Background
The provision of adequate antenatal care and skilled attendance at delivery are widely accepted strategies for preventing infant and maternal mortality and morbidity [
1‐
3]. Adequate antenatal care is associated with better infant survival and is an important determinant of safe delivery. While antenatal care cannot predict all potential obstetric complications, it is widely accepted that antenatal care presents opportunities to recognise and identify pregnancy risks, provide education about recognising and acting on danger signs, and monitor and support women’s health [
4‐
6]. The risk of maternal death is highest immediately postpartum and in the following 48 h [
7]. The presence of skilled birth attendants, whether a woman gives birth at home or in a health facility, is a vital intervention for preventing maternal and infant mortality [
2].
There are many factors affecting the uptake of antenatal care and skilled birth attendance in low and middle-income countries (LMICs) [
1,
4,
8‐
11]. In Vietnam, previous research has shown that factors such as belonging to an ethnic minority group, low income, low education, and living in a rural area are significantly negatively associated with maternal health care utilisation [
12‐
14]. Ethnic minority status has been highlighted as a key structural determinant of inequity in health outcomes in Vietnam [
15]. This is within a global context where ethnic and racial minorities in high-income countries as well as LMICs experience inequities in regards to health outcomes generally [
16‐
18], and often maternal, neonatal and child health specifically [
19‐
24].
Ethnic minority women in Vietnam are much less likely than those from the Kinh majority to attend antenatal care and to give birth with a skilled attendant present [
25]. Low levels of maternal health care utilisation have also been attributed to living far from health facilities and lack of access to transport, however this overlooks the fact that utilisation of services is often low in facilities that are in close proximity to villages [
26,
27]. Traditional customs and cultural differences between ethnic minority patients and health professionals are also cited. However, representations of ethnic minority practices as barriers to service utilisation can play into stereotypes that ‘other’ ethnic minorities [
27]. The scapegoating of ethnic minority practices also suggests a reluctance to examine issues of satisfaction, quality, and appropriateness of health services that may contribute to low levels of service utilisation [
27].
Dien Bien Province (DBP) is a mountainous province of Vietnam, predominantly populated by ethnic minority groups [
28] who experience poorer health and economic outcomes than the Vietnamese average [
29]. Maternal and child health outcomes in DBP are particularly poor [
29‐
31]. Family networks play an important role in both facilitating and delaying the decision to seek maternal health care, particularly during labour, and that there are opportunities for community education around facilitators and barriers to seeking care, particularly preventive maternal care, i.e. antenatal care and skilled birth attendance/facility-based delivery for an anticipated normal birth [
32]. While there has been research describing the barriers to accessing maternal healthcare services [
12‐
14,
26,
27], there has been little research conducted that analyses how and why ethnic minority women access maternal healthcare services [
27]. One previous study exploring utilisation of maternity services did not include communities located further than 3 km from a district hospital [
27]. This study uses a qualitative approach to explore how and why ethnic minority women in DBP currently use and do not use maternal healthcare services, in order to illuminate the factors that influence ethnic minority women and their families in their decisions to seek and not seek maternal healthcare services, and the barriers and facilitators to preventive care seeking.
Discussion
These results show that ethnic minority women’s reasons for accessing and not accessing maternal health care are multifaceted and complex, with barriers beyond the physical accessibility and availability of health facilities. The main themes from our thematic analysis point towards three key findings: 1. the perceived role of health facilities generally is to provide treatment for illness or problems. This perception was found in both community members and health professionals. In the maternal health context, this perception can result in women and their families not seeing the value of accessing antenatal care and facility-based delivery. 2. Inequities exist between ethnic groups, with some communities (particularly Hmong communities) overlooked by the modernising narrative (i.e. the perception that traditional practices have been progressively and increasingly replaced and/or complemented by modern, medicalised practise), and continuing with their traditional practices. 3. Women’s and health professionals’ perceptions of quality of care in health facilities is an important factor in determining which services are utilised. Perceptions of low quality of health stations generally resulted in women either choosing or being referred by health professionals to the District Hospital to give birth, with many of those women unable or unwilling to travel making the choice to deliver at home. We do not intend to imply that delivering at home is the ‘wrong’ choice. Rather, this is an equity issue in that some women and families have the option to deliver in hospital and others do not.
The perceived role of health services as treatment-oriented rather than preventive is not unique to this population [
40], and may speak to a lack of clarity about the role of health stations and health services more generally in DBP. Ethnic minority women who participated in this study generally had a non-medicalised view of pregnancy and childbirth as normal, healthy physiological states, and this was linked to their views about accessing health services. This relationship has been found to be a factor in the non-utilisation of health facilities elsewhere. Qualitative meta-syntheses of the evidence on antenatal care utilisation [
4] and facility-based delivery [
41] also found that women in LMICs generally viewed pregnancy as a healthy state, and so saw little reason to visit health facilities or consult health professionals during pregnancy. They also found that women resisted risk-averse approaches to maternal care and health care generally [
4].
In Vietnam, a recent United Nations Population Fund (UNFPA) report on barriers to accessing maternal health and family planning services [
26] found similar views among ethnic minority women across the country. The report collected data from 27 ethnic minority groups in six provinces, including the two provinces that neighbour DBP and have similar ethnic makeup and mountainous terrain. Qualitative findings from this report suggest that ethnic minority women typically access health services when they experience a complication during pregnancy, but otherwise do not find antenatal care necessary. Women who participated in the UNFPA’s qualitative study also made comments consistent with our finding that commune health stations were perceived to be primarily treatment-oriented, and associated with illness, rather than health. The UNFPA also found a belief that antenatal care and ultrasound examination could determine whether pregnancies (and subsequently labour) would be ‘easy’, indicating that further antenatal care was unnecessary, and the baby could be delivered at home, or ‘complicated’, indicating that a facility delivery was necessary [
26]. We found that fear and worry were also a motivating factor for women who decided to give birth in a health facility, for both primiparous and multiparous women. The UNFPA also found that fear and lack of experience was a factor in choosing to deliver at a health facility for nulliparous women; no women in their sample had delivered at home, and then subsequently in a health facility [
26].
There is evidence from countries including Sri Lanka and Malaysia that suggests that as facility-based maternity care becomes available, women tend to stop using traditional, home-based maternity care [
42,
43]. Bohren and colleagues’ synthesis of the qualitative evidence on facilitators and barriers to facility-based deliveries in LMICs [
41] also found the ‘desire for modernity’ [
41] p.71 to be a factor driving women in some contexts towards a facility-based delivery, which they perceived as contemporary and aspirational. However, this shift has also been found to be contingent on facility-based care being accessible and of good quality; proximity and availability of services are not sufficient [
26,
27]. This behavioural shift was aided by initiatives such as free transportation to health facilities and robust quality assurance measures [
42]; lack of transportation and poor quality services have been found to be barriers to service utilisation among ethnic minority women in Vietnam. [
26,
32,
44].
Our data found that Hmong participants, and health professionals working in predominantly Hmong communes, were less positive overall than Thai participants when speaking about the state of maternal health care and outcomes in their communities. Findings indicated a lack of available, accessible, acceptable, and affordable resources for Hmong people in DBP. In Vietnam, ethnic minority groups as a whole have much lower antenatal care attendance and skilled attendance at birth coverage than the overall national indicators and the Kinh majority [
25], but there is a lack of research that disaggregates results and examines disparities between ethnic minority groups. The UNFPA found that among 27 ethnic minority groups surveyed, Hmong people were among those with the worst maternal health indicators (e.g. percentage of women attending antenatal care, percentage of births attended by skilled personnel) [
26]. Research on communication between primary health care professionals and ethnic minority women in the maternal health context also indicates that Hmong women face greater communication barriers with health professionals that require culturally and contextually targeted intervention [
33].
Participants (both community members and health professionals) mentioned several indicators that suggest the perception of low quality of health facilities, including long waiting times, the absence of doctors, lower level of staff expertise, no ultrasound availability, cramped facilities, absence of appropriate equipment, and lack of hygiene. We found that women who lived close to the District Hospital often preferred to attend antenatal care at the hospital, rather than the commune health station. The perception of poor quality of commune health stations, and the relative perceived high quality of the hospital was suggested as an important reason for this preference. Some women also accessed private facilities, for reasons of quality and convenience. The UNFPA also found that the main reason for the use of private facilities over commune health stations was the perceived low quality of health station services generally, with specific mention of the lack of availability of ultrasound examination, which was often used to determine whether a woman should deliver in a health facility or at home [
26]. The emphasis on ultrasound examination is indicative of a wider trend in Vietnam. A study of obstetricians’ views of ultrasound use in pregnancy found that obstetricians perceived that Vietnamese women associated ultrasound with pregnancy management at the expense of other clinical examinations, resulting in missed opportunities to identify potential pregnancy complications [
45]. This complements our finding that women who could choose between services often chose the District Hospital or a private facility, citing the availability of ultrasound as one of the reasons motivating their preference. Some health professionals also cited the lack of availability of ultrasound at health stations as an indicator of lower quality care, both for antenatal care and delivery services, as staff doubted their own ability to detect complications and so pre-emptively referred all labouring women to the hospital.
For many of the women in our sample, the District Hospital presented an accessible, acceptable alternative to giving birth at the commune health station. The UNFPA found that a minority of their sample delivered at a health facility, and of those the majority chose to deliver at a District Hospital [
26]. However, the relative direct and indirect costs of delivering at the hospital are considerable, and out of reach of many, with an average cost of USD 130, compared to USD 20 at the commune health station, and USD 10 at home [
26]. For women who are unable to access the District Hospital due to reasons of accessibility, affordability, and acceptability, their options are limited to poor quality, possibly unacceptable services at the commune health station, or home. Home births provide the psychosocial benefits of culturally appropriate family-provided care, usually without a skilled birth attendant present [
26]. As such, women may perceive that there is little advantage to giving birth at the health station, compared to giving birth at home, and continue to see home as their best and/or only option [
25‐
27].
Strengths of this study include a heterogeneous sample, involvement of health professionals and community members, a rigorous analysis process, and the involvement of local collaborators. This study had several limitations. Firstly, Vietnamese is not the first language of the ethnic minority people living in this community. All health professionals and most ethnic minority women who participated in the study spoke Vietnamese; some ethnic minority women needed to speak through local interpreters. The use of local interpreters may have resulted in some distortions in women’s responses, either self-imposed or interpreter-imposed. This is a cross-cultural study, and as such, some responses may have been misinterpreted by the authors. We have attempted to limit misinterpretations by conducting an independent translation of audio data and collaborating with a Vietnamese co-author. Any actual or potential misunderstandings were discussed by authors in regular meetings during data collection. Additionally, self-reported practice in interviews and focus group discussions may differ from actual behaviour, and there may be a related element of social desirability bias. We have tried to minimise this through use of a neutral facilitator and assuring participants about confidentiality. Finally, due to logistical and language constraints, systematic member-checking [
46] and verification of our data and themes was not possible.
Conclusions
The challenges to achieving equitable access to maternal health services are numerous and complex, and barriers exist on both the supply-side (health facilities) and demand-side (communities). However, barriers to access and improved communication between health professionals and communities are often perceived to exist mainly on the side of ethnic minority communities [
27,
33], and are based on cultural stereotypes and assumptions about the cultural ‘otherness’ of ethnic minorities [
27].
Health promotion approaches in DBP are often focused on increasing service utilisation by ethnic minority people, and can be didactic, one-way, and paternalistic [
33,
34], perpetuating a health professional-centred approach in which health professionals tell women what they need rather than ask them what they want. This study, and others, have found that although health facilities are technically available and accessible to women, these services are likely to be underutilised if they are perceived to be of low quality [
4,
26], and are not appropriately aligned with women’s social and cultural context as ethnic minority women [
4,
27]. The evidence around interventions to provide culturally-appropriate maternity care is mostly from high-income countries and of low-quality [
47] but it shows that culturally appropriate interventions have positive effects on the utilisation of skilled maternity care, particularly antenatal care [
48]. Simple interventions, such as respecting certain preferences, e.g. accommodating traditional birthing positions and allowing relatives to be present at delivery (health facilities refusing to accommodate these practices has previously been identified as a factor that discourages ethnic minority women in Vietnam from facility-based deliveries [
27]) can have a positive impact on patient satisfaction and service utilisation [
49]. Interventions should be designed based on empirical data, and with the input of affected communities through participatory approaches [
50]. Despite the low-quality evidence, the WHO has made a strong recommendation for providing culturally appropriate skilled care, and highlighted the need for ongoing dialogue with communities in defining culturally appropriate, high quality care that incorporates the communities cultural preferences [
47].
Services will continue to be underutilised if the perceived benefits of attending a health facility are not seen by women and their families to outweigh potential harms and costs [
26,
32]. Those who have the means to bypass commune level facilities may access higher quality facilities that are further away, potentially reinforcing inequities. The desirability of existing services in remote, difficult to access areas can be improved through addressing and improving the quality of local care, staff training, and communication between health professionals and communities at the primary care level [
51].
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.