Background

The number of African-born immigrants to the US grew 40-fold between 1960 and 2007, and currently numbers approximate 1.5 million [1]. African-born immigrants are disproportionately from war-affected nations, and refugees and asylum seekers are at particular risk for health and health care disparities within this population. Refugees can be defined as those outside of their country of citizenship or place of last habitual residence, who are seeking protection based on persecution due to race, religion, nationality, political opinion, or social group. Asylum seekers are those who seek refugee protection from inside the United States, or from a port of entry [2].

One-third of all refugees and asylum seekers arriving in the United States are African-born [3]. African immigration, to the state of Massachusetts specifically has increased 62% from 2000 to 2010, with a total of 77,375 African-born immigrants in Massachusetts in 2010 [4]. African refugees represent 37% of all refugees arriving to the United States [5] and 46% of all refugees arriving to the state of Massachusetts in most recent available estimates [6].

A majority of these immigrants and refugees are women of childbearing age who underutilize gynecologic care [7]. We have previously reported that less than half of women in the Congolese and Somali populations we serve have received a routine gynecologic examination in the past year, only 23% of women over the age of 45 have received a mammogram, and one-third or more of women report a lack of knowledge of mammography or cervical cancer screening [8]. Immigrant women in general are less likely to be insured or have a regular source of health care [9,10,11], are less likely to have received a pap smear or health evaluation in the last year [12], and may be at increased risk for adverse pregnancy outcomes [9, 13, 14].

There is high prevalence of “high risk sexual behavior” and poor access to care among post-conflict populations [15, 16]. Effective health services utilization to address this risk may be further constrained by language barriers, discordance between the cultural beliefs of providers and patients, difficulties acculturating, and lack of appropriate insurance coverage and financial resources [17, 18]. One qualitative study of Somali immigrant and refugee women living in the United States elucidated patterns of discordant health beliefs and expectation between reproductive health providers and patients [19].

The current study is a secondary analysis of prior work conducted in Congolese and Somali immigrant communities in Boston. We analyzed focus groups for themes relevant to (a) gynecologic health-seeking behaviors, and (b) barriers to accessing reproductive health services, in order to better understand decreased service utilization and address reproductive health disparities in this at-risk population.

Theoretical Framework

We used the Integrated Model of Behavior as the conceptual model for qualitative analysis [20]. This framework examines a person’s intentions, knowledge, and constraints in order to understand why a behavior is or is not performed. Relevant components from this framework that emerged in our analysis are listed in parenthesis in the “Results” section.

Methods

Data for this study were obtained from the UJAMBO Project, which has been described elsewhere [5]. The UJAMBO project included focus groups with male and female community members to assess health care acquisition knowledge, as well as attitudes and social norms related to general, reproductive and mental health concerns. This study focuses on findings relating to gynecologic health-seeking behaviors and barriers to accessing reproductive health services. Findings relating to other broad themes such as experiences of sexual violence, mental health, and trauma are presented separately [21].

Certain principles of community-based participatory research informed the original UJAMBO Project. Community members identified through community-academic partnerships with participating organizations informed topics addressed in educational DVDs and focus group discussion questions. Staff from two partnering non-profit organizations, the Somali Development Center and the Congolese Women’s Association of New England, contributed to program development and focus group design (for example, the decision to separate focus groups by age, national background, and gender), were trained as interviewers who co-led and interpreted focus groups, facilitated recruitment of study participants, provided space for focus groups, and attended project leadership meetings. Study findings were shared with partner organizations to facilitate dissemination of study findings to key stakeholders and inform next steps.

Participants

From June through July 2008, trained research staff from partnering Somali and Congolese community organizations recruited a convenience sample of eligible (i.e., age 18 years or older and not cognitively impaired) community members at community events. All recruited individuals were asked to attend a focus group on a designated day. Immediately prior to focus group participation, verbal informed consent was obtained.

In order to ensure comfort within peer discussions, focus groups were divided by gender and nationality, then further divided into the following age cohorts: 18–25, 26–35, and 36 years and older. A series of six women’s focus groups were conducted and were included in this analysis given our focus on gynecologic care.

Data Collection

The study team used an interview guide that was developed in English. All focus groups were conducted in English by a study co-investigator with an interpreter simultaneously facilitating bilingual discussions. This methodology of “cross-cultural focus groups” using a real-time interpreter who acts as a culturally familiar “health broker” to interpret women’s experiences and narratives, with note-taking instead of audio-recording, has been described and validated elsewhere as a way to honor principles of community-based participatory research, by allowing members of the affected community to inform the best way to create an exchange among themselves, researchers, and participants [22]. Co-investigators spoke English and were doctoral-level researchers trained in qualitative data collection. Bilingual discussions were held in English and French, Lingala, or Somali as preferred by participants. Each focus group met for one session and ranged from 85 to 195 min. Participants were given referrals to Boston Medical Center for health care and a $20 gift card for their time.

Focus groups were not audio-recorded out of concern for the participants’ comfort and potential hindrance of discussion; discomfort with recording devices in this population has been described elsewhere [7]. For data analysis, detailed notes were taken in English by a designated note-taker and the co-investigator group leader.

Analysis

We analyzed notes using a grounded theory approach [23]. Two coders identified mutually exclusive codes or themes across focus groups, then independently coded data. Additional codes and sub-codes were identified iteratively and reapplied to previous groups. Inter-coder reliability across coding was reached via standard approach and arbitration by the lead investigator [24].

Measures

A manual coding process was used to sort coded data into health domains (Fig. 1). During the coding process, tree diagrams were constructed to depict study domains, codes specific to these domains, and sub-codes within each code.

Fig. 1
figure 1

Diagram of coding domains, themes and sub-themes used

Full size image

This study was conducted as a partnership between academics at Boston University School of Medicine and School of Public Health and two community-based organizations in Greater Boston: the Congolese Women’s Association of New England and the Somali Development Center. All study procedures were reviewed and approved by the Boston University Institutional Review Board.

Results

Sample Characteristics

Thirty one women participated in the focus groups. Fifteen women were from the Democratic Republic of the Congo and 16 from Somalia. The average length of US residence for the study participants was 6.57 years (SD 4.23; range < 1 year to 15). The average age for both groups was comparable: 34.5 years (range 18–50) for Congolese women 34.1 years (range 18–59) for Somali women.

Disparities in education were noted; 60% of the Congolese women had attended university (partial or full), compared to only 12.5% of the Somali women. Congolese women were generally asylum seekers or U.S. citizens, while Somali women mostly identified with being refugees or U.S. citizens. Eighty percent of Congolese women read English very well or somewhat well, in contrast to 20% of the Somali women. 86.6% of Congolese and 61.9% of Somali women reported employment.

Housing status varied between the groups. For the Congolese, 40% had unstable housing (staying with someone or move from place to place), 26.6% owned their own home and 20% rented. In the Somali group, 81.2% rented homes and the remaining reported unstable housing (living in a shelter or move from place to place). Marital status was comparable across both groups.

Key themes emerging from our qualitative analysis are presented below.

Domain 1: Gynecologic Health Care Seeking Preferences

Theme 1: Gynecologic Care Utilization Most Common During Pregnancy (Knowledge/Salience)

Somali and Congolese women described seeking reproductive and gynecologic care in the context of pregnancy, rather than as part of routine care. Participants expressed that they did not feel that seeking care was warranted without pregnancy as a pretext. “Usually we don’t see doctors unless we are giving birth,” a Somali participant in the 25–36 age group stated. A Congolese woman in the same age group similarly told us, “In general, the only time a Pap or GYN exam is done is when women are pregnant.” This finding differed in focus groups by age. Younger participants were consistently more likely to be aware of recommendations to seek care between pregnancies, but many still articulated not acting on this recommendation: “Most understand you should go [to the doctor] when not pregnant [i.e., preventive care], but usually only go when pregnant,” one Somali participant in the 18–25 age group shared.

Theme 2: Care not Warranted in the Absence of Pain (Knowledge/Salience)

Participants in both ethnic groups and all age groups repeatedly expressed that a certain level of discomfort should be endured prior to presenting to the health care system for gynecologic treatment. “If I don’t have any pain, I don’t feel like going to the doctor,” one Somali participant said. These statements were met with agreement from the group. “People feel if healthy or no pain, it is not necessary to go to the doctor. You only go to the doctor if there is something wrong,” a Congolese woman echoed. Reflective of similar views in young as well as older participants, a younger Somali participant reflected on lack of trust in gynecologic providers as healers rather than provokers of pain: “Some will go to the doctor [for pelvic pain]. Others believe pain will increase if you go to the doctor.”

Overcoming reluctance to present to care without a concrete cause or source of discomfort was tied to acculturation in addition to age. “There are still some people who have a hard time adapting. They have their old ways with the mentality of not seeking health care. These people ‘settle’ with the mentality of not getting help.” Younger participant recognized an implicit tension between an “old” way, where care would not traditionally be sought, and a new context where it would be considered acceptable to seek preventive care.

Theme 3: Female Providers Strongly Preferred for Gynecologic Care (Experiential Attitude)

When questioned regarding gender preference for OB/GYN providers, both Congolese and Somali participants in all age groups predominantly preferred female providers. “I feel shame to go to a man. [I want] a woman just like me. She has what I have; she knows where I am coming from,” one Somali participant told us. “[Congolese women] will want a women doctor. Most immigrant females feel more comfortable with a female doctor,” stated a Congolese participant. Endorsements of shame and stigma and the role of experiences of sexual trauma and experiences of violence were more common in Congolese focus groups.

Theme 4: Experience with Female Genital Cutting (Experiential Attitude)

For Somali women and not Congolese women, discussion focused on the importance of having a gynecologic health care provider experienced with and accepting of female genital cutting, also described in some literature as female genital modification or mutilation (FGM). Comments on FGM were not prompted by discussion guides, arising spontaneously leading to further exploration in focus groups; “circumcision” was the term most often used by participants. All comments reflected women’s desires to receive educated, non-judgmental care. “A doctor might say, ‘What the hell happened to you?’ Many doctors do not know about [FGM],” one Somali participant said. “I feel bad [when a doctor is shocked by FGM],” another shared. Somali women in all age groups shared the same sentiments.

Domain 2: Barriers to Gynecologic Health Care Acquisition

Theme 1: Stigma Associated with Seeking Gynecologic Care (Normative Beliefs)

The stigma of seeking gynecologic care emerged as a major barrier to utilization of such services for both communities and in all age groups. “Even for me, a married woman with children, if I tell my mom I am going to the gynecologist while my husband is away; she is going to ask me what I did wrong,” said one Somali woman.

“Once she tells someone about it, they will talk about her. They gossip. People might criticize or make up stories,” a Congolese participant reflected. Many participants described facing stigma within families and in the community around gynecologic care seeking behaviors.

Theme 2: Not Being Married a Barrier to Seeking Gynecologic Care (Normative Beliefs)

Unmarried women in particular described pressure to not seek gynecologic care and cervical cancer screening specifically for fear it would indicate pre-marital sexual activity, viewed as a major taboo in both Somali and Congolese communities—this was significantly more obvious in the responses in younger age groups, reflecting a higher proportion of unmarried women in these groups. “If she is married then she will go; otherwise no reason. If she is not married she will not go,” a younger Somali participant stated. “Our family expects us to be virgins when we marry. So they would not go to gynecologist before marriage,” a Congolese woman said. The distinction between unmarried ‘girls’ and married ‘women’ was implicit in several narratives: “In the Somali culture, if you tell your mom ‘I go to the hospital now,’ she’ll tell you, no you’re not. You’re a girl and you can’t go anywhere in that culture.” Another elaborated, “For young [unmarried] women, it would mean she is having an affair. Women will avoid [going to gynecologist] to prevent people from spreading the rumor that you are having an affair.”

Theme 3: Seeking Gynecologic Care a Threat to Modesty (Normative Beliefs)

Privacy and the pressure to avoid exposure of female anatomy due to culturally valued sexual modesty was a significant barrier to presenting for and utilizing gynecologic care faced by both Somali and Congolese participants; this was more commonly expressed by older participants but was noted to some degree in all age groups. “The culture is very strong: you never go to the doctor. You don’t want to be naked, it’s uncomfortable,” one Somali participant told us. “Going to the GYN, being touched down there, is felt not appropriate,” said another. “Some people say ‘Why should I allow someone to go into my private part?” a Somali participant asked.

Theme 4: Lack of Partner Support (Control Beliefs)

Many participants’ described male partners as potential barriers to care. “Their husbands might not let them go. They are jealous. He may feel threatened or jealous or worried she is having an affair,” said one Congolese participant. “Sometimes a man feels a woman should not go to the doctor because she will expose herself,” reflected another older participant. These narratives invoked discussions of stigma and the role of marriage in defining care-seeking goals and behaviors described above, and were less commonly raised in the younger age groups.

Theme 5: Lack of Resources to Seek Care (Environmental Constraints)

Other significant logistical barriers included transportation, cultural limitations on mobility, cost, insurance coverage, and childcare. “Most have family (to help with childcare and transportation), some don’t. If you don’t have family it’s harder,” one participant stated. “Lack of insurance is the reason why don’t go. There is the obligation to be covered in Massachusetts, but not everyone knows that,” said another, resulting in broad agreement. Themes of environmental constraints and lack of resources were more commonly raised by Somali women, though logistical barriers were prominent in both groups.

Discussion

Our study elicited perspectives on structural and social barriers to utilizing gynecologic care in the years following resettlement in the United States. Using the Integrated Behavior Model, the factors most relevant to our participant perspectives in care fell under categories of knowledge/salience, experiential attitudes, normative beliefs about others’ expectations, perceived control of partners, and environmental constraints. Personal agency and self-efficacy did not frequently enter narratives around why participating women did not frequently utilize gynecologic care in the United States.

Consistent with research demonstrating limited acquisition of gynecologic services and delayed registration for prenatal care among African immigrants in high-resource settings [25,26,27,28] our participants expressed a common theme of gynecologic care being most acceptable in the context of pregnancy. While some participant responses did touch upon the issue of acculturation, because we did not formally collect information on acculturation in our demographic assessment, the role of acculturation will need to be further explored in future research.

Participants expressed rational preferences to not access routine gynecologic health care services between pregnancies, a known challenge in the broader population as well, with up to a third of women not accessing routine primary care prior to or between pregnancies in some studies [29]. Factors contributing to these preferences in this population highlight a mismatch between provider expectations of recommended gynecologic care for preventive health, and participant expectations as structured by cultural considerations and norms.

The fear of stigma as a normative belief was consistent in participant responses around avoiding preventive care. This theme of immigrants not being fully accustomed to the idea of preventive care (particularly cancer screening), and therefore underutilizing services has been affirmed in other studies and could be in part related to the differences between economic and health systems in host versus native settings in addition to the cultural barriers [14, 17].

The theme of patient-provider discordance in expectations also helps us to contextualize the frustration of our participants around multifactorial and difficult-to-treat issues such as pelvic pain. The issue of the inability of structures of Western medicine to recognize, validate, and address immigrant narratives of pain has been previously explored in the biomedical literature, including in specifically African refugee communities in high-resource settings [30,31,32] and manifested in our participant responses as well.

Both Congolese and Somali women expressed a preference for female providers, a barrier that can immediately be addressed in care delivery settings and with outreach to affected communities. In our study, Somali women specifically and consistently voiced the need for gynecologic care by a provider familiar with FGM. This preference for FGM-proficient providers is well documented in the literature [17].

Other barriers to care noted by our participants included transportation, social mobility, childcare, insurance coverage, and financial resources. These logistical barriers have also been noted extensively elsewhere in the literature [33, 34]. While federal refugee resettlement programs may provide aid for the displaced during the initial resettlement period, more support for social and economic assistance programs as well as community-based education about insurance eligibility is needed to ease acculturation barriers. Interestingly, our participants did not identify language barriers as a major barrier as has occurred in other study groups [17, 35, 36].

The ability to generalize the findings of our study and determine the prevalence of the perspectives we elicited is constrained by our limitation to two African immigrant communities and our qualitative focus group study design. In addition, all participants were recruited via convenience sampling and were thus linked to those community organizations serving these geographic locations. Findings may not be generalizable to those not of this geographic region, these communities, or linked community organizations. However, it is important to note that specific cultural barriers can never be generalized to the greater immigrant population or across immigrant groups. Of note, the similarities in the domains assessed across Somali versus Congolese ethnicity despite underlying demographic differences were stronger than differences between the two ethnic groups. Potential confounders such as timing of migration and relationship with acculturation and perceived barriers could not be well assessed using our focus group design and may be better assessed using individual semi-structured interviews and surveys in future research.

A further limitation of our study design is the communal context of focus groups. Given that the goal of this study was to generate insight and hypotheses into normative beliefs and behaviors in these communities, this degree of bias may be acceptable. Social desirability of responses could also have been a limiting factor, though we attempted to mitigate the role of social desirability bias by (1) using focus group facilitators and moderators who were not community members, thus addressing the potential role of social stigma within small communities in generating bias when facilitators come from within communities, and (2) separating focus groups by age and ethnicity to allow for comfort within potential subgroups in sharing personal perspectives on issues of reproductive and gynecologic health.

Despite associated constraints, the value of focus groups as a strategy to generate data on health services in minority communities such as immigrant groups, and give these groups a voice in shaping care while minimizing bias and stigma, has been broadly recognized [22]. The limited availability of data on this subject allows our qualitative work to contribute to a more nuanced understanding of gynecologic health care utilization in African immigrant communities in the US. Further research to determine the prevalence of our findings and impact on resultant disparities in health outcome is imperative.

New Contribution to the Literature

The current study elucidates specific barriers to seeking and accessing reproductive health care in a population at high risk for disparities. Findings can inform structuring of programs that are more directly responsive to patient and community-identified needs. In our setting, these findings will directly inform service delivery in women’s health through provider education, staffing and priority setting in a dedicated Women’s Refugee Clinic where obstetric and gynecologic services are offered, and in shaping community engagement activities. We aim to increase the number of providers who are comfortable with the health consequences and management of FGC, and who practice trauma-informed care. Central to trauma-informed care is building a trusting relationship with a client prior to invasive examination or testing. The Women’s Refugee Clinic will continue to maintain of a core of female providers who work closely with social work colleagues to ensure access to acceptable care and address resource constraints. Community engagement activities will focus on raising awareness within community and religious organizations on the importance of preventive gynecologic healthcare throughout the life cycle—from menarche through post-menopause. The current study additionally calls for the development of integrated peer support to address social stigma around gynecologic care for prevention and for unmarried and older women. Finally, our participant perspectives call for mentorship and training of primary care clinicians and development of integrated informed systems in the broader primary care settings where many immigrant women in our area receive preventative gynecologic services.

In short, centering of patient and community perspectives on care may allow health providers to address the predisposing, enabling, and social determinants of gynecologic health care utilization in recently arrived African immigrants and refugees. Our findings emphasize the importance of culturally humble communication, engagement with issues of stigma, presence of acceptable providers with specific skills, and tailored care delivery innovation prioritizing individualized counseling, communication, and transparent setting of expectations in order to invite improved engagement in care.