Abstract
Purpose: Community health centers (CHCs) provide critical health care access for people who experience high risks during and after pregnancy, however it is unclear to what extent they provide prenatal care. This study seeks to describe clinic and patient characteristics associated with longitudinal prenatal care delivery in CHC settings.
Methods: This retrospective cohort study utilized electronic health record (EHR) data from the Accelerating Data Value Across a National Community Health Center Network (ADVANCE) between 2018 to 2019 to describe prenatal care provision among CHCs (n = 408), and pregnant CHC patients (n = 28,578) and compared characteristics of patients who received longitudinal prenatal care at CHCs versus those who did not.
Results: 41% of CHCs provided longitudinal prenatal care; these CHCs were more likely to be larger, have multidisciplinary teams, and serve higher proportions of nonwhite or non-English speaking patients. Patients who received longitudinal prenatal care at CHCs were racially and ethnically diverse and many had comorbidities. Patients who received longitudinal prenatal care at CHCs (compared with pregnant patients who did not) were more likely to be white or Latinx and more likely to have non-English language preference.
Conclusions: Many CHCs in this national network provide prenatal care and serve pregnant patients at high risk of pregnancy-related complications, including people of color, those with low income, and those with comorbidities. CHCs provide critical access to care for vulnerable populations and will be an important partner in work addressing inequities in maternal morbidity and mortality.
- Data Set
- Family Medicine
- Health Services Accessibility
- Newborns
- Obstetrics
- Postpartum
- Pregnancy
- Prenatal
- Retrospective Studies
- Vulnerable Populations
Introduction
Maternal morbidity and mortality in the US are high and racial and economic disparities in pregnancy outcomes are rising.1⇓⇓–4 Healthy People 2030 calls for decreasing maternal morbidity and mortality by promoting high-quality interdisciplinary preventive and prenatal care for people of reproductive age.5⇓⇓–8 Currently, nearly 15% of pregnant patients in the US receive inadequate prenatal care and 11 to 51% of patients do not receive postpartum care.9⇓⇓⇓⇓⇓⇓–16 Understanding the settings in which pregnancy care is received, particularly among individuals with higher risk of adverse pregnancy outcomes, is critical to addressing maternal morbidity and mortality and closing existing disparities.
Community health centers (CHCs) – Federally Qualified Health Centers (FQHCs), Rural Health Clinics, and FHQC ‘look alike’ clinics – are a cornerstone of primary care access in the US, delivering health care services across the lifespan.17⇓⇓–20 CHCs provide access to care regardless of ability to pay or health insurance status and tend to serve a racially and ethnically diverse population that has a high prevalence of low income and uninsurance.18 People of reproductive potential comprise a large proportion of the CHC patient population, and even before the Affordable Care Act expansions, CHCs provided care to nearly one in three low-income reproductive-aged females in the US.20⇓⇓–23 Compared with other primary care clinic settings, CHCs tend to care for more individuals who are nonwhite, who smoke and use other substances, and who have multiple chronic medical and mental health conditions,18 all risk factors for severe maternal morbidity.24
The majority of CHCs are staffed by family medicine physicians18 who are trained to provide care across the lifespan including preconception, prenatal, postpartum, and contraceptive care. Despite this training, decreasing proportions of family physicians provide obstetric care25⇓⇓⇓–29 and it is unclear to what extent CHCs provide pregnancy care. The Uniform Data System (UDS) provides a limited view of prenatal care provision among Federally Qualified Health Centers (FQHCs), reporting only patients’ age and trimester of prenatal care initiation.30 Electronic health record (EHR) data provide a unique opportunity to use real-time clinical care information to describe prenatal care delivery in CHCs, the characteristics of CHCs that provide longitudinal prenatal care, and characteristics of the patients receiving longitudinal care in CHC settings across each of their pregnancies.
A better understanding of prenatal care provision in CHCs, including clinical measures such as longitudinal or episodic care, and patient-level demographics as well as health status and prenatal care during each completed pregnancy, could help better align care with patient needs and focus future interventions to reduce disparities in maternal morbidity and mortality.
This study aims to describe the extent to which US CHCs provide prenatal care, and identify characteristics of clinics providing this care and the patients receiving this care during their pregnancies. This information is critically needed to consider the role of CHCs in addressing disparities in maternal morbidity and mortality.
Methods
Data Source and Analytic Sample
We used electronic health record (EHR) data to conduct a retrospective cohort study using individual patient-level data from the Accelerating Data Value Across a National Community Health Center Network (ADVANCE) Clinical Research Network, a member of Patient-Centered Outcome Research Network (PCORnet).31,32 ADVANCE is a multi-center, CHC-focused collaborative led by OCHIN (not an acronym) in partnership with Health Choice Network (HCN), Fenway Health, and Oregon Health & Science University.31,33 ADVANCE is demographically and clinically similar to CHC patients nationally.18,31 In this analysis, we limited inclusion to OCHIN and HCN networks because of the completeness of their reproductive health data. Health centers were included if they were ‘live’ on the EHR throughout the study period (January 1 2018 – December 31 2019), and provided primary care to women of reproductive age (age 15 to 44). With a time frame before the COVID-19 pandemic, this study provides a glimpse into care before pandemic-related disruptions ensued. This study was approved by Western Institutional Review Board.
Outcome Variables
The primary outcome of interest was clinic-level provision of longitudinal prenatal care, defined as providing more than 1 prenatal visit (confirmed by CPT codes) during a single pregnancy. These prenatal visit codes specify provision of pregnancy-related ambulatory care (see Online Appendix for codes utilized). First, we categorized clinics (as the unit of analysis) as either providing longitudinal prenatal care (delivering 2 or more prenatal visits per pregnancy), providing occasional prenatal care (delivering only 1 prenatal visit per pregnancy for all observed pregnancies) or as not providing prenatal care (no prenatal care visits observed). In the ADVANCE Network, a “clinic” typically represents a unique delivery location that may or may not be affiliated with a larger health system.
Second, we assessed patient characteristics among females during each completed pregnancy in the study period (pregnancy episode at the unit of analysis). Pregnancies were identified using indicators established by the Office of Population Affairs34 and supplemented by EHR laboratory data (eg, positive pregnancy tests), and indicator(s) of a live birth (documented delivery date, presence of a postpartum ICD-10 code, or presence of a postpartum visit). Pregnancies without any indication of live birth were excluded because the need for prenatal care would have been uncertain. For patients with more than one pregnancy during the study period, we defined the second pregnancy as one occurring at least 10 months after the end date of the preceding pregnancy. Detailed descriptions of these definitions are included in the Online appendix.
Other Variables
Additional variables were selected based on prior studies of clinic- and patient-level factors that impact primary care delivery.35⇓⇓–38 Clinic-level factors included the size of the clinic (total number of unique patients and total number provided visits), scope of practice (eg, care of children), team structure (eg, presence of a multidisciplinary team), participation in the Title X program,37 clinic rurality based on ZIP code of address,38 geographical region of the US determined by US Census Bureau, and aggregate patient population characteristics (eg, proportion reproductive aged women, proportion low-income, etc.). At the patient level, we used EHR data including ICD-10 codes to describe demographic characteristics (age, race, ethnicity, language preference, income, and insurance status), health before pregnancy (presence of any medical, mental health, and substance use disorders, and multiple morbidity using an updated version of the Charlson Comorbidity Index adapted for ambulatory care),39 and health during pregnancy (gestational diabetes, hypertensive disorders of pregnancy). A detailed list of ICD-10 codes used and definitions applied is included in the Online Appendix.
The primary purpose of this study was to describe the characteristics of CHCs that provided longitudinal prenatal care and describe the characteristics the pregnancies for which they provided longitudinal care. However, because natural comparisons arise between the 3 clinic groups (longitudinal vs occasional vs none) and between the 2 pregnancy groups (presence or absence of longitudinal prenatal care at the CHC) we opted to provide some unadjusted descriptive statistics to help identify between-group differences that may be important for future consideration. We conducted Chi-squared tests to compare differences between pregnancy-provision groups. Analyses were conducted in SAS, version 8.3.
Results
Findings Among Clinics
In our sample of 408 clinics across 21 states, 41% (n = 168) delivered longitudinal prenatal care, 53 clinics (13%) provided occasional pregnancy-specific care but did not have repeated prenatal visits, and the remaining 187 clinics (45%) did not provide pregnancy-specific care during the study period (Table 1). On average, clinics that provided longitudinal prenatal care were larger than clinics that did not (mean, 4402 active patients per year vs 1370 active patients per year), although their populations included similar proportions of female patients of reproductive age (13.4% vs 14.7%, P < .001). Clinics providing longitudinal pregnancy care were more likely than other clinics to have multidisciplinary teams and participate in the Title X program. Clinics providing longitudinal prenatal care were more likely to be from the western US (54.2% vs 47.1% overall) and less likely to be in rural areas (3.6% rural vs 6.4% nonrural), P = .014). Though there were statistically significant between-group demographic differences in patient populations, the magnitudes were small and lacked consistency. All clinics provided care to diverse populations (23.2% nonwhite, 28.2% Latinx, 22.1% non-English language preference). Compared with clinics that provided no pregnancy care, clinics providing longitudinal prenatal care served fewer patients with Medicaid (19.2% vs 25.2%, P < .001) and without health insurance (16.9% vs 27.5%, P < .001).
Characteristics of Community Health Centers (CHCs) by Provision of Prenatal Care, 2018-2019
Findings Among Pregnancies
There were 28,578 pregnancies among 28,064 included patients; 514 patients had 2 observed pregnancies in the study time frame. Most pregnancies identified within the EHR received longitudinal prenatal care at CHC clinics (92%), however 8% of pregnancies did not receive longitudinal prenatal care at included CHCs. Pregnant patients who received longitudinal prenatal care from CHCs received an average of 10 prenatal visits and 2 postpartum visits per pregnancy. Patients who did not receive prenatal care at CHCs were still likely to receive postpartum care at CHCs with an average of 1.6 postpartum visits in this group (Table 2).
Characteristics of Pregnant Patients at Community Health Centers (CHCs) During Each Completed Pregnancy*, 2018-2019
Pregnant patients who received longitudinal prenatal care at CHCs were more likely to be white (59.1% vs 50.4%), Latinx (52.6% vs 32.8%), have Spanish language preference (35.5% vs 18.3%), and have private health insurance (13.1% vs 3.7%) compared with women who did not receive longitudinal prenatal care at CHCs. Overall, many pregnant CHC patients experienced chronic conditions before pregnancy including hypertension (1.8%), diabetes (0.8%) mental health diagnoses (13.6%), and substance use disorders (tobacco 2.6%, other substances 3.4%). Aside from small increased prevalence of diabetes and gestational diabetes among pregnancies receiving longitudinal prenatal CHC care, there were not significant differences in health status (before or during pregnancy) between pregnancies receiving longitudinal CHC clinic care versus those receiving occasional or no prenatal care.
We performed a sensitivity analysis which identified only minor demographic differences between patients with 1 observed pregnancy and those with 2 pregnancies (Online Appendix Table 4).
Discussion
The majority of CHCs in our large national network (54%) provide prenatal care, and most of these clinics (76%) provide longitudinal prenatal care. Similar to the population of CHC patients overall, pregnant CHC patients are racially and ethnically diverse, and experience comorbidities and psychosocial challenges. Pregnant CHC patients also tend to be at high risk of adverse pregnancy-related outcomes which disproportionately impact people of color, people with fewer resources, and people with comorbid conditions.1,40⇓⇓⇓–44 As disparities in pregnancy outcomes continue to widen, CHCs are well-positioned to provide critically-needed interventions through their strong relationships with this high-risk population.20,45
The fact that CHCs providing longitudinal prenatal care in our network tended to be larger, more likely to have a multidisciplinary team, more likely to receive Title X funding, and have fewer patients with Medicaid or no health insurance, suggests that clinics providing longitudinal prenatal care had access to more well-developed resources overall. Title X provides reimbursement to clinics for delivery of reproductive health care, but does not provide coverage for prenatal care so finding a higher proportion of Title X recipients among CHCs providing longitudinal prenatal care demonstrates an important overlap between clinics providing reproductive care outside of pregnancy and those providing prenatal care. Similarly, clinics providing longitudinal prenatal care were more likely to care for children, suggesting more comprehensiveness overall at these clinics.
Many patients who received longitudinal prenatal care at CHCs in this study had comorbid mental health or substance use conditions. This high prevalence is similar to previously published averages in CHC populations and is higher than that in the general public.18,46⇓–48 Many CHCs in this study provided a multidisciplinary team, consistent with national efforts to expand multidisciplinary primary care through positions like resource specialists, care managers, health educators, behavioral health teams and more.18,49 These services may be particularly useful in caring for pregnant patients with complex medical and social needs and has previously been associated with better health outcomes.36 These resources may also be important for improving screening and treatment for postpartum depression, which is a Healthy People 2030 initiative goal.5
The average number of prenatal visits (among patients receiving longitudinal prenatal care at CHCs) was 10, which is similar to the US national average number of prenatal visits (11) and fewer than the number recommended by the American College of Obstetrics and Gynecology (ACOG) at the time of the study (12 to 14).50 The relationship between number of prenatal visits and care quality has been debated and new guidelines were published in 2021 and endorsed by ACOG and others.51 However, these new guidelines are based on expert opinion and still require rigorous validation to determine how they relate to health outcomes, particularly among higher risk populations. This information will be important to a more complete assessment of pregnancy care delivery at CHCs.
Of physician staff at CHCs nationally, 46% are family physicians18 who are uniquely trained to provide comprehensive longitudinal care for pregnant people – including care for medical comorbidities, mental health issues, and substance use disorders as well as primary prenatal, postpartum and pediatric care. Provision of pregnancy care among family physicians has declined substantially, and in 2012 only 9% of family physicians reported providing these services. Though we did not measure provision of pregnancy care at the individual physician level, our findings suggest that family physicians who practice in CHC settings may be more likely than other family physicians to provide pregnancy care. As family medicine specialty groups consider strategies to ensure a continued pipeline of family physician staff to our nation’s CHCs, they should consider the importance of skills in pregnancy care for this important practice setting. As policy makers strive to improve maternal morbidity, mortality, and infant health, adequate support and staffing to CHCs should be a high priority.
Though the majority of CHCs in this study provided longitudinal pregnancy care, 46% of CHCs did not provide this service. Describing the reasons for this were beyond the scope of this study, but merit further research, particularly since CHCs tend to serve proportionally more patients of reproductive potential and people for whom pregnancy may pose health risks.18 Future research is also needed to assess the quality and effectiveness of pregnancy care delivered at CHCs to determine the best ways to improve pregnancy outcomes in this population.
Limitations
The use of EHR data were a strength of this study as it is free from sampling and recall biases that affect survey studies. Still, ambulatory EHR data provide a number of challenges in measuring pregnancy outcomes. Because pregnancy start-dates were sometimes extrapolated or estimated from other data (such as postpartum care) and precise delivery dates were unknown, we were unable to accurately quantify trimester of presentation to care. Because comprehensive delivery information such as birth outcome, is not reliably available within the EHR, assessment of these outcomes was beyond the scope of this study. Though ICD-10 codes are widely used to measure health conditions of all kinds, there is a well-documented risk of misclassification52 that has not been specifically quantified for pregnancy-related diagnoses. Among patients with limited prenatal care, it is likely that pregnancy-related comorbidity is underestimated as many comorbid conditions (eg, gestational diabetes, preeclampsia) are typically diagnosed later in pregnancy. Because pregnancies were considered the unit of analysis and some individuals experienced 2 pregnancies in the study period, there may be a slight overrepresentation of the pregnancy characteristics among these individuals, though this impacted only a small number of individuals.
Use of national sample of CHCs serving a very large number of patients is a clear strength of this study, though use of ‘big data’ like this comes with 2 important limitations. First, statistically significant differences may or may not equate to clinically significant differences and studies with large sample sizes often demonstrate statistical significance at lower levels of absolute difference. Second, the study may not be generalizable to all CHCs or geographic regions. Though clinics came from 21 states across major geographic regions of the US, they were disproportionately clustered in western states and tended to be located in nonrural areas, which may have over-represented the proportion of CHCs providing longitudinal prenatal care. More research is needed to assess the impact of regional and geographical differences in practice patterns at CHCs.
Conclusions
CHCs provide longitudinal pregnancy care to a large population of patients who may be at higher risk for pregnancy-related complications. With increasing investment in multidisciplinary strategies to combat maternal morbidity and mortality in the US, CHCs are critically important partners for addressing disparities and ensuring access to high quality longitudinal prenatal and postpartum care. Ensuring continued support for CHCs and a pipeline of comprehensively trained staff should be high priorities in the fight to reduce maternal morbidity and mortality.
Appendix.
Inclusion/exclusion details:
–Clinics were in 21 states, specifically AK, CA, CO, FL, GA, HI, IN, KS, MA, MD, MN, MO, MT, NC, NM, OH, OR, RI, TX, WA and WI.
–Receipt of Title X funding was determined by cross-referencing clinic names and addresses with data from the Office of Population Affairs for 2018.
–School-based health centers and correctional facilities were excluded.
–Pregnancies without any indication of live birth were excluded because the need for prenatal care would have been uncertain in that group (eg, pregnancies that end in early miscarriage or abortion).
Patient Covariate Details
–Gestational complications were categorized using ICD10 codes (gestational diabetes, O24; preeclampsia, O11, O14, O15; hypertensive disorders of pregnancy O10, O13, O16) from the patients’ problem list during the time frame of each pregnancy. For chronic conditions, each was captured at the patient level during the study time period.
–Charlson Comorbidity Index was calculated from active problem list diagnoses before the study end date, and includes added weights for transplantation history, inflammatory bowel disease, seizures, sickle cell anemia, hemophilia, muscular dystrophy, Down syndrome, cystic fibrosis, Tay-Sachs disease, developmental delay, mental retardation, cerebral palsy, autism, schizophrenia, bipolar disorder, and drug or alcohol abuse.
–Diagnoses during pregnancy were ascertained exclusively during the period 30 days before the pregnancy start to 30 days after the pregnancy end date from the patients’ problem lists.
Sensitivity Analysis Interpretation
Pregnancies to women with 2 observed pregnancies had an average of 6.8 prenatal care visits per pregnancy, compared with 9.3 visits to women with 1 observed pregnancy (P < .001). Women with 2 observed pregnancies were more likely to be Black (23.9% compared with 21.0%) and not to be Latinx (53.8% compared with 45.2%, These mothers had higher proportions of all prepregnancy health comorbidities (diabetes, hypertension, mental health disorder, substance use disorder, tobacco use) and had higher Charlson Comorbidity Index scores. In each separate pregnancy, pregnancies to women with 2 pregnancies have greater proportions of gestational diabetes, gestational hypertension and preeclampsia (P < .001 for all).
Codes (Number (Percent)) Used to Identify Prenatal Care Encounters
Codes (Number (Percent)) Used to Identify Births
Codes (Number (Percent)) Used to Identify Postpartum Encounters
Sensitivity Analysis: Demographic and Health Characteristics by Number of Observed Pregnancies During Study Time Period
Notes
This article was externally peer reviewed.
Funding: This work was funded by the Agency for Healthcare Research and Quality award 1R01HS025155-01 and supported by the Accelerating Data Value Across a National Community Health Center Network (ADVANCE) Clinical Research Network (CRN). ADVANCE is a CRN in PCORnet®, the National Patient-Centered Clinical Research Network. ADVANCE is led by OCHIN in partnership with Health Choice Network, Fenway Health, and Oregon Health & Science University. ADVANCE’s participation in PCORnet® is funded through the Patient-Centered Outcomes Research Institute (PCORI), contract number RI-OCHIN-01-MC.
Conflict of interest: None.
To see this article online, please go to: http://jabfm.org/content/36/4/574.full.
- Received for publication January 20, 2023.
- Revision received March 22, 2023.
- Accepted for publication March 27, 2023.
