What Young African American Women Want in a Healthcare Provider

Abstract

The Institute of Medicine’s report on racial and ethnic disparities in health care encourages enhancing patient provider relationships by building trust. We explored factors important to 19–24 year old African American women (N=40) in choosing a health care provider. Eight focus groups were held in seven Ohio counties. Discussion was aided by photographs of client-provider interactions: two African American and two White providers (man or woman in each); in similar settings, attire, and pose; with a young African American woman client. Participants commented on what was happening in the photographs, how the woman felt, and their perceptions of each provider. Fongwa’s Quality of Care model guided analysis. Women providers were favored; race was not of primary concern. Provider proximity, perceived interest, and understandability were persistent preferences. Trust, awareness of body language, interest in client, and clearly conveying information clearly are critical for providers caring for young African American women.

African Americans have poorer health and birth outcomes in comparison to their White counterparts (US Department of Health and Human Services [USDHHS], 1999). The Centers for Disease Control’s 2006 National Vital Statistics Report (Martin et al., 2009) notes that infant death and low birth weight rates for African American woman are twice the rate of White women and African American women are more likely to have preterm labor and preterm births. African American women also have a maternal mortality rate that exceeds that of White women by three to four times (USDHHS, 1999). Socioeconomic status (SES) does play a significant role in understanding the difference in health outcomes, but SES cannot fully explain this gap (Institute of Medicine [IOM], 2003).

The 2003 IOM report on racial and ethnic disparities in health care was published with the intent to assess the extent of disparities, identify possible sources, and provide recommendations on possible interventions. The report identified patient, care process, and systemic level variables that can impact health outcomes (IOM). Patient level factors include experiences of discrimination, mistrust in the system, willingness to follow recommended treatments, and personal preference in care and in providers. Care process issues include provider biases and stereotyping (conscious and unconscious) and clinical uncertainty. Systemic variables include issues related to insurance/payment, accessibility and availability of care, and the fragmentation of the system. We sought to identify personal preferences important to 19–24 year old African American women when choosing a health provider. Findings presented are part of a larger study focusing on pre-conception and inter-conception issues among young African American women.

Review of Literature

The literature concerning racial concordance and preference of provider is inconsistent. Saha, Taggart, Kamaromy, and Bindman (2000) found that minorities often seek providers who are members of the same ethnic/racial background as themselves. Gerbert et al. (2003) displayed videos of different providers with clients and asked participants of different races which provider they preferred. Whereas only 23% of White participants chose an African American provider, 52% of African Americans chose a provider of their race. In each of the different ethnic groups surveyed, a majority of participants chose a provider of the same race (Whites 56%, Latinos 44%).

Patient-provider race concordance has been related to satisfaction with care. Individuals were more likely to choose a provider of the same race (LaVeist & Nuru-Jeter, 2002) and those who were of the same race as their physician reported higher satisfaction with their doctors (Benkert, Hollie, Nordstrom, Wickson, & Bins-Emerick, 2009; LaVeist & Nuru-Jeter). Others have found that African Americans who had same race providers were more likely to rate their doctor as excellent (Chen, Fryer, Phillips, Wilson, & Pathman, 2005; Cooper-Patrick et al., 1999; Saha & Komaromy, 1999), more likely to report receiving all needed medical care (Saha & Komaromy), and had more participation in a “partnership” with their physician (Cooper-Patrick et al.). African Americans with high perceptions of discrimination in the health care system were more likely to prefer a physician of the same race (Chen et al.).

However, other researchers found that race was not important. Bornstein, Marcus, and Cassidy (2000) identified factors important to choosing a primary care physician. There was little difference between African American and White respondents. Factors identified as most important were related to professional skills (hospital of admission, neatness of office and doctor, board certification, recommendation), and the least important factors were related to demographic characteristics (race, gender, religion, age, and marital status). No statistically significant relationship was found in selection of provider related to gender (that is, women wanting women providers) or race (African Americans wanting an African American provider). Bornstein et al. found that professional skill and office management were of greater concern than personal attributes of the provider.

Most research regarding women selecting a provider has primarily related to selection of an obstetric and gynecology (OB/GYN) provider and the results are mixed. Although some findings support women seeking providers who mirrored their sex and age (Schantz, Murphy, O’Sullivan, & Sorosky, 2007; Zuckerman, Navizedeh, Feldman, McCalla, & Minkoff, 2002), others found no gender preference in seeking an OB/GYN provider (Howell, Gardiner, & Concato, 2002; Johnson, Schantz, Kelsey, & Ohannessian, 2005).

Further complicating selection of and satisfaction with a provider are personal characteristics and communication style. Schantz et al. (2007) found that with no professional information about providers, women most frequently choose women providers. However, when provided with information on the providers’ “humanistic qualities and competence,” a significant number of women changed their provider of choice without regard to provider gender (Schantz et al.). Mast, Hall, and Roter (2007) assessed differences in client satisfaction based on gender and communication style of providers and found that women were more satisfied with a woman provider as long as she adopted a “caring” style of communication. However, none of these studies (Howell et al., Johnson et al., Mast et al., Schantz et al.) specifically addressed the preferences of African American women.

Researchers have identified general issues related to provider choice for African Americans (Bornestien et al., 2003; Chen et al., 2005; Gerbert, 2003; LaVeist & Nuru-Jeter, 2002; Saha et al., 2000) and for women seeking an OB/GYN (Howell et al., 2002; Johnson et al., 2005; Schantz, et al., 2007), yet research exploring characteristics young African American women seek in a provider is lacking. Understanding these characteristics may provide insight into why and when young African American women seek care. The aims of this qualitative focus group (FG) study were to: (a) identify characteristics important to young African American women in selecting health care providers; and (b) describe factors in a health care encounter important to young African American women.

Fongwa’s Quality of Care model, which guided this study, links quality to culture and is useful in identifying ways to improve health care services (Shoultz, Fongwa, Tanner, Noone, & Phillion, 2006). Fongwa (2000) maintained that clients’ perceptions of quality of care influence their relationship with their provider. The model, which is an extension of Holzemer’s Outcomes model (Holzemer, 1996), is depicted in a matrix format with inputs, processes, and outcomes on the horizontal axis and client, provider, and setting on the vertical axis (Fongwa, 2000). Inputs include the personal characteristics, skills, values, and beliefs that are brought into an interaction (Holzemer). Holzemer identified both client and provider-client processes as critical to satisfaction. Processes are concerned with the application of “good” medical care (Donnabedian, 1966) including level of satisfaction, behavior, knowledge, and preferences (Holzemer). Using the Shoultz et al. adaptation of the Fongwa model for the interpretation of the data, inputs are direct quotes from FG participants, processes are themes derived from the quotes, and outcomes are the practice implications regarding participants’ preferences in provider characteristics and in client provider encounters (Shoultz et al.). The Shoultz et al. adaptation was used in this study due to its previous application in qualitative health services research.

Methods

Institutional review boards of The Ohio State University and the Ohio Department of Health (ODH) approved the study. Eight FG with a total of 40 African American women were held in seven urban Ohio counties. The counties were in three different regions of the state as chosen by ODH, the funder of this study. Counties were chosen based on the highest African American populations in Ohio. We used purposeful sampling to recruit eligible women who were served by Women, Infants, and Children (WIC) clinics, Maternal Child health programs, community health centers, and local health departments. Purposeful sampling was conducted to enhance confidence that findings reflected the phenomenon of interest (Richards & Morse, 2007). Eligibility criteria were African American woman, between the ages of 19 and 24 years, and able to read and write English.

The eight FG were held at locations that provided public services to the communities, were well known in the area, and had easy access via public transportation. Half of the FG had six to seven participants; the remaining groups had 5 participants or less. All FG were moderated and co-moderated by African American members of the study team. Both the moderator and co-moderator were skilled in interviewing. Researchers purposefully chose to use a moderator and co-moderator who were women and African American to enhance participants’ comfort with the research process and to reflect the racial homogeneity (i.e., all African American women) of the sample.

Informed consent and demographic information were obtained from all participants prior to beginning each FG. Participants were provided with a $25 gift card and $10 for transportation. Support for child care was provided if needed.

During the FG, participants were shown four 8 × 11 inch photographs of client-provider interaction. Each photograph depicted a different provider: an African American woman, an African American man, a White woman, and a White man in a similar scenario-the same office-like setting, wearing a white lab coat, sitting at a table with a young African American woman who posed as the client. The provider and client appeared to be having a discussion. The four photographs were displayed together on a single tri-fold board. Participants were asked to describe the following aspects of each picture: (a) what was occurring, (b) how the young woman felt, and (c) their perceptions of each provider. Probes related to gender and race were posed if that information was not spontaneously forthcoming in the FG discussion and to elicit discussion regarding tentative explanations developed across successive FG regarding perceptions of providers and the health care encounters. At the conclusion of each FG, participants were asked to comment on an oral summary of the key discussion points presented by the co-moderator. This provided participants with an opportunity to consider whether the summary reflected their perspectives and to correct or add to the capture of the key points of the discussion.

All FG sessions were recorded. Recordings and co-moderator notes were transcribed by a professional transcriptionist and then verified by the researchers. All participant names were removed from the transcripts and notes. A minimum of three FG--subject to saturation of data--were judged to be adequate to identify any potential pattern of responses by region (Krueger & Casey, 2000). Saturation of data was attained by the midpoint of data collection (end of the fourth FG). No salient differences were noted within or among the regions during data collection and analysis as the study proceeded across all eight FG. The remaining four FG, which were required due to contractual obligations, confirmed saturation and supported researchers’ confidence regarding the strength of the analysis and conclusions (Richards & Morse, 2007).

Data were analyzed using content analysis techniques (Kruger & Casey, 2008). Transcripts were coded independently and then reviewed as a group. Discrepancies in coding of data were resolved via consensus. As data collection and analysis proceeded through the successive FG, data collection strategies reflected identified themes. Probes were used to elicit possible alternative explanations of identified themes. Using Fongwa’s Quality of Care (Schoultz et al., 2006) model as a guiding framework, constructed processes and outcomes were identified. All members of the research team reviewed processes and outcomes for concurrence. Participant exemplar quotes are included in the results below and additional quotes are in the tables to illustrate the linkages among the inputs, processes, and outcomes revealed in the data. Tables 1 and 2 display selected key quotes (inputs) in the first column, processes as key themes (second column), and outputs as practice implications (third column).

Table 1.

Provider characteristics important to young African American women.

Inputs	Processes	Outcomes
"A male doctor is rough too."	Men were not always perceived as trustworthy; more comfort was found in another woman as a provider.	Gender and race sensitivity are important in provider-client relationships.
"Probably I'm more comfortable with a woman."	While there was some concordance and preference for same race, priority was gender concordance.	Critical to nurture trust in the relationship, especially critical for men providers.
“She’s a women, she’s the same color as me, looks like she knows what she’s talking about.”
“Men seem to be more cold, like they just tell you what’s the matter and leave. I feel a woman will talk it out.”
“Because a man he just doesn't care, but women they know, they seem like they know more about your body.”
"There is a certain level of comfort maybe having the person be like the client, a Black woman."

Table 2.

Health care encounter factors important to young African American women.

Inputs	Processes	Outcomes
“She's sitting in closer to her.”	Providers perceived as comfortable in the interaction, interested in the client, and competent were valued.	Provider awareness of messages conveyed via body language, actions, and interactions is needed.
"She's really explaining to her making her more involved with whatever she’s going through.”	Positive body language and attitude were important.	Providers identified as more interested, competent, and involved may gain greater receptivity from clients.
“He’s not making eye contact. He’s just reading to her."	Ability to understand the provider and the information conveyed was essential.	Provider must assure client has clear understanding of information conveyed.
“She's talking to her more on an adult level.”
“He seems comfortable in his environment."
"When I was pregnant with my daughter, I couldn't understand anything, he'd just be talking and I couldn't understand."
“But not only language, English is important.”
“Give me layman’s terms.”
“Can you break it down because I really don’t understand you.”

Results

A majority (68.5%) of the 40 FG participants reported being single, unemployed (62.5%), and enrolled in Medicaid (75%). About 18% did not graduate from high school, and 50% attended or graduated from college or a technical school. Most had been pregnant at some point (87%). Seventy-two percent had given birth to a live baby and 77.5% had at least one child. Almost half of the participants (42.5%) received health care primarily from their own private doctor, nurse practitioner, or physician’s assistant, approximately one-third (30%) from a clinic, and about 18% from an urgent care center or emergency department. Results are presented below per research aim.

Characteristics Important in Selecting a Health Care Provider

Table 1 displays inputs (selected quotes), processes, and outcomes specific to provider characteristics. Provider race was not considered of primary importance to most FG participants, but across all FG gender was critical. A woman provider was preferred by 90% of participants (n=36). One participant noted:

because she’s female, maybe she’s a little more comfortable about opening up about the information and she seems more involved and can relate to you on like a more personal level about how you feel and how much you get out of the conversation.

Participants in other FGs made comments such as “For me, I feel more comfortable with a female doctor than a male doctor,” “I’m uncomfortable with a male,” and “But you can’t really trust a male doctor.”

A young woman from another FG stated “I think race plays a role, but I don’t know, like I still feel gender plays a stronger role than race with the men and maybe race [doesn’t] play a role, I don’t know.” Participants identified wanting a provider that could be trusted and with whom they were comfortable. Having a White woman provider was preferred by 60% of the participants as illustrated in the following quote:

I think the White woman is feeling more comfortable with sharing information, which means she’s probably making her more involved with whatever she’s going through. Because I noticed also with the different providers I have with this pregnancy, it seems like the White woman has me more involved and it’s like when I have a Black doctor, it’s like this is the way it is, we’re going to do this and I’m trying to explain that I know my body, this is going on. But they don’t want to listen.

Although women were overwhelmingly preferred as providers, four participants favored men providers as exemplified by the following:

Personally I don’t like going to a woman. Like if you do try to explain something, from my experience it seems like the woman, she wants to be a part and try to steer you into another direction. Whereas a male, he might be more lenient because he actually doesn’t have to go through it, he’s trying to help the situation. So he listens to certain things more than she would.

The outcomes derived from the FG data regarding provider characteristics suggest the importance of gender and race sensitivity and nurturing trust as a precursor to strong client-provider relationships, especially for men providers.

Factors Important in a Health Care Encounter

After viewing the four photographs, participants commented on their perceptions of the health care encounter depicted in the photographs. Table 2 presents selected quotes as inputs, as well as the processes, and outcomes related to health care encounter factors. Provider choice was affected by the women’s perception of the provider’s competency and perceived interest in and their comfort with the client. Descriptors such as being “comfortable” with the client were frequently voiced as valued in a health care encounter. One participant commented about a woman provider in the photographs “She is comfortable like she is teaching her client what she needs to know.” A similar comment from a participant in another FG also embodied the concept of comfort “She seems comfortable like she wants to help the client.”

Body language such as eye contact, gestures, and proximity to the client could convey interest or lack of interest in the client “[provider’s] just pouring stuff out, not really going over it, just pointing at things.” Examples from another FG include “body posture looks uncomfortable” and “if you’re not looking at somebody you’re really not paying attention.” Perception of attitude and responsiveness of the provider are valued encounter factors “…he’s telling her, you know, this is and this, and not explaining” and “Because she’s telling you what you need to know, page by page, make sure you don’t miss any details.” When asked what about the provider matters to you, a participant responded, “Their attitude, the way they carry themselves. If they speak clearly and get you to understand what they are saying, I’m OK.”

Providers who appeared to have a positive attitude and respond to the client depicted in the photograph were favored more often in the FG discussions. Participants noted that being respected, listened to, and involved in their own care was important. “I think it’s imperative to tell me everything, don’t come in there winging it, I don’t care what race you are.”

Although not initially a topic of the FG discussion, concerns were raised about a provider’s ability to communicate in English. One participant commented:

They gave me a doctor. I couldn’t tell you what she said her name was. I didn’t know what she was saying. I was asking them, “Can I get another doctor because I don’t know what she’s saying?”

Further probing on this topic revealed it was critical for the provider to not only be able to speak English but also use understandable terms. Participants stated that it was imperative that their health care information be conveyed in non-medical terms so that they could understand their conditions as well as their treatment plans. One participant commented “Because I need to learn information, so I don’t want somebody just showing me something I need to understand so I can take care of myself.” Another participant shared this perception “If they speak clearly and get you to understand what they are saying, I’m OK.” A different young woman commented “The fact that they speak a different language is not a problem as long as I can understand them.”

Outcomes related to health care encounters included provider awareness of how messages are being conveyed. Using appropriate body language, being comfortable, displaying genuine interest and competence were important. Other provider actions that were described as critical were using clear language and assuring the client understands the information discussed.

Discussion

We explored the characteristics of providers and factors in a health care encounter that are important to young African American women. Schoultz et al. (2006) adaptation of Fongwa’s Quality of Care model (2000) was used to guide categorization of themes of input, processes, and outcomes. Women in this study identified gender as an important provider characteristic, whereas race was less important when choosing a provider. Congruent with previous research (Wheatley, Kelley, Peacock, & Delgado, 2008) participants identified specific factors in a provider-client encounter that they believed indicated whether a provider was competent: communication style, demonstrating understanding and listening skills, and showing interest in the women. The themes identified in our FGs closely mirrored those found in previous research regarding what marked “quality” in prenatal care (McKenzie & Oliphant, 2010; Wheatley et al., 2008) and the defining of “support” by African American women in pregnancy (Coffman & Ray, 1999).

In this study gender concordance was seen as a critical component of choosing a provider, while others have found that racial concordance--and not gender--was paramount (Bornstein et al., 2000; Christen, Alder, & Bitzer, 2008; Howell, et al., 2002; Saha et al., 2000). Although our study did not specifically indicate that the client in the four photographs was seeing a women’s health specialist, the participants referred to their own experiences, primarily in OB/GYN care. Bornstein et al. acknowledged that decisions on provider preference might differ when seeking an OB/GYN.

Participants identified aspects of interactions with providers that influenced their perceptions. Consistent with previous research (Christen, et al., 2008; Copeland, Scholle, & Binko, 2003; Howell, et al., 2002; Mavis, Vasilenko, Schnuth, Marshall, & Colavito, 2005), verbal and nonverbal communication were critical to a successful client-provider interaction. English speaking ability of non-native providers was of primary concern. Participants described having providers speaking in verbiage that they could understand to be of equal importance to speaking English. Participants identified that they would not elect to see specific providers, because those providers appeared to be “telling” not explaining. Separation between provider and client in the pictures was seen as being “uncomfortable” and not representing a “caring” interaction.

A strength of this study is that saturation of data was attained by the end of the fourth FG. Investigator triangulation also enhanced the rigor of the study (Lincoln & Guba, 1985). All members of the research team reviewed the transcripts, concurred that there was active discussion and expression of opinions within each FG, and concurred on the study findings.

A limitation of this study was that participants were recruited intentionally from urban areas of Ohio’s major cities. In addition, we focused on women who were served by the safety net and public programs such as WIC, local health departments, and community health clinics and therefore only represents young African American women who are served through these programs. African American women of different income levels may have different perceptions and beliefs.

Implications

Understanding what African American women are seeking in a provider may provide insight into understanding poor health outcomes in African American women (USDHHS, 1999; IOM, 2003). This study provides information about African American women’s perceptions of provider characteristics and the impact on those characteristics on outcomes of health care encounters. Further investigation is needed to answer research questions such as: Do African American women avoid seeking care if they perceive the available providers to be lacking in gender sensitivity? To what extent do both objective and subjective outcomes of care for African American women differ by provider gender and race? These questions are particularly important when women are limited in the number of providers that they can access, such as when sthey must rely on safety net providers.

Conclusion

Women that participated in this study described physical characteristics and mannerisms that were important to them when choosing a provider. Participants discussed characteristics of the patient-provider interaction as more important than the physical attributes of the provider. These findings support the need for future studies on understanding the relationship between clients and their health care providers and the impact of these relationships on health outcomes.

Biographies

Helen E. Dale, MSN, CNM, FNP-BC RN is a current doctoral candidate at The Ohio State University, Columbus, Ohio, USA.

Barbara J. Polivka, PhD, RN is an associate professor at The Ohio State University College of Nursing, Columbus, Ohio, USA.

Rosemary V. Chaudry, PhD, MHA, RN is an assistant clinical professor at The Ohio State University College of Nursing, Columbus, Ohio, USA.

Gwenneth Simmonds, PhD, CNM, RN is an assistant professor at South Carolina State University Program of Nursing, Orangeburg, South Carolina, USA.