Race/Ethnicity and Perception of Care: Does Patient–Provider Concordance Matter?

Keywords

racial concordance - obstetrics - patient and provider - discordant dyads

Significant racial and ethnic disparities in maternal morbidity and mortality persist in the United States. Black pregnant people are three to four times more likely to die from pregnancy-related causes than White women.[1] Hispanics/Latinx pregnant people (herein referred to as Hispanics) have been shown to have higher rates of severe maternal morbidity and are more likely to die from hypertensive diseases in pregnancy.[2] [3]

The various dimensions of structural racism are a foundational cause of health disparities, impacting health at various levels.[4] One specific area where structural racism has an appreciable effect is at the level of the patient–provider encounter.[5] [6] Physician-held implicit biases[7] and negative racial stereotypes influence attitudes toward and expectations of patients and, most, notably providers' diagnostic and treatment decisions.[8] A reduction in bias with cultural and linguistic congruency in racially and ethnically concordant patient–physician dyads may increase trust, the basis of quality care.

Currently, research into the effect of racial concordance on experiences and outcomes of birthing mothers is lacking. Given trends in the role of racial concordance in various fields and the accentuation of the racial concordance benefit in complex medical cases,[9] [10] we seek to explore the effect of racial and ethnic concordance on the experiences of birthing people undergoing obstetrical procedures in the second stage of labor (e.g., operative vaginal delivery or intrapartum cesarean delivery). The rationale for studying individuals undergoing operative interventions in the second stage is that these individuals required unplanned, time-sensitive counseling with direct impacts on maternal and, potentially, neonatal health. Thus, optimizing care in this setting requires high-quality communication skills from a clinician and trust from the patient. We hypothesize that individuals cared for by providers belonging to their perceived same racial group (i.e., racial concordance between individual and their obstetric provider) will report higher levels of satisfaction in their care compared with racially discordant counterparts.

Materials and Methods

This is a prospective observational cohort of newly postpartum people who underwent operative vaginal delivery or cesarean delivery in the second stage of labor at Northwestern Memorial Hospital in Chicago, IL. This hospital is a quaternary care center in which approximately 12,000 individuals give birth each year, including a socioeconomically and racially diverse patient population. Over 150 obstetric providers participate in labor and delivery care, including obstetrics and gynecology residents, maternal-fetal medicine fellows, obstetrics and gynecology specialty faculty, maternal-fetal medicine subspecialty faculty, and certified nurse midwives. When patients require operative interventions in the second stage of labor, any of these individuals may be the primary person responsible for counseling and consenting for intervention, although attending physicians are involved in the care of all patients.

All postpartum individuals aged ≥18 years old who spoke English or Spanish and who underwent an operative vaginal delivery or cesarean section in the second stage were eligible for participation in the study. Each participant was identified by trained research assistants during the postpartum hospitalization. Patients who did not speak English were approached by a certified bilingual research assistant, and all study protocols were conducted in the participant's language of choice. Biomedical and birth outcome data were abstracted from the electronic medical record. Participants completed a survey collecting sociodemographic information, as well as self-perceived race or ethnicity. As part of the survey, participant were asked to identify their primary obstetrician during the intrapartum period and to assign the individual a particular race or ethnicity. Given that our study was conducted in an academic center, resident and fellow physicians may play an equal or even greater role in intrapartum counseling. Patients may not be aware of the academic hierarchy, and thus, we felt that it was important not to direct participants to select their attending obstetrician. Rather, they were instructed to complete the survey based on who they felt provided the bulk of their intrapartum counseling, regardless of training level or practice type.

Participants were also administered the modified Interpersonal Processes of Care (IPC) survey, a 29-item instrument that assesses several subdomains of communication, patient-centered decision-making, and interpersonal style.[11] It has been validated in both English and Spanish, various racial groups and primarily primary care (i.e., nonobstetric) settings to assess factors that contribute to health outcomes and examine the impact of quality improvement endeavors. The IPC survey is scored “so that higher scores indicate higher frequency of the labeled interpersonal process, which means that higher scores sometimes indicate better processes (e.g., decided together) and sometimes indicate worse processes (e.g., lack of clarity”[11]; [Supplementary Appendix A] [available in the online version]). Subdomains include those of communication (hurried communication, elicited concerns, explained results), decision-making (patient-centered decision-making), and interpersonal style (compassionate, respectful, discriminated, and disrespectful nursing staff).[11]

The exposure was participant-assigned racial concordance or discordance with their obstetrical provider (i.e., participant–physician dyad). The primary outcome was the median IPC subscore focused on participant assessment of discrimination due to race or ethnicity. Secondary outcomes were all median IPC domain subscores. Assuming an α of 0.05 and 80% power, we estimated a need for 74 persons (37 in each group) to detect a difference of 0.3 points on the IPC survey between racially concordant versus discordant groups. Bivariable analyses were performed using chi-square or Fisher's exact test, for categorical variables, and t-test or Wilcoxon rank-sum test, for continuous variables. Statistical significance was set at p<0.05. All data were analyzed in R (version 3.6.2). IRB approval was obtained from Northwestern University (STU00212861).

Results

In total, 168 patients were approached, 107 (63.6%) agreed to participate, and 87 (81.3%) completed the survey (n=49 racially discordant, n=38 concordant). The final sample included 87 individuals, of which 49 were in racially discordant dyads and 38 were in racially concordant dyads.

When compared with racially discordant dyads, participants in racially concordant dyads were more likely to identify as White (76.3 vs. 11.4%, p ≤ 0.001), be older (p=0.017), have completed a higher degree of education (p=0.03), and have a higher household income (p=0.002; [Table 1]). Intrapartum characteristics, including mode of delivery, were not significantly different between groups ([Table 2]).

Table 1 Sociodemographic characteristics

Variable	Concordant (n=38)	Discordant (n=49)	p-Value
Age (y)
18–24	1 (2.6)	12 (24.5)	0.017
25–29	8 (21.1)	10 (20.4)
30–35	22 (57.9)	17 (34.7)
36–40	7 (18.4)	8 (16.3)
40 or older	0 (0.0)	2 (4.1)
Race/ethnicity
White	29 (76.3)	5 (11.4)	<0.001
Black	5 (13.2)	10 (22.7)
LatinX	3 (7.9)	20 (45.5)
Asian	1 (2.6)	7 (15.9)
Pacific Islander	0 (0.0)	2 (4.5)
Participant primary language
English	35 (97.2)	40 (83.3)	0.071
Spanish	1 (2.8)	8 (16.7)
Education
High school or less	4 (10.5)	10 (20.4)	0.03
Some college	4 (10.5)	14 (28.6)
College or advanced degree	30 (78.9)	25 (51.0)
Income ($)
<25,000	2 (5.3)	11 (22.4)	0.002
25,001–40,000	7 (18.4)	9 (18.4)
40,001–75,000	2 (5.3)	12 (24.5)
≥75,000 or more	27 (71.1)	17 (34.7)
Insurance type
Private payor	32 (84.2)	27 (55.1)	0.005
Government payor	6 (15.8)	22 (44.9)
Provider race/ethnicity
White	29 (78.4)	39 (83.0)	0.273
Black	5 (13.5)	4 (8.5)
LatinX	2 (5.4)	0 (0.0)
Asian	1 (2.7)	4 (8.5)
Language provider used
English	38 (100.0)	47 (95.9)	0.502
Spanish	0 (0.0)	2 (4.1)
Interpreter used by provider (if applicable)	0 (0.0)	2 (4.1)	0.502
Participant language preference
English	37 (97.4)	46 (93.9)	0.629
Spanish	1 (2.6)	3 (6.1)

Note: Data displayed as n (%).

Table 2 Clinical characteristics

Variable	Concordant (n=38)	Discordant (n=49)	p-Value
BMI (kg/m2)
18.5–25	2 (5.3)	3 (6.1)	0.255
25–30	18 (47.4)	14 (28.6)
30–35	11 (28.9)	13 (26.5)
35–40	5 (13.2)	11 (22.4)
40+	2 (5.3)	8 (16.3)
Nulliparous	33 (86.8)	32 (65.3)	0.026
Mode of delivery
Cesarean delivery	25 (65.8)	32 (65.3)	1.0
Operative	13 (34.2)	17 (34.7)
Primary cesarean delivery	24 (96.0)	27 (84.4)	0.215
Mode of operative delivery
Forceps	12 (92.3)	14 (82.4)	0.613
Vacuum	1 (7.7)	3 (17.6)
Failed operative delivery	0 (0.0)	4 (13.3)	0.114
Use of neuraxial anesthesia	37 (97.4)	48 (98.0)	1.000
Third or fourth degree laceration	4 (30.8)	5 (29.4)	1.0
Surgical complications[a]	5 (13.2)	13 (26.5)	0.183

Abbreviation: BMI, body mass index.

Note: Data presented as n (%).

^a Exploratory laparotomy, postpartum hemorrhage (defined as estimated blood loss>1L)[16], emergent hysterectomy, and/or visceral injury at time of laparotomy.

No significant difference was noted between racially concordant and discordant dyads in the IPC survey domains of communication, shared decision-making, interpersonal style (compassion and respect), or discrimination ([Table 3]). Importantly, participants reported a generally a favorable perception toward lack of discrimination, as well as communication, patient-centered decision-making, and interpersonal style. In the secondary analysis comparing participants by race/ethnicity, there was no statistical significance in IPC survey results between groups ([Table 4]).

Table 3 IPC survey results, stratified by racial concordance

Variable	Concordant (n=38)	Discordant (n=49)	p -Value
	Median score [IQR]	Median score [IQR]
Discriminated
Discriminated due to race/ethnicity[a]	4.00 [4.00, 4.00]	4.00 [4.00, 5.00]	0.551
Disrespectful nursing staff[a]	4.00 [4.00, 4.00]	4.00 [4.00, 4.00]	0.829
Communication
Hurried communication—lack of clarity[a]	6.00 [5.00, 8.00]	7.00 [5.00, 9.00]	0.527
Elicited concerns, responded[b]	15.00 [13.00, 15.00]	15.00 [13.00, 15.00]	0.955
Explained results, medications[b]	19.00 [17.00, 20.00]	18.00 [15.00, 20.00]	0.152
Decision-making
Patient-centered decision-making—decided together[b]	17.00 [14.00, 20.00]	17.00 [14.00, 20.00]	0.649
Interpersonal style
Compassionate, respectful[b]	25.00 [23.00, 25.00]	25.00 [22.00, 25.00]	0.631

Abbreviations: IPC, Interpersonal Processes of Care; IQR, interquartile range.

Notes: For discriminated subdomain: discriminated due to race/ethnicity.

For communication subdomain: hurried communication scores range from 5 to 25, elicited concerns, responded scores range from 3 to 14, and explained results, medications scores range from 4 to 20.

For decision-making subdomain, scores range from 4 to 20.

For Interpersonal Style subdomain, scores range from 5 to 25.

^a Higher score indicates a less favorable perception of the interpersonal care process.

^b Higher score indicates a more favorable perception of the interpersonal care process.

Table 4 IPC survey results, stratified by postpartum participant race/ethnicity

Variable	Black	LatinX	White	p -Value
	Median score [IQR]	Median score [IQR]	Median score [IQR]
Discriminated
Discriminated due to race/ethnicity[a]	4.00 [4.00, 5.00]	4.00 [4.00, 4.00]	4.00 [4.00, 5.00]	0.216
Disrespectful nursing staff[a]	4.00 [4.00, 4.00]	4.00 [4.00, 4.00]	4.00 [4.00, 4.00]	0.576
Communication
Hurried communication—lack of clarity[a]	7.00 [5.00, 8.00]	7.50 [5.75, 9.25]	6.00 [5.00, 8.00]	0.249
Elicited concerns, responded[b]	15.00 [11.00, 15.00]	15.00 [13.75, 15.00]	14.00 [13.00, 15.00]	0.774
Explained results, medications[b]	19.00 [17.00, 20.00]	18.00 [15.50, 20.00]	18.50 [16.00, 20.00]	0.806
Decision-making
Patient-centered decision-making—decided together[b]	17.00 [14.00, 19.00]	17.50 [12.50, 20.00]	17.00 [14.00, 20.00]	0.774
Interpersonal style
Compassionate, respectful[b]	25.00 [23.00, 25.00]	25.00 [22.50, 25.00]	25.00 [23.00, 25.00]	0.696

Abbreviations: IPC, Interpersonal Processes of Care; IQR, interquartile range.

Notes: For discriminated subdomain, scores range from 4 to 20.

For communication subdomain: hurried communication scores range from 5 to 25, elicited concerns, responded scores range from 3 to 14, explained results, medications scores range from 4 to 20.

For decision-making subdomain, scores range from 4 to 20.

For Interpersonal Style subdomain, scores range from 5 to 25.

^a Higher score indicates a less favorable perception of the interpersonal care process.

^b Higher score indicates a more favorable perception of the interpersonal care process.

Discussion

We found that older, White, English-speaking persons with higher degrees of education and income were more likely to be racially concordant with their providers. In contrast to our hypothesis, we identified no meaningful differences in patient-reported IPC by racial concordance or self-reported race/ethnicity.

Previous data, mostly derived from the nonobstetric literature, emphasize the role of racial concordance in different health outcomes. White, Black, Hispanic, and Asian patients who were racially/ethnically concordant with their provider reported greater satisfaction with their physician compared with discordant groups.[12] Racial and ethnic concordance was associated with an increased likelihood of seeking preventive care,[13] using a higher volume of health services,[12] and a lower likelihood of postponing or delaying care in Black and Hispanic patients.[14] Within the realm of obstetrics, Fryer et al demonstrate that under the care of Black physicians, the mortality rate for Black newborns is significantly lower as compared with those cared for by White physicians.[10]

Strengths and Limitations

A major strength of the study is the inclusion of patients who underwent operative vaginal deliveries or cesarean deliveries conducted in the second stage. This specific scenario was chosen because these are bounded procedures that require more in-depth counseling in a stressful period, making the role of communication quality much more critical than in less stressful health care interactions. Importantly, due to the presence of one or two clinicians during these scenarios and the close proximity of study participation to the clinical events, provider communication could be assessed more specifically, thereby reducing recall bias or confusion regarding counseling by multiple providers.

This study was not without limitations. Despite planned oversampling of Black and LatinX patients, our team was still unable to obtain necessary sample sizes to fully explore these concordant groups due to disproportionately fewer minority providers. Therefore, our study is likely underpowered to detect a difference in the primary outcome. An additional limitation is that the IPC survey has not been validated in obstetrics and, thus, may not capture key obstetric processes of care. Newly developed metrics for obstetrical quality of care aimed at identifying and remediating obstetric racism, such as the Patient Reported Experience Measure of OBstetric racism (PREM-OB) Scale, emphasize the role of communication[15] but have yet to be tested within the clinical scenario in which our study was conducted—thus limiting our study's generalizability and external validity. Finally, as with any survey-based study, residual recall and selection biases limit the generalizability of our findings.

Conclusion

The racial disparities in obstetrics warrant further examination of patient–provider factors that may contribute to poor outcomes, as well as those that may limit or prevent further discrepancies in morbidity and mortality among different racial/ethnic groups. Studies performed in nonobstetric fields have demonstrated that racial concordance between patients and providers has a positive impact on patient adherence, trust, and use of health care services. While our study did not detect differences in perceptions of IPC between racially concordant and discordant groups, further research examining racial concordance in obstetrics is essential. As future studies examine the role of racial concordance in obstetrics, the disparities in the health care provider workforce remain a challenge.

• References

• 1 Callaghan WM. Overview of maternal mortality in the United States. Semin Perinatol 2012; 36 (01) 2-6
• 2 Creanga AA, Bateman BT, Kuklina EV, Callaghan WM. Racial and ethnic disparities in severe maternal morbidity: a multistate analysis, 2008-2010. Am J Obstet Gynecol 2014; 210 (05) 435.e1-435.e8
• 3 Howell EA, Egorova NN, Janevic T, Balbierz A, Zeitlin J, Hebert PL. Severe maternal morbidity among Hispanic women in New York City: investigation of health disparities. Obstet Gynecol 2017; 129 (02) 285-294
• 4 Gee GC, Ford CL. Structural racism and health inequities: old issues, new directions. Du Bois Rev 2011; 8 (01) 115-132
• 5 Dominguez TP. Race, racism, and racial disparities in adverse birth outcomes. Clin Obstet Gynecol 2008; 51 (02) 360-370
• 6 Howell EA. Reducing disparities in severe maternal morbidity and mortality. Clin Obstet Gynecol 2018; 61 (02) 387-399
• 7 Green AR, Carney DR, Pallin DJ. et al. Implicit bias among physicians and its prediction of thrombolysis decisions for black and white patients. J Gen Intern Med 2007; 22 (09) 1231-1238
• 8 Smedley BD, Stith AY, Nelson AR. et al. Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care. Washington, DC: Institute of Medicine, National Academies Press; 2002
• 9 Greenwood BN, Hardeman RR, Huang L, Sojourner A. Physician-patient racial concordance and disparities in birthing mortality for newborns. Proc Natl Acad Sci U S A 2020; 117 (35) 21194-21200
• 10 Fryer Jr GE, Green LA, Vojir CP. et al. Hispanic versus white, non-Hispanic physician medical practices in Colorado. J Health Care Poor Underserved 2001; 12 (03) 342-351
• 11 Stewart AL, Nápoles-Springer AM, Gregorich SE, Santoyo-Olsson J. Interpersonal processes of care survey: patient-reported measures for diverse groups. Health Serv Res 2007; 42 (3, pt. 1): 1235-1256
• 12 Laveist TA, Nuru-Jeter A. Is doctor-patient race concordance associated with greater satisfaction with care?. J Health Soc Behav 2002; 43 (03) 296-306
• 13 Ma A, Sanchez A, Ma M. The impact of patient-provider race/ethnicity concordance on provider visits: updated evidence from the medical expenditure panel survey. J Racial Ethn Health Disparities 2019; 6 (05) 1011-1020
• 14 LaVeist TA, Nuru-Jeter A, Jones KE. The association of doctor-patient race concordance with health services utilization. J Public Health Policy 2003; 24 (3,4): 312-323
• 15 White VanGompel E, Lai JS, Davis DA. et al. Psychometric validation of a patient-reported experience measure of obstetric racism (The PREM-OB Scale suite). Birth 2022; 49 (03) 514-525
• 16 Committee on Practice Bulletins-Obstetrics. Practice Bulletin No. 183: postpartum hemorrhage. Obstet Gynecol 2017; 130 (04) e168-e186