Person-Centered Maternity Care and Health Outcomes at 1 and 6 Months Postpartum

Abstract

Objectives:

Person-centered care has been increasingly recognized as an important aspect of health care quality, including in maternity care. Little is known about correlates and outcomes of person-centered care in maternity care in the United States.

Materials and Methods:

Data were from a prospective cohort of more than 3000 individuals who gave birth to a first baby in a Pennsylvania hospital. Person-centered maternity care was measured via a 13-item rating scale administered 1-month postpartum. Content validity was established through exploratory factor analysis. The resulting scale had scores ranging from 13 to 54, with Cronbach's alpha of 0.86. Using linear and logistic regression models to control for covariates, we examined associations between participants' characteristics and person-centered maternity care and between person-centered maternity care and postpartum outcomes.

Results:

Participants had a mean total score of 47.80 on the person-centered maternity care scale. Patient factors independently associated with more person-centered maternity care included older age, more positive attitude toward vaginal birth during pregnancy, and spontaneous vaginal birth. In adjusted models, higher person-centered maternity scale scores were strongly associated with many positive physical and mental health outcomes at 1 and 6 months postpartum.

Conclusions:

Our findings underscore the importance of person-centered maternity not just due to its intrinsic value but also because it may be associated with both mental and physical health outcomes through the postpartum period. Results suggest that policy efforts are necessary to ensure person-centered maternity care, especially for delivery hospitalization experience.

Introduction

Person-centered care has been increasingly recognized as an important aspect of health care quality, including in maternity care.^1–4 Person-centered care includes high-quality communication, respectful treatment, involving patients in decisions about their care, and respecting bodily autonomy.^1,3,5 Research has found that person-centered care is associated with increased patient satisfaction, trust in the provider, treatment adherence, and health care utilization.^6–18 Aspects of person-centered care can also lead to better health outcomes, although the literature assessing this has had mixed findings.^11,19–23

Nearly 4 million people give birth in the United States each year,²⁴ and the rates of maternal morbidity and mortality have increased over time, with large racial disparities.^25,26 Recently there has been increasing attention to disrespect, abuse, and obstetric violence globally, as well as in the United States.^27–30 Research on person-centered maternity care among individuals giving birth in facilities in low- and middle-income countries has found variation in the receipt of overall and specific aspects of person-centered care by country, but indicates the need to increase the provision of person-centered maternity care.^31,32 An emerging literature has begun to examine person-centered maternity care in the United States. Findings from a national survey indicate that between 40% and 66% of birthing people report communication problems in the prenatal period with their maternity care providers.³³ In addition, national and state-specific surveys have shown that between 8% and 24% of birthing people experience some type of discrimination in a health care setting during the perinatal period.^33,34 People from minoritized racial and ethnic groups, people with lower levels of education, and people with public insurance are less likely to report shared decision–making during birth, and more likely to experience mistreatment.^35,36 Qualitative studies conducted among birthing people from minoritized racial and ethnic groups have found that patient–clinician interactions are often characterized by experiences of disrespect and discrimination, and clinicians fail to share adequate information to enable patient autonomy in decision-making about their care.^37–40 Individuals with more obstetric interventions during labor also report experiences that are less consistent with person-centered care.^35,36

Recent statements from the American College of Obstetricians and Gynecologists (ACOG) and the American College of Nurse-Midwives (ACNM) affirm these organizations' commitment to providing person-centered care for individuals during the perinatal period.^41,42 However, because data on aspects of person-centered care in maternity care are not routinely collected, a limited number of studies have examined correlates and outcomes of person-centered care in maternity care in the United States . Modifiable factors related to postpartum well-being, such as the provision of person-centered care, are important to better understand, given that mental and physical health problems postpartum are common and frequently inadequately addressed.⁴³

In this analysis, we use data from a prospective cohort of people giving birth for the first time to (1) examine the correlates of person-centered maternity care, including sociodemographic characteristics, interventions during birth, and medical complexity using a novel scale, and (2) assess the relationship between person-centered maternity care and mental and physical health.

Materials and Methods

We used data from the First Baby Study, a prospective longitudinal cohort of 3006 individuals in Pennsylvania who gave birth to their first baby in a hospital between 2009 and 2011. Recruitment methods and the study population details have been described in previous publications.^44,45 Baseline interviews were conducted during participants' third trimester of pregnancy and additional interviews at 1, 6, 12, 18, 24, 30, and 36 months postpartum. Birth certificate and hospital discharge data were linked to interview responses. The present study uses data from the baseline and 1- and 6-month surveys, as well as birth certificate and hospital discharge data. Participants were included in the analytic sample for this study if they had no missing values on covariates (N = 2,892). The First Baby Study was approved by the Penn State College of Medicine Institutional Review Board (IRB) and the IRBs of participating hospitals; this analysis, which used deidentified data, was determined to be exempt from review by the University of Massachusetts Amherst IRB.

Measures

Person-centered maternity care scale

The measure of person-centered maternity care consisted of 13 items, shown in Table 1, that were asked at the 1-month postpartum interview. An initial draft of the instrument was developed through extensive review of literature and documents related to “patient-centered care” (the terminology in widespread use at the time), refined through focus group discussions, and pilot tested and modified based on pilot results. Items included statements such as “My needs and preferences were respected,” and “I was provided with the information that I needed to care for my baby and myself.” Response choices ranged from “strongly agree” to “strongly disagree” on a 4-point Likert scale. Scores for negative items were reversed, and scores for all 13 items were added together, resulting in a scale with a theoretical range of 13–52, with higher scores representing a more person-centered maternity care experience.

Table 1.

Person-Centered Maternity Care Scale Items

Item	Agree (strongly or somewhat)	Disagree (strongly or somewhat)
1. My family was welcomed and treated with respect.	2847 (98.7)	38 (1.32)
2. I had trouble getting my questions answered.a	237 (8.2)	2655 (91.8)
3. I was treated with dignity and respect.	2846 (98.4)	46 (1.6)
4. I received as much information as I wanted.	2779 (96.1)	113 (3.9)
5. The hospital staff were busy and rushed.a	717 (24.8)	2175 (75.2)
6. My needs and preferences were respected.	2824 (97.6)	68 (2.4)
7. I had trouble getting help when I needed it.a	326 (11.3)	2566 (88.7)
8. My questions and concerns were answered promptly.	2776 (96.0)	116 (4.0)
9. I was provided with the information needed to care for my baby and myself.	2830 (97.9)	62 (2.1)
10. The medical care that my baby and I received was excellent.	2855 (98.7)	37 (1.3)
11. I felt like I was being rushed to have my baby quickly.a	202 (7.0)	2687 (93.0)
12. I was in charge of how things went during my labor, delivery, and hospital stay.	2523 (87.2)	369 (12.8)
13. I felt like the treatment decisions that were made during my labor and delivery were made mostly for the convenience of the hospital doctors and nurses.a	332 (11.5)	2557 (88.5)

The potential responses to these questions were: strongly agree (4), somewhat agree (3), somewhat disagree (2), and strongly disagree (1). Negative items were reversed to create a total score such that the higher the score the more positive they were toward the care they received. Total scores could range from 13 to 52.

^a Indicates that item was reversed.

For participants missing a response for one item (n = 13), we used person-mean substitution to impute scale scores.⁴⁶ No participants were missing a response to more than one item. Content validity was established through exploratory factor analysis. Cronbach's alpha for the resulting scale was 0.86, which indicates good internal-consistency reliability. Actual scores ranged from 13 to 54. The mean total score was 47.80, and the 50th percentile was 49.0.

As a check for construct validity, we examined the association between person-centered maternity care scale scores and satisfaction with hospital care using a correlation test and linear regression. Hospital care satisfaction was assessed by asking participants at the 1-month postpartum interview, “Overall, how satisfied are you with the care that you received while you were in the hospital having your baby?” Responses were rated on a 4-point scale from “extremely dissatisfied” to “extremely satisfied.” These measures were strongly correlated, with a correlation coefficient of 0.64. Each additional point on the care satisfaction measure was associated with a 5.9-point increase on the patient centered maternity care scale.

Other variables

We created variables measuring the following maternal sociodemographic characteristics: age (18–24, 25–30, or 31–36 years), race/ethnicity (White, Black, Latina, or other/multiple races), education level (high school degree or less, some college, or bachelor's degree or higher), insurance type (private or nonprivate), and partnership status (married and living together, not married but living with partner, or other). At the baseline interview, participants were asked if they had experienced a number of medical conditions before or during pregnancy. Based on those responses, we created indicator variables for each of the following: prepregnancy or gestational diabetes, prepregnancy or gestational hypertension, prepregnancy obesity, and anxiety or depression before the index pregnancy. We also measured a number of prenatal factors. Psychosocial stress and social support during the index pregnancy were measured using modified versions of validated scales.^47,48 Another scale assessed participants' prenatal attitudes toward vaginal delivery,⁴⁹ with higher scores indicating a more positive attitude toward vaginal delivery.

In addition, we assessed birth-related characteristics: birth mode (spontaneous vaginal, assisted vaginal, cesarean), and labor support from a doula, and whether there were self-reported maternal and infant complications during the birth. Shared decision-making during birth was assessed by the Delivery Decision Making Scale, with higher scores indicating higher levels of shared decision-making.³⁵ This scale was dichotomized into a score of six compared with a score less than six. We also created a set of indicator variables for the delivery hospital.

Postpartum outcomes

Outcomes related to physical health and mental health were examined at 1 and 6 months postpartum. At the 1-month and 6-month postpartum interviews, participants were asked various questions relating to their physical health. First, participants were asked to rate their overall health as “excellent,” “very good,” “good,” “fair,” or “poor.” We used responses to create a binary variable for overall health (excellent/very good/good vs. fair/poor). Then, participants were asked questions about how often in the past week they had (1) been in pain; (2) been sick and unable to get out of bed; (3) been vigorous and healthy; (4) been tired from lack of sleep, and (5) been able to do all of their normal activities. Response choices for these items were “all of the time,” “most of the time,” “some of the time,” “a little of the time,” or “none of the time.”

We created the following binary variables: (1) in pain some of the time or more (vs. a little of the time or none of the time); (2) sick and unable to get out of bed some of the time or more (vs. a little of the time or none of the time); (3) not vigorous and healthy most of the time or all of the time (vs. a little of the time or none of the time); (4) tired from lack of sleep some of the time or more (vs. a little of the time or none of the time); and (5) not able to do normal activities most of the time or all of the time (vs. a little of the time or none of the time).

At 1 and 6 months postpartum, depressive symptoms were evaluated using the Edinburgh Postnatal Depression Scale.⁵⁰ We coded respondents as having depressive symptoms if they scored 13 or greater, as this is suggestive of postpartum depression.⁵⁰ Respondents were also asked a series of questions to determine whether they were experiencing symptoms of birth-related post-traumatic stress at one month postpartum.⁵¹

Analysis

We first examined bivariate associations between participants' characteristics, person-centered maternity care, and postpartum outcomes using chi-square tests and t-tests. We then estimated a multivariable linear regression model to examine the association between participants' characteristics and person-centered maternity care scale scores. Finally, we estimated multivariable logistic regression models to estimate the relationship between person-centered maternity care scores and each postpartum outcome, controlling for covariates. Multivariable models used clustered standard errors to account for correlation of respondents within hospitals. All analyses were conducted in Stata 16.

Results

Associations between participant characteristics and person-centered maternity care are shown in Table 2. The youngest age group (18–24 years) had the lowest mean score on the person-centered maternity care scale (p < 0.001). Participants with higher levels of education had higher scores on average (mean 47.0 for individuals with a high school degree or less vs. mean 48.0 for individuals with a Bachelor's degree or higher, p < 0.001), as did individuals with private health insurance (48.1 for those with private health insurance vs. 46.8 for those without private insurance, p < 0.001). Social support during pregnancy and more positive prenatal attitudes toward vaginal birth were positively associated with person-centered maternity care scale scores, while psychosocial stress was negatively associated.

Table 2.

Characteristics of Participants by Person-Centered Maternity Care Scale Score (N = 2,892)

	N	%	PCMC mean (SD)	p
Sociodemographic characteristics
Age				<0.001
18–24	740	25.6	47.1 (5.2)
25–30	1432	49.5	48.2 (4.4)
31–36	720	24.9	47.8 (4.6)
Race/ethnicity				0.014
White	2437	84.3	47.9 (4.6)
Black	198	6.9	47.1 (5.3)
Latina	146	5.1	47.6 (4.3)
Other, multiple races	111	3.8	46.9 (6.1)
Education level				<0.001
High school degree or less	453	15.7	47.0 (5.3)
Some college	767	26.5	47.9 (5.2)
Bachelor's degree or higher	1672	57.8	48.0 (4.3)
Insurance type				<0.001
Private	2228	77.0	48.1 (4.3)
Nonprivate	664	23.0	46.8 (5.9)
Partnership status				<0.001
Married and living together	2083	72.0	48.1 (4.3)
Not married but living with partner	494	17.1	47.0 (5.9)
Other	315	10.9	47.4 (5.3)
Medical characteristics
Prepregnancy or gestational diabetes				0.946
No	2703	93.5	47.8 (4.7)
Yes	189	6.5	47.8 (5.3)
Prepregnancy or gestational hypertension				0.192
No	2686	92.9	47.8 (4.7)
Yes	206	7.1	47.4 (5.3)
Prepregnancy obesity				0.466
No	2273	78.6	47.8 (4.6)
Yes	619	21.4	47.7 (5.2)
Anxiety or depression before index pregnancy				<0.001
No	2223	76.9	48.0 (4.5)
Yes	669	23.1	47.2 (5.3)
Prenatal factors
Psychosocial stress during pregnancy			−0.167a	<0.001
Social support during pregnancy			0.255a	<0.001
Prenatal attitude toward birth mode (higher indicates more positive toward vaginal birth)			0.846a	<0.001
Birth-related factors
Birth mode				<0.001
Spontaneous vaginal	1805	62.4	48.2 (4.3)
Assisted vaginal	256	8.9	47.4 (5.0)
Cesarean	831	28.7	47.1 (5.3)
Had doula support in labor				0.097
No	2712	93.8	47.8 (4.7)
Yes	180	6.2	47.2 (5.5)
High level of shared decision-making				<0.001
No	1043	36.1	45.9 (5.8)
Yes	1849	63.9	48.9 (3.5)
Maternal complication during birth				<0.001
No	2478	85.7	48.0 (4.5)
Yes	414	14.3	46.5 (5.7)
Infant complication during birth				<0.001
No	2474	85.6	48.0 (4.4)
Yes	418	14.5	46.8 (6.0)
Hospital				0.038
Hospital 1	504	17.4	47.6 (4.6)
Hospital 2	487	16.8	48.1 (4.4)
Hospital 3	304	10.5	48.3 (4.4)
Hospital 4	354	12.2	48.1 (4.4)
Hospital 5	404	14.0	47.3 (5.0)
Hospital 6	316	10.9	47.9 (4.4)
Other	523	18.1	47.9 (5.3)

^a Beta coefficient reported for continuous variables.

PCMC, person-centered maternity care; SD, standard deviation.

Several birth-related factors were also associated with person-centered maternity care. Mean person-centered maternity care scale scores were about one point higher among participants who gave birth vaginally compared to those with cesarean births (p < 0.001). There was some variation across hospitals in scale scores, with a one-point difference between the lowest mean hospital score and the highest.

Table 3 presents bivariate associations between postpartum outcomes and person-centered maternity care. Higher person-centered maternity care scale scores were associated with better mental health and physical health outcomes across most of the measures examined, at both 1 and 6 months postpartum.

Table 3.

Outcomes by Person-Centered Maternity Care Scale Score (N = 2892)

			Mean person-centered maternity care scale score
Outcome	N	%	If no for outcome	If yes for outcome	p
Mental health at 1 month postpartum
Post-traumatic stress symptoms	227	7.6	48.0 (4.4)	44.7 (7.1)	<0.001
Depressive symptoms	111	3.7	47.9 (4.6)	44.9 (6.5)	<0.001
Physical health at 1 month postpartum
Overall health is fair or poor	125	4.2	47.9 (4.6)	45.0 (6.5)	<0.001
Pain more than a little or none of the time	437	14.5	47.9 (4.5)	47.0 (5.5)	<0.001
Sick and unable to get out of bed some of the time or more	63	2.1	48.9 (4.7)	45.2 (6.6)	<0.001
Not vigorous and healthy all or most of the time	373	12.4	48.0 (4.6)	46.3 (5.2)	<0.001
Tired from lack of sleep some of the time or more	2449	81.5	48.4 (4.3)	47.7 (4.8)	<0.001
Not able to do normal activities all or most of the time	1397	46.5	48.2 (4.6)	47.3 (4.8)	<0.001
Mental health at 6 months postpartum
Post-traumatic stress symptoms	81	2.7	47.9 (4.6)	44.1 (7.9)	<0.001
Depressive symptoms	108	3.7	47.9 (4.6)	45.4 (6.9)	<0.001
Physical health at 6 months postpartum
Overall health is fair or poor	146	5.0	47.9 (4.6)	45.9 (6.8)	<0.001
Pain more than a little or none of the time	227	7.8	48.0 (4.5)	45.7 (6.7)	<0.001
Sick and unable to get out of bed some of the time or more	46	1.6	47.8 (4.7)	46.9 (5.8)	0.207
Not vigorous and healthy all or most of the time	262	9.0	47.3 (4.6)	46.4 (5.6)	<0.001
Tired from lack of sleep some of the time or more	1787	61.4	48.2 (4.7)	47.6 (4.7)	0.001
Not able to do normal activities all or most of the time	249	8.6	47.9 (4.6)	46.5 (5.5)	<0.001

In the adjusted multivariable model presented in Table 4, there were fewer associations between participants' characteristics and person-centered maternity care scale scores than in the bivariate analyses. Individuals aged 25–30 had higher person-centered maternity care scale scores compared to individuals aged 18–24 (β = 0.62, 95% confidence interval [CI]: 0.07–1.18). Education level and insurance type were not associated with person-centered maternity care scale scores. More psychosocial stress during pregnancy was associated with lower person-centered maternity care scale scores (β = −0.09, 95% CI: −0.13 to −0.05), while more social support during pregnancy was associated with higher person-centered maternity care scale scores (β = 0.17, 95% CI: 0.11–0.22).

Table 4.

Adjusted Associations Between Participant Characteristics and Person-Centered Maternity Care Scale Scores

	β	95% CI
Sociodemographic characteristics
Age category (Ref = 18–24)
25–30	0.62*	(0.07 to 1.18)
31–36	0.38	(−0.04 to 0.80)
Race/ethnicity (Ref = White)
Black	0.56	(−0.64 to 1.75)
Latina	0.44	(−0.37 to 1.25)
Other	−0.40	(−1.72 to 0.93)
Education level (Ref = High school or less)
Some college	0.28	(−0.33 to 0.89)
Bachelor's degree or higher	−0.05	(−0.50 to 0.41)
Insurance type (Ref = Private)
Nonprivate	−0.61	(−1.22 to 0.01)
Partnership status (Ref = Married)
Not married but living with partner	0.02	(−0.38 to 0.41)
Other	0.62	(−0.03 to 1.28)
Medical characteristics
Prepregnancy or gestational diabetes	0.37	(−0.18 to 0.91)
Prepregnancy or gestational hypertension	−0.19	(−0.70 to 0.31)
Prepregnancy obesity	0.09	(−0.27 to 0.44)
Anxiety or depression before index pregnancy	−0.21	(−0.73 to 0.32)
Prenatal factors
Psychosocial stress during pregnancy	−0.09***	(−0.13 to −0.05)
Social support during pregnancy	0.17***	(0.11 to 0.22)
Prenatal attitude toward birth mode (higher indicates more positive toward vaginal birth)	0.49*	(0.07 to 0.90)
Birth-related factors
Birth mode (Ref = Spontaneous vaginal)
Assisted vaginal	−0.43	(−0.94 to 0.07)
Cesarean	−0.70***	(−0.99 to −0.41)
Labor support from a doula (Ref = No)	−0.34	(−1.04 to 0.35)
High level of shared decision-making	2.55***	(2.30 to 2.80)
Maternal complication	−1.11***	(−1.53 to −0.70)
Infant complication	−0.73*	(−1.31 to −0.14)
Hospital (Ref = Other)
Hospital 1	−0.19	(−0.59 to 0.22)
Hospital 2	−0.04	(−0.41 to 0.33)
Hospital 3	0.40	(−0.01 to 0.81)
Hospital 4	0.40*	(−0.01 to 0.78)
Hospital 5	−0.46*	(−0.87 to −0.04)
Hospital 6	−0.01	(−0.39 to 0.37)

^* p < 0.05; ^**p < 0.01; ^***p < 0.001.

CI, confidence interval.

Having a more positive attitude toward vaginal birth prenatally was associated with higher person-centered maternity care scale scores (β = 0.49, 95% CI: 0.07–0.90). Participants with cesarean births had lower person-centered maternity care scale scores compared with participants with spontaneous vaginal births (β = −0.70, 95% CI: −0.99 to −0.41). Individuals who reported a high level of shared decision-making during their birth had higher scores on the person-centered maternity care scale (β = 2.55, 95% CI: 2.30–2.80). Individuals who experienced maternal and infant complications had lower person-centered maternity care scores.

In adjusted models, high person-centered maternity care scale scores continued to be strongly associated with nearly all measures of physical and mental health at both 1 and 6 months postpartum (Table 5). For example, every one-point increase in the person-centered maternity scale was associated with a 7% decrease in the odds of rating overall health as fair or poor at 1 month postpartum and a 5% decrease in odds of fair or poor overall health at 6 months postpartum. Individuals who reported higher levels of person-centered maternity care had lower odds of depressive symptoms at 1 and 6 months postpartum (adjusted odds ratio [AOR] = 0.93, 95% CI: 0.91–0.95 for 1 month postpartum, AOR = 0.95, 95% CI: 0.92–0.98 for 6 months postpartum). Higher scores on the person-centered maternity care scale were also associated with lower odds of post-traumatic stress symptoms at both 1 and 6 months postpartum (AOR = 0.92, 95% CI: 0.90–0.94 and AOR = 0.94, 95% CI: 0.91–0.98, respectively).

Table 5.

Adjusted Odds of Outcomes by Person-Centered Maternity Care Scale Score

Outcome	AOR	95% CI
Mental health at 1 month postpartum
Post-traumatic stress symptoms	0.92***	(0.90–0.94)
Depressive symptoms	0.93***	(0.91–0.95)
Physical health at 1 month postpartum
Overall health is fair or poor	0.93***	(0.90–0.97)
Not in pain a little or none of the time	0.98*	(0.96–0.99)
Sick and unable to get out of bed a little or none of the time	0.92***	(0.89–0.95)
Not vigorous and healthy all or most of the time	0.96***	(0.94–0.98)
Tired from lack of sleep some of the time or more	0.96***	(0.95–0.97)
Not able to do normal activities all or most of the time	0.96***	(0.95–0.97)
Mental health at 6 months postpartum
Post-traumatic stress symptoms	0.94***	(0.91–0.96)
Depressive symptoms	0.95**	(0.92–0.98)
Physical health at 6 months postpartum
Overall health is fair or poor	0.95**	(0.92–0.98)
Not in pain a little or none of the time	0.94***	(0.92–0.95)
Sick and unable to get out of bed a little or none of the time	0.98	(0.94–1.03)
Not vigorous and healthy all or most of the time	0.96***	(0.94–0.98)
Tired from lack of sleep some of the time or more	0.98*	(0.96, 1.00)
Not able to do normal activities all or most of the time	0.96*	(0.93, 0.99)

Models adjusted for: age, race/ethnicity, education level, insurance type, partnership status, prepregnancy or gestational diabetes, prepregnancy or gestational hypertension, obese before pregnancy, birth mode, doula support in labor, prenatal attitude toward birth mode, maternal complication in birth, infant complications at birth, birth hospital; standard errors are clustered by hospital.

^* p < 0.05; ^**p < 0.01; ^***p < 0.001.

AOR, adjusted odds ratio.

Discussion

In adjusted models, we found that person-centered maternity care was higher among individuals aged 25–30 (vs. 18–24) who experienced higher levels of social support during pregnancy and a high level of shared decision-making during birth. Person-centered maternity care was lower among birthing people who experienced more psychosocial stress during pregnancy, who delivered by cesarean, and who experienced a maternal complication during birth. Higher levels of person-centered maternity care during the childbirth hospitalization were associated with indicators of better mental and physical health at 1 and 6 months postpartum.

In bivariable analyses, participants who identified as Black and other or multiple races reported lower levels of person-centered maternity care than did White participants, as did participants with lower (vs. higher) education levels and nonprivate insurance. However, these associations were attenuated in the adjusted models. While some prior literature has reported that people of color and people with lower socioeconomic status (SES) are less likely to receive person-centered care,^35,52–54 other studies have not found these associations.^33,55–57

There are several potential reasons for these mixed findings in general, as well as the null findings in our study. Patient reports of health care experiences can be impacted by differential response tendencies across groups,^58,59 or by differing expectations of care across groups.⁶⁰ This can lead to similar reports of health care experiences on quantitative measures even if actual experiences are different; qualitative research has consistently documented that perinatal care experiences are substantially impacted by racism, as well as by discrimination based on other characteristics such as SES and age.^37–40 The measure we used of person-centered maternity care was not specifically designed to capture the experiences of birthing people of color, and the sample was nearly 85% White; both of these factors may have limited the ability of our study to detect differences in experiences of person-centered maternity care by race and ethnicity in multivariable models.

In addition, as sociodemographic factors such as age, race and ethnicity, education, and insurance type can be correlated, disparities in person-centered maternity care by all of these characteristics in bivariable analyses are still worthy of notice and need to be addressed. Individuals in the youngest age group reported lower levels of person-centered maternity care, which is consistent with findings from low- and middle-income countries.⁶¹

Since the First Baby Study, there has been increased attention to the importance of people's experiences during childbirth and interest in improving data collection on this topic, including more accurately capturing the experiences of birthing people of color.^36,62,63 Projects have been undertaken to generate patient-reported outcomes specific to childbirth and for people of color specifically,^64–66 although data collection based on these item banks has not yet been implemented on a broad scale.

Participants with cesarean births reported lower levels of person-centered care compared to participants with spontaneous vaginal births. The previous literature on this topic has been mixed; one study found lower levels of shared decision-making (one specific possible dimension of person-centered care) reported by individuals with unplanned cesarean births,³⁵ while another study found no relationship between cesarean birth and a number of interpersonal processes of care measures.⁶⁷ While we found that perceptions of shared decision-making were associated with our measure of person-centered maternity care, this relationship was not deterministic, and these were clearly two distinct constructs. We also found that individuals with a more favorable attitude toward vaginal birth prenatally reported more person-centered care, independent of birth mode. While our results do not speak directly to the reasons for this, it is possible that people with more favorable attitudes toward vaginal birth were more informed about birth or more likely to have chosen a hospital where they were likely to have an experience that was a match for their values.⁶⁸

Individuals who experienced high levels of person-centered maternity care in the hospital had better mental and physical health outcomes at both 1 and 6 months postpartum. A 2019 commentary noted that there was a need for literature documenting a connection between respectful maternity care and postpartum post-traumatic stress disorder (PTSD), since the very type of experiences that may result from disrespectful maternity care (e.g., loss of dignity, experience of intense helplessness) are those that could be factors in triggering postpartum PTSD.⁶⁹ Our results demonstrate just such a connection; that is, more person-centered maternity care experiences were protective against developing post-traumatic stress symptoms at both 1 and 6 months postpartum. Person-centered maternity care was also associated with a decrease in odds of depressive symptoms postpartum.

We also found a relationship between person-centered maternity care and better physical health postpartum, even after accounting for birth mode and maternal medical characteristics. People who experience person-centered maternity care may feel safer and more confident in their care, leading to different patterns of management and care seeking for postpartum physical health issues.⁷⁰ It is also possible that person-centered maternity care experiences are more likely in settings that prioritize this as part of an overall effort to deliver high-quality care⁶⁹; in this case, higher levels of person-centered maternity care may have been indicative of overall better care quality during the birth hospitalization, in ways that may have positively impacted postpartum health.

Since person-centered maternity care was assessed at the same time that 1-month health outcomes were reported, these associations could be partially attributable to retrospective reporting bias; however, our findings are bolstered by the fact that these associations persisted for health outcomes at 6 months postpartum. Our findings also underscore the importance of the data collection initiatives mentioned above not just for providing more reliable information on treatment in childbirth due to the intrinsic value of person-centered care but also because person-centered care may be associated with health outcomes through the postpartum period.

Our findings suggest that policy efforts are needed to ensure access to person-centered maternity care. For example, implementation of the patient-reported measures discussed above may be helpful in providing timely feedback that can be used in quality improvement efforts at the practice or health system level,⁷¹ as well as training for clinicians and other patient-facing staff addressing communication skills⁷² and concepts such as cultural humility and antiracism.^73,74

A variety of enhanced maternity care models have been tested that may improve person-centeredness, including group prenatal care, maternity care homes, and birth center-based care through the Strong Start for Mothers and Newborns Initiative.⁷⁵ While the focus of these models on prenatal care is certainly important for maternal and newborn health, our results point to the importance of the experience of care during the birth hospitalization as well and indicate that this might be a reasonable target of future policies. Innovative perinatal care models that seek to address racial inequities in care have done so by prioritizing the provision of relationship-centered care, including by supporting the agency of the pregnant person and displaying cultural humility, and these models have encompassed the perinatal period more broadly.^76–78

Limitations

Our findings should be considered in light of some limitations. The timing of the assessment of person-centered maternity care could have impacted participants' perceptions of their care; for example, prior research has found that reports of disrespect and abuse were lower when participants were surveyed at facility discharge versus 5–10 weeks postpartum.⁷⁹ However, as all First Baby Study participants were interviewed at ∼1 month postpartum, respondents would not have been differentially impacted by the amount of time since the birth. As mentioned above, person-centered maternity care was assessed at 1 month postpartum, as were several outcomes; it is possible that individuals who were having a more difficult recovery recalled their childbirth hospitalization less favorably. In addition, participants with more complications during birth may have both experienced care as less person centered and had worse physical health in the postpartum period; however, our adjusted analyses included multiple medical factors indicating complexity.

The data used in this analysis are based on survey responses by people who gave birth in 2009–2011, and there have been changes to maternity care in the intervening years, including an increased emphasis on postpartum care.⁴³ However, the First Baby Study is unique as a large prospective cohort that captured a variety of measures of patient experiences, enabling the current analysis. Finally, the First Baby Study is not a population-based sample, and participants were more likely to be White and privately insured compared with the overall population of people giving birth in Pennsylvania.⁴⁵

Conclusions

We examined correlates of and outcomes associated with person-centered maternity care, using a novel scale that measured experiences during the childbirth hospitalization. While sociodemographic characteristics were largely not independently associated with experiencing person-centered maternity care, individuals who gave birth by cesarean reported less person-centered maternity care. We found that individuals who experienced person-centered maternity care had better health at 1 and 6 months postpartum across a variety of measures, emphasizing the importance of person-centered maternity care for mental and physical health.

Disclaimer

The funder had no role in study design, analysis, interpretation of data, or in the preparation, review, or approval of the article, or the decision to submit the article for publication. The article's contents are solely the responsibility of the authors and do not represent the official views of the NICHD.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

The First Baby Study was funded by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) R01 HD052990. Research reported in this publication was supported by NICHD R03 HD098392.