1 INTRODUCTION

Healthcare inequities are preventable differences in the quality of healthcare and health outcomes in socially disadvantaged individuals, and these inequities can result in differences in healthcare access, disease occurrence, and healthcare outcomes [1]. Adverse perinatal outcomes affect birthing individuals and their children, with Black birthing individuals two to three times more likely to die from pregnancy complications compared to White birthing individuals [2]. Additionally, in the United States, children of Black, non-Hispanic women have a disproportionately high neonatal mortality rate of 6.85/1000 live births which is a nearly threefold increased risk of neonatal death compared to the national rate of 3.69/1000 live births [3]. Persistent inequities also exist despite education, and the pregnancy-related mortality ratio for Black birthing individuals with at least a college degree is five times as high as White birthing individuals with similar education [2]. Racial inequities in perinatal morbidity and mortality have been present in the United States for over 100 years and have links to structural and systemic racism impacting care [4].

The Military Health System (MHS) has an annual budget of $55 billion, and covers 89,000 births, with 75,000 births in the purchased care (civilian) market with Tricare insurance [5]. In the MHS, despite universal insurance coverage and stable employment, perinatal healthcare disparities and inequities exist, with Black pregnant individuals experiencing higher rates of cesarean delivery, admission to the intensive care unit, and maternal morbidity compared to White pregnant individuals [6]. Black, Asian Pacific Islander, and other races also experienced higher rates of cesarean delivery, postpartum hemorrhage, and severe maternal morbidity compared to White birthing individuals in the MHS [7]. Furthermore, causes of pregnancy-associated deaths in active duty individuals varied by race, with higher rates of deaths from other pregnancy-related causes, assault, and transportation accidents in Black birthing individuals [8]. Some studies have also identified healthcare inequities related to military rank (socioeconomic status), with higher military rank associated with shorter hospitalizations and lower in-hospital mortality in patients with ischemic stroke admissions [9]. The neonatal mortality rate within the MHS is lower than in the United States overall and is comparable to other developed nations. However, neonatal racial inequities also persist in the MHS, placing infants born to Black service members at a significantly increased risk of death in the neonatal period even when accounting for socioeconomic status and universal healthcare [3].

Despite persistent perinatal healthcare inequities in the MHS, prior studies have not explored the lived experiences of birthing individuals and how experiences such as trauma, military deployments, and other stressors may affect birth experiences and outcomes. The purpose of this study was to conduct a causal network analysis to understand the antecedent and mediating factors that contribute to healthcare inequities.

2 MATERIALS AND METHODS

Participants included individuals who delivered newborns within the past 5 years at a military treatment facility or civilian facility using TRICARE insurance in the United States and at worldwide facilities (both military and civilian). Additional inclusion criteria included age of 18 years or older and English speaking. Participants were recruited using social media posts, email blasts, snowball sampling, and flyers posted at military hospitals. Individuals were recruited with maximum variation and homogeneous sampling to represent diverse viewpoints based on socioeconomic status and self-identified race and ethnicity. Maximum variation was used to assess shared patterns between participants across a variety of perspectives, and homogeneous sampling was used to provide an in-depth description of participant groups [10]. Participants reviewed a study information sheet prior to making an informed, voluntary decision to participate.

An interprofessional group of clinicians and a qualitative researcher convened to conduct this study using a grounded theory approach [11] to understand participant's birthing experiences. The study team included physicians, residents, a midwife, and a qualitative researcher. Reflexivity statements are included in Supporting Information S1.

Semi-structured interviews were conducted from February 2023 to June 2023, and included questions related to participants’ birth experience, and prenatal and postpartum care. The data were part of a larger study focused on military beneficiaries’ birth experiences [12]. The interview guide was developed by the authors to investigate participants’ specific lived experiences during perinatal care and delivery to capture insights into potential healthcare inequities and is shown in Supporting Information S2. Interviews included open-ended questions about their pregnancy experience, childbirth, postpartum care, newborn care, personal and family history, and military experiences as applicable. Though participants were required to have a birth within the past 5 years to be eligible to participate, they were not excluded from talking about earlier birth experiences that they deemed important. The semi-structured interviews lasted between 45 and 90 min and were conducted with trained study investigators using Google Meet (Mountainview, CA). Participants gave verbal informed consent to participate in the study, including video recording and transcription using Google Meet software. All participants were free to end the interview early or decline to answer questions during the interview. At the end of the interview, participants were asked demographic questions including their age, self-reported race and ethnicity, active-duty service status, branch of service for themselves or their partner, rank for themselves or their partner, and travel time to their obstetric clinician and delivery location. Following the interview, participants were assessed for safety and provided counseling resources as needed. Participants were offered a $25 electronic gift card for their participation in the study. Interviews were conducted until thematic saturation was reached [13]. All interview transcripts were de-identified.

Each interview was coded by a trained investigator to assess causal themes, with causation coding and deductive methods used to generate a variable list of antecedent variables, mediating variables and outcomes. These variables were then organized into a flow chart to determine what generally leads to what, and then further organized into the final causation model [14]. Causation coding allows researchers to understand participants’ rationale for why things are the way they are from their perspective [14]. Causation coding is based in attribution theory, which includes the causal explanations produced by participants related to their experiences and perspectives, and allows the generation of plausible causes of outcomes [14, 15]. Coding was completed as interviews were conducted, with two authors coding each interview. Following completion of the first four interviews, the team reviewed the coding and made modifications to the interview guide to ensure participant understanding and depth of responses. To ensure consistency and accuracy of coding, group discussions were held at approximately 4-week intervals, and the interview guide was iteratively adjusted. Group discussions continued until consensus on coding was reached. Themes were organized into causal chains based on participant stories. A causal network was developed using cross-case mapping to generate a thematic narrative from a systematic comparison of within-case causal networks [14]. The steps of the causal network analysis are listed in Supporting Information S3. Initial coding and analysis was conducted, with attribution sequences plotted including antecedent conditions → mediating variables → outcomes. After initial categorization, similar outcomes were combined into major categories that led to specific outcomes, answering the ‘why’ question. The research team iteratively reviewed the causation coding to ensure accuracy, with cycles of codebook revision and iterative revisions of the causal network. Throughout the process the team reflected on their involvement as clinicians, military personnel, and mothers impacted their interpretation of participant narratives, with the team challenging these perspectives through discussion and finding opposing examples in the data. Interviews were conducted until saturation was reached with no new substantiative information obtained [10]. The study was approved by the Institutional Review Board of the Uniformed Services University of the Health Sciences and conforms with the U.S. Federal Policy for the Protection of Human Subjects.

3 RESULTS

A total of 77 individuals expressed initial interest in participation during the study period, with 36 individuals completing interviews, and 41 individuals declining participation or not following up to schedule an interview. Participant demographics are shown in Table 1. Most participants identified as non-Hispanic White, active duty servicemembers in the U.S. Navy. Care was received in both military and civilian care settings with some individuals experiencing care overseas, and some individuals experiencing care in both military and civilian settings, either due to transfer from one system to the other or in different pregnancies.

The resultant causal network is shown in Figure 1. The network can be organized into three main themes: (1) Delays in care (including delays in behavioral healthcare), (2) anxiety and stress, and (3) mistrust of the medical system. Additional causal networks centered around a behavioral health theme, which included (1) delays in care, (2) anxiety and stress, and (3) increased risk of relapse, as shown in Figure 2. Finally, a human milk feeding theme emerged with causal factors leading to success at human milk feeding and stopping early, as shown in Figure 3. Additional participant quotes are shown in Tables 2–5.

4 DISCUSSION

The unique experiences of birthing individuals in this study provide important insights into health and healthcare inequities in the MHS. Participants reported challenges related to care delays (Table 2), increased anxiety, stress, and mistrust (Table 3). Many themes and concerns centered around the ability of participants to access and utilize healthcare in the perinatal setting. Despite universal healthcare coverage, individuals report significant difficulties with referrals to the civilian network and availability of appointments in the MHS and civilian settings. The system to complete referrals was not easy to access and was especially challenging during military moves and when individuals needed care in both the direct care and purchased care markets. Improvements in the availability of appointments, ease of scheduling and referrals, and improved communication could decrease delays in care, decrease anxiety and stress, and decrease mistrust.

Current retrospective analyses of care and associated patient outcomes are limited by the fact that they only evaluate patients who sought and received care [16]. Individuals unable to schedule or keep appointments were not included and may have different experiences or outcomes. Thus, learning about individual experiences seeking and experiencing care provides an important perspective. Lack of appointment availability was commonly reported by participants, and improving MHS staffing shortages [17] remains a critically important step in improving access to care. Furthermore, half of active duty military installations are located within federally designated health professional shortage areas (HPSA) [18], and 35% of US counties are considered maternity care deserts [19], making it even more important to maintain military healthcare staffing in the setting of relative civilian staffing shortages. Recent reports highlight some of the ongoing staffing challenges, including data that 88% of Department of Defense facilities have nursing shortages [20].

Improving the systemic infrastructure to access appointments and increasing the availability of appointments (with improved staffing) in the MHS may also improve patient access to care. In a recent publication using the Joint Outpatient Experience Survey data from Army active duty servicemembers, the authors reported improvements in access to care with improved ease of scheduling appointments, the time between scheduling and actual visit, and wait time after scheduled appointment time [21]. The complexities of the referral and scheduling system in the MHS are also challenging for individuals scheduling perinatal and behavioral healthcare. Similar challenges have been reported in the Department of Veterans Affairs Health system, with comparisons to “finding your way through the labyrinth.” [22] Other opportunities to improve access to and completion of care include increasing patient navigators to assist patients with referrals and scheduling appointments and minimize inequities based on social factors such as age, education, and health. A recent study of access to care in the MHS using administrative data and patient satisfaction surveys reported that patient-reported access to care was influenced by gender, age, education, and self-reported health status, with increased age and good health status corresponding positively with improved access [23]. Potential solutions to improving access and satisfaction in the MHS are discussed in Tables 2–5, and include improving call center performance for appointment scheduling and medical questions [24].

Challenges with scheduling appointments and completing referrals in the MHS are further exacerbated by frequent moves required of military servicemembers, and variable command support to attend medical appointments, which are unique social determinants of health for military service members and their families. Improving command support for patients and improving the ability to schedule appointments during moves would be beneficial. The recent COVID-19 pandemic was another factor that contributed to delays in healthcare, and attention must be paid to maintaining care for routine services even in the setting of pandemics and other healthcare emergencies.

Participants reported that factors that decreased anxiety and stress included improved communication (Table 3), finding other individuals or peers who looked like them and faced similar struggles and supportive programs such as CenteringPregnancy group prenatal care and doula care. Increasing supportive care through support groups, doula access, and CenteringPregnancy are important components of improving care, and potentially mitigating racial inequities in the MHS. A recent randomized controlled trial of CenteringPregnancy group prenatal care found that increased participation in group prenatal care resulted in lower rates of preterm birth and low birthweight infants for Black individuals [25]. Similarly, use of doulas positively affects birthing outcomes including decreased need for analgesics, shorter labors, decreased cesarean delivery rates and improved patient satisfaction [26, 27]. Ultimately doula support may decrease health disparities and decrease costs for health systems [28], thus the MHS and Tricare could consider expanding current programs which offer six visits with a labor support visit. Increasing awareness of and use of support groups and doulas may further improve patient care for birthing individuals who are adapting to family and work changes following childbirth.

Participants also reported antecedent factors of microaggressions, differential treatment due to race, knowledge of poor outcomes for Black pregnant individuals and inadequate treatment of pain resulting in mistrust of the medical system. Such experiences then led individuals to question other aspects of care, or not to return for care with specific clinicians. Microaggressions stand out as an actionable item that healthcare workers can be aware of and address. Microaggressions are subtle, often unintentional, comments or actions that express bias or stereotypes, typically directed toward marginalized groups [29]. They may seem harmless, but can reinforce discrimination and make people feel excluded or devalued. These stigmatizing actions can create significant harm and mistrust in healthcare settings, and additional training and advocacy to decrease microaggressions in healthcare and society are important components of improving care. Addressing issues of systemic racism and microaggressions are important to improve trust and rebuild relationships in healthcare and mitigate health inequities. There is a biologic basis for the negative impacts of racism and microaggressions on health outcomes. The concept of “weathering” was developed by Dr. Arline Geronimus in the early 1990s [30]. She noted that societal racism can impact Black women resulting in “premature biological aging” ultimately leading to poor maternal health and pregnancy outcomes [30]. Allostatic load, or chronic stress, has further been shown to adversely contribute to the health of Black people [31]. Structural determinants of health (including structural racism, and aversive institutional policies) and social determinants of health such as food stability, education, environmental factors, and income, interact to influence Black maternal and neonatal outcomes [32]. Being Black itself isn't the issue. Black women exist at the intersection of race, gender and class, and this intersectionality of minoritized identities may also uniquely impact their well-being [32-34].

Despite not asking direct questions about behavioral health, multiple behavioral health themes emerged from participant narratives with concerns such as privacy and the impact of medications and counseling on their pregnancy and their careers [35]. This is not surprising as perinatal mental health conditions affect an estimated 10%–20% of birthing individuals in the United States, with perinatal anxiety affecting another 8%–20% [36, 37]. Participants cited privacy concerns and some admitted to not being honest on screening forms, thus it is imperative that providers consistently complete screening and follow up negative responses with reassurances that patients may always return for help in the perinatal period if their mood changes. When participants had supportive interactions with clinicians, they found behavioral health services and medications to be helpful.

Participant experiences related to human milk feeding varied, with some individuals reporting command support, access to lactation rooms and adherence to military policies supporting human milk feeding and pumping. However, others reported challenges including deployments and lack of schedule and facility accommodations that resulted in early cessation of human milk feeding. Increasing adherence to policies and improved support of human milk feeding in the workplace are important components to ensure success and decrease stress and anxiety. While TRICARE covers outpatient lactation services, many military facilities have infrequent and unpredictable lack of inpatient lactation services which likely contributes to active duty women having lower breastfeeding initiation rates than the national averages [38]. This responsibility for lactation support falls onto already overburdened inpatient nursing staff, where positions are reportedly filled at 68% rates with OBGYN nursing amongst the lowest filled positions [39]. Further enhancement and availability of lactation support services and lactation consultants both inpatient and outpatient would also be beneficial.

Furthermore, this study included individuals who received care in military and civilian settings, and some cases where individuals received care in both settings. Positive and negative experiences existed across care locations and there was not a clear difference between care received in these settings or other participant characteristics. Individual experiences were categorized to develop the reported causal networks and answer the “why” question: Why do persistent healthcare inequities exist in the military health system? The answer includes multiple causes and outcomes, and based on our analysis, we have developed plausible causes of outcomes and probable outcomes from the individual causes [14].

The strengths of this study include the important findings and detailed experiences of individual participants. Multiple opportunities exist to improve access to care and timely appointments and to improve clinician interactions and communication with patients, and additional support for unique social determinants of health, including frequent military moves. Although military members have stable employment, they must balance work commitments and deployments, which may require additional support. The military health system covers 89,000 annual births, with approximately 75,000 births in the purchased care (civilian) market with Tricare insurance [5], and in this study, 47% of participants received military care, 33% received civilian care, and 13.9% received both military and civilian care, thus the challenges and experiences are relevant for both military and civilian obstetric clinicians. Another strength was the use of qualitative methodology, which includes investigators’ positionality, where personal backgrounds shape the research lens and findings. Throughout the study we reflected on our involvement as clinicians, military personnel, and mothers, and how this influenced the study. We engaged in critical self-reflection and actively sought diverse perspectives from participants to increase our knowledge in this area. Finally, we report on the lived experiences of 36 participants, with an in-depth and rich understanding of their individual experiences. These detailed experiences provide transferability [40] to other patients.

The limitations of this study include the possibility of different experiences in others and the possibility that other social determinants impacting participant health experiences were not discussed in our interviews. Another limitation was the use of multiple interviewers for this study, which was done to facilitate feasibility, and all interviewers received training and used an interview guide to standardize interviews. The interview guide (Supporting Information S2) was designed to assess overall participant birth experiences and did not specifically ask questions related to behavioral health, thus participant answers may have been different if specifically asked about behavioral health.

5 CONCLUSIONS

In summary, opportunities for improvements include addressing staffing shortages and availability of appointments, improved infrastructure and support for scheduling and referrals, improved communication, addressing systemic racism and microaggressions, encouraging command support, and maintaining the privacy of patients. Future research should investigate the effects of such improvements on patient experiences and outcomes in perinatal care.

AUTHOR CONTRIBUTIONS

All authors conducted interviews that were included in this study. All authors contributed to manuscript drafting and editing. All authors read, edited, and approved the final manuscript.

CONFLICT OF INTEREST STATEMENT

Some authors are federal employees or members of the armed forces. The views expressed are those of the authors and do not reflect the official policy or position of the U.S. Army Medical Department, Departments of the Army, Navy and Air Force, Defense Health Agency, Department of Defense, the Uniformed Services University of the Health Sciences, or the U.S. Government. The investigators have adhered to the policies for protection of human subjects as prescribed in 45 CFR 46.

Several of the authors are military service members. This work was prepared as part of their official duties. Title 17 U.S.C. 105 provides that ‘Copyright protection under this title is not available for any work of the United States Government.’ Title 17 U.S.C. 101 defines a United States Government work as a work prepared by a military service member or employee of the United States Government as part of that person's official duties.