Background
Reduction of maternal mortality is one of the key Millennium Development Goals (MDGs) and is measured by two indicators: a reduction in the maternal mortality ratio (MMR) by three quarters and the proportion of deliveries attended by a skilled health person [
1]. Nepal has made significant progress with regard to the first of these indicators, reducing the MMR from 539 per 100,000 live births in 2006 (one of the highest in the world) [
2] to 170 per 100,000 in 2010 [
3]. Possible explanation for this decline include: the rapid drop in total fertility, increase in average age of marriage, the 2002 Abortion law, and the high proportion of Nepali men working abroad [
4‐
7]. However, one of the major challenges remains the underutilisation of Skilled Birth Attendants (SBA) and the health facilities where they invariably practise [
8].
Traditional barriers to facility birth in low resource settings include costs, transportation problems, and sociocultural norms [
9], and a lack of necessary infrastructure, equipment, supplies, drugs and systems for referral that comprise an enabling environment [
10]. However, recent literature points to staff behaviour as a significant deterrent to women entering facilities [
11]. Research conducted in Nepal suggests that poor quality of services, unavailability, and inaccessibility of SBAs, minimal staff support, lack of medicine and equipment and poor referral systems lead to a low uptake of skilled attendance at birth [
12,
13]. Other constraints to providing effective maternal health services include staff knowledge and competence, lack of proper training and development, inadequate pay, and lack of support from management and colleagues [
14]. The cost of maternity care in an institution, either real or perceived, is also a factor affecting the uptake of facility birth [
15‐
19].
Previous studies have explored women’s views of these barriers to facility birth. Recently, Morrison et al. [
19] interviewed women who had birthed at home in rural Nepal to explore their reasons for delay in seeking care (the ‘first delay’) and concluded that decisions were based on a mix of considerations. These included a lack of family support around the time of birth, difficulty in securing funding for transport, the thought of shaming their family by displaying their body parts in the facility, and past poor encounters. Women’s accounts and experiences provide valuable data in the search for strategies to overcome the barriers to accessing skilled care, however further information from the provider perspective is needed. Although it is recognised that women’s voices from marginal communities are seldom heard, the voices of staff who offer services to these women have had even less recognition in the literature. Reducing barriers to facility based care requires not only a commitment from health care providers but also an understanding of their perspectives and awareness if practices are to be changed to improve care for women. This study sought to elicit the views of health care providers in two maternity units in Nepal, one semi-rural (SR) and one urban (U), regarding barriers to facility birth. Acknowledging the value of the staff’s first-hand experience not only adds to the available evidence but can determine possible improvements for service provision if any deficits are highlighted.
Methods
This mixed methods study comprised 20 qualitative interviews with ten members of staff in each birthing facilty and 25 h of non-participant observation in both facilites. Written informed consent was ontained from all praticipants. A comprehensive literature search was conducted to place the study within the wider body of knowledge.
Literature search
The Advanced Search, with no time restriction, was conducted on MySearch, a federated search engine, provided by EBSCO. Databases covered included Global Health, MEDLINE Complete, Science Citation Index, Social Sciences Citation Index, CINAHL, ScienceDirect, PsycINFO, guided by a librarian. Medical Subject Headings (MESH) and key words included SBA OR Skilled Birth Attendan* (truncation), health personnel or obstetrician* or gyn*cologist* or p*diatrician* or nurs* or midwi* AND developing countr*AND “point* of view*” or perspective* AND Childbirth NOT “wom*n* perspective*” Due to resources the search was limited to English language papers only.
Setting
The study was conducted in one not-for-profit semi-rural community hospital in Kathmandu Valley (hospital SR) and one small private urban hospital (hospital U), in Kathmandu. These hospitals were chosen because they are community hospitals dealing with low risk women from poorer communities. The hospitals represent the type of facilities that the government of Nepal is currently advocating to increase facility birth. In hospital SR all members of staff work full time, usually forty hours per week without night shift or 48 h with a night shift. Nursing staff included axillary-nurse midwives, community medical assistants and health promoters who are supported by receptionists and cleaners for example. There is only one general doctor on site covering 24 h a day, seven days a week. There is no paediatrician, obstetrician or gynaecologist, however every Saturday morning one female obstetric gynaecologist (ObGy) is scheduled to come from Kathmandu city to conduct a gynaecology clinic that includes an antenatal clinic. Hospital SR only has provision for normal vaginal deliveries; they do not have the equipment, facilities or trained personnel to support caesarean sections or assisted deliveries. In addition, they do not have paediatric cover or intensive care facilities to receive and care for sick babies. The hospital provides an on call ambulance manned by two ambulance drivers covering a 24 h, seven day a week service. Neither drivers have had any first aid nor other health related training.
In contrast Hospital U, established 18 months prior to data collection, is located near a major and extremely busy intersection in Kathmandu. The hospital is open 24 h a day, seven days a week. The facilities include a four-bedded antenatal room where women also labour, a delivery room that can accommodate one birth at a time, and a separate four bedded ward that doubles as both a postnatal ward and mixed-sex medical/surgical ward. There is also an operating theatre. Additional services available between 4 and 6 o’clock in the evenings include obstetric, paediatric and anaesthetic cover but there are no intensive care facilities. There is no ambulance service.
Data collection
Data were collected through one-to-one, face-to-face, semi-structured interviews with health care providers, conducted over a period of one month in two hospitals (September 2013). Non-participant observation was used to provide further insights into the barriers portrayed by the health care providers.
A purposive sample of staff was interviewed on each site. Staff ranged from SBAs, including auxiliary nurse midwives and doctors, to support staff such as laboratory technicians and receptionists. Questions for the semi-structured interviews were developed in English and were guided, in part, by themes emerging from the literature around barriers from women’s perspectives. The questions were pre-tested on members of the research team in English. Once translated by a native Nepali speaker, questions were further tested and reviewed by three Nepali colleagues and edited accordingly prior to the interviews being conducted. Interviews were conducted in English, where possible, with those staff whose English was good enough, in order to reduce any bias arising through interpretation [
20]. A Nepali interpreter who had received all her schooling in English was used for non-English speaking staff. Interviews were audio-taped (with permission) and, where necessary, translated before being transcribed. Four interviews in Nepali were transcribed twice by separate translators for quality control.
Birth registers were reviewed to determine how many women used the facility for birth and the outcomes for both mothers and babies. Both sites were able to produce evidence from their respective birth registers covering the last 12 months up to the point of the research taking place.
Non-participant observations of women and staff interacting in both facilities were also undertaken; however few interactions were witnessed due to low numbers of labouring women accessing either facility. All kinds of staff interactions were observed, and no members of staff objected to being observed. These observations were captured through field notes taken as events occurred or shortly afterwards. Additional observations were undertaken at an antenatal clinic delivered on site at Hospital SR and an outreach antenatal clinic provided by staff from Hospital U.
A thematic approach to analysing the qualitative data was used [
21]. Three researchers (LM, JI and EvT) coded all of the transcripts independently. Emergent themes were then discussed and agreed and during this process it became apparent that key themes corresponded with Thaddeus and Maine’s three Phases of Delay so quotations, extracted from participants’ (P) transcripts, are presented using this framework to illustrate themes [
22].
Ethical considerations
Ethical approval was given from both the Nepal Health Research Council and Bournemouth University’s ethics committee. Each participant received a study information sheet, written in Nepali assuring that confidentiality would be maintained.
Discussion
The three phases of delay emerged from the data rather than being used to tailor pre-determined interview questions. The interviews highlight a number of key barriers to all phases of delay that are common to both sample sites, despite their different geographical locations, and that resonate with the literature predominantly collated from women’s perspectives [
10,
11,
18,
19]. Less commonly reported themes included poor services, embarrassment and financial issues related to the most recently employed financial incentives [
31] (first Phase Delay); by-passing the facility in favour of one further away (second Phase Delay) and both absence of an enabling environment and disrespectful care (third Phase Delay).
Several barriers, such as poor services, were linked to several different phases of delay. For example, participants believed some decisions to access care were influenced by women’s previous poor experiences with their local facility or by stories told by others who had used the facilities and recounted poor services during the third delay. Thus influencing other people’s decisions to either birth at home (the first delay) or by-pass facilities and go elsewhere for subsequent births (the second delay). We know that pregnant women are often influenced by the experiences and stories of friends and family, however in Nepal the mother-in-law and husband tend to be the most influential people [
8]. Ingrained cultural factors are very difficult to change and there is a need for greater awareness and solutions that focus on cultural competence rather than change. For example, Nepal might be encouraged to train more female doctors in obstetrics and services encouraged to employ them.
A number of factors appear likely explanations to why women by-pass their local birthing facility in favour of government maternity hospitals further away. A practice initially identified with ill people in Africa [
32]. Research conducted in rural Tanzania [
33] recognised women frequently by-pass both their local antenatal and birthing facility in favour of services offered at higher level facilities further away regardless of the cost and effort involved in accessing them. Women in Kruk et al.’s study perceived quality of care to be poorer and staff to be less trustworthy at their local facility. Similarly, participants in our study noted that some women having initially accessed local facility antenatal care went on to complete the remainder of their antenatal care and birth in a tertiary hospital in the capital. It was not possible from our study to determine the reasons women had for doing this; however staff hypothesised that it was to do with the greater availability of services and equipment in Kathmandu. Knowing the staff in the local hospital and being embarrassed was also suggested to be a reason for by-passing the service.
What was evident from observations and interviews was the absence of the ObGy in hospital SR and the limited availability of the ObGy in hospital U. However, one very practical issue in both hospitals is that ObGyn doctors also undertake fulltime work elsewhere in main government birthing facilities in Kathmandu and regularly do not turn up for scheduled clinics. This practise is likely to be a result of the lack of qualified ObGy’s nationally coupled with the opportunity to subsidise incomes [
34]. Moreover, participants perceived many women had time constraints due to competing demands of daily family life, house chores and working in agriculture. It is possible that the lack of emergency obstetric care locally means people think ahead and plan to go directly to larger facilities in Kathmandu which they may perceive to offer more reliable services assuming they have the money for transport and other associated out of pocket expenses.
Further to this staff relayed a major incident at hospital SR and were aware that stories of this event had been circulated amongst community members and possibly contributed to by-passing which in turn influenced a reduction in their monthly birth numbers.
A notable number of quotes also highlight the use of the term “better care/treatment” used by staff to describe larger tertiary facilities and the use of such descriptors when talking to women and their families may also encourage women to by-pass. In addition both staff and communities’ philosophy on what constitutes safe birth and quality services perhaps focus on a more medicalised model of care and the opportunity to get caesarean section.
As previously mentioned, embarrassment was a key finding. It has been reported in the literature that Nepali women feel shy about allowing others to examine their bodies and is tied in with women’s status and their duty not to bring shame on their family (Morrison et al. 2014). This notion of being embarrassed to show private parts to doctors has also been reported as a barrier to accessing local medical care by Nepalese female sex industry workers [
35]. Moreover, several respondents at hospital SR also professed to birthing in a main facility in Kathmandu rather than in the facility that they worked at due to embarrassment and role modelling this ‘flight’ may have also influenced other women to do the same.
Reasons underpinning the decision to by-pass have been postulated. It may be a reflection on perceived poor local services or seemingly better services in the city and further research is needed to ascertain this and to determine to what extent either of these views contribute to the decision and who makes the decision. Having said this, it is clear that there are women who can afford the one hour taxi fare from their semi-rural residence (half an hour from hospital U) to a facility further away. Therefore an unknown proportion of women presumably birth with a SBA albeit at another facility and are disputably exercising freedom of choice.
The effects on by-passing for both the local facilities and tertiary hospitals necessitate further consideration. If the trend to by-pass continues local facilities may be in danger of becoming unviable whilst tertiary staff and facilities, already purported to be overstretched and under-resourced, may find it increasingly difficult to provide adequate care.
One possible solution for the inconsistency of specialist doctor availability for antenatal care is to train midwives adequately to care for low risk women and refer women with risk factors to a reliable tertiary service however this necessitates a reliable ambulance service for transfers and the availability of a SBA for escort.
Financial issues have long been established as a major barrier to accessing SBA care. The Government’s Safe Motherhood Programme, Aama Aurakshya Karyakram, was implemented to alleviate this and offers women 500 rupees in the Terai (plains) and 1,500 rupees in the mountains if they birth in a facility. Concurrently, health facilities receive 1000 rupees (US $11) for a normal birth, 3,000 rupees (US $34) for managing obstetric complications and 7,000 rupees (US $80) per caesarean section [
36]. In addition, having a caesarean section at Hospital U necessitates women and their families to purchase pharmaceutical supplies from the neighbouring private pharmacy at a cost of 4,834 rupees (US $46). In contrast Government tertiary hospitals also offer free caesarean section in addition to the same financial incentives as local birthing units and this could be a further factor influencing by-passing.
Barriers within the hospital were less obvious to staff. Although staff highlighted a number of absent environmental effects necessary for them to perform their duties adequately, one major observation noted was the lack of a clean working environment. Only one respondent mentioned the importance and difficulties in maintaining hygiene. It is possible that no other participants highlighted cleanliness as an issue because this reflects the standard facility norm. There is evidence that for a proportion of rural women, standards of cleanliness experienced in government hospitals is higher than where they are required to give birth at home as settings include cowsheds, in keeping with the tradition of ‘chaupadi’ [
37]. Whether cleanliness is
perceived to be better in the tertiary hospitals compared to the local facility is unclear, although a recent study suggests otherwise [
38].
Prevention of cross infection is crucial to the working environment, yet observations showed the current circumstances to be incongruent with the idea of promoting safe birth in a facility with a SBA and the belief that this would reduce maternal death will be impeded if cleanliness and the prevention of cross infection are not addressed adequately. In other words, women birthing in facilities may reduce the incidence of primary postpartum haemorrhage for example one of the current major direct cause of maternal death in Nepal [
39], but this may be at the expense of increasing the likelihood of women dying later, following discharge, from a hospital acquired infection [
40] or indeed a secondary postpartum haemorrhage triggered by infection.
A further observation was the lack of privacy observed at the urban hospital’s antenatal outreach clinic which may be construed in some cultures as disrespectful but in this particular context no pregnant woman appeared phased by this. What may constitute disrespectful behaviour in one culture may be perceived as acceptable, respectful behaviour in another [
11]. Staff were aware of disrespectful care occurring but not in their facility and none was observed during the study, although it is possible that staff were ‘performing’ for the observer and the lack of child birthing women meant few interactions were witnessed. Having said this, discussion around disrespect care led some staff to identify that they themselves felt disrespected at times by service users. A recent study carried out in two major government maternity hospitals in Kathmandu focused on female nurse SBAs concepts of respectful maternity care reported that they understood the concept of disrespectful care, but because they were overworked they placed women’s safety over comfort as a priority [
41]. Release of the WHO statement on the ‘prevention and elimination of disrespect and abuse during facility-based childbirth’ recognizes the problem and “calls for greater co-operation among governments, healthcare providers, managers, professional associations, researchers, women’s advocates, international organizations and women themselves to end disrespect and abuse during facility-based childbirth” [
42].
Limitations and strengths of study
The scale of this study did not permit the elicitation of women’s and their families’ perceptions which may or may not corroborate with the perceptions of the staff however a number of the findings are supported by both the primary literature review and observations undertaken during the study. A strength of the study is that it did not set out with any preconceived ideas about the barriers from health providers’ perspectives, but it was surprising how well the themes that emerged mapped to the model proposed by Thaddeus and Maine [
22].
One weakness is the use of a translator during some interviews as meanings and accuracy of information may be lost, although the same translator was used throughout providing consistency. Accuracy of translation was verified by two Nepalese translators transcribing four of the interviews independently of each other. On the positive side, using a translator enabled access to the wider staff body whom all have a role to play in providing childbirth services and thus influencing women’s perceptions.
A further strength of this study was the concurrent use of observations and semi structured interviews. Whether observations should be conducted prior to the interviews or vice versa is debatable however, entering into the ‘natural world’ and the unpredictability of the nature of this world with regards to the workload of staff for a relatively short period of time necessitated a pragmatic approach and opportunities to interview staff and undertake observations were taken as they arose. Consequently, all participants were interviewed during their working day at their convenience. On several occasions this meant staff members were interviewed one after another, as a result, staff were unable to confer with each other although responses to questions were very similar and thus credible.
Acknowledgements
We would like to thank the following people for their invaluable support; All the respondents who kindly participated in the study; Ram Chandra Silwal, Country Programme Director, Green Tara Trust, Nepal; Sarita Khadka, Programme Coordinator and translator, Green Tara Trust, Nepal; Ray Acharya and Baghabati Ghimire and Bhawani Adhikary for translation. Jan Hutt, Academic Support Librarian, Health and Social Care, Bournemouth University, England, UK.
The study was funded by the Wellbeing of Women chairty and the RCM and represents their first International Fellowship Award.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
EvT and VH conceived the study and provided scientific advice on study design and analysis. LM collected data. LM, JI and EvT independently coded transcripts. LM drafted the initial manuscript with the support of EvT, VH, PS and JI who all contributed to the writing of this paper. All authors approved the final version of the manuscript.