR], 1.8; 95% confidence interval [CI], 1.5-2.0; RR, 1.9; 95% CI, 1.5-2.3) compared with White birthing people.

What does this add to what is known?
This study highlights the racial and ethnic disparities in emergency mental healthcare escalation in birthing people who have delivered preterm infants. Targeted interventions in high-risk groups may promote equitable outcomes.   Risk ratios of mental healthcare utilization by race and ethnicity, <37 weeks