Setting of the study

Two different communities from the hill and the Terai–the flatland region of Nepal bordering southern India—regions were selected. In the hill region, Dhading District was chosen, as the place is well known for its success in implementing the FCHV programme despite having a difficulty in terms of the distance to the nearest facility with some places located one to five hour walk away [23, 24]. In Dhading, Gajuri Village was chosen as a main study site with some additional interviews held with people from surrounding villages. Each village unit (the smallest unit of village) has one FCHV with a mix of ethnic groups: Bramhin, Dalit, Tamang and Chepang, which allowed us to capture diverse views.

In the Terai, Sarlahi District was chosen because the place has easy access to healthcare services compared to the hill region. The number of FCHVs in the Terai is dependent on its population, with one FCHV for about every 400 people. The district has a total population of 769, 729 with ethnic groups such as Tamang, Kshatri, Muslim and Madhesi [14]. Madhesi has its own caste division, which is beyond the scope of this paper. Yet inclusion of these diverse populations allows us to compare and contrast experiences of different ethnic communities regarding the use of healthcare services.

Research ethics

Ethical approval was obtained from the Nepal Health Research Council on the 10^th May 2013 (registration number 32/2013). Informed consent was obtained from all participants. The purpose of the research was explained to each participant and each was provided with a ‘participant information sheet’ and a ‘written consent’ form. They were informed that their participation was voluntary. Written permission was sought from everyone. However, for the participants who could not read and write, the first author (SP) read out the informed consent form and obtained oral consent. The names and exact age of the participants have been removed from the manuscript. The individuals in this manuscript have given written informed consent (as outlined in PLOS consent form) to publish these case details, when possible.

Data collection

Data were collected between April 2014 and September 2014 in two regions of Nepal using qualitative methods: interviews, FGDs and field notes. In total, 71 individuals participated in this study. Interviews were conducted with 20 FCHVs, 26 service users, and 11 health workers who had supervised the volunteers. An additional 15 FCHVs were included in four FGDs. Table 1 shows the number of participants in interviews and FGDs from the Hill and Terai regions.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Participants involved in interviews and focus group discussions.

https://doi.org/10.1371/journal.pone.0217070.t001

A semi-structured topic guide was used to collect the data (Panday et al., 2017). Both Nepali and English versions of the topic guides are attached (S1 Appendix and S2 Appendix) and show questions used in the previous doctoral study [13], from which the current paper is developed. Most study participants were interviewed because they were based in remote areas, e.g. some volunteers were living a six-hours walk from the health centre (Table 2). All participants were purposely selected to reflect diversity in age, education and years of volunteer experience (Tables 2 and 3).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Socio-demographic characteristics of Female Community Health Volunteers (FCHVs).

https://doi.org/10.1371/journal.pone.0217070.t002

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Demographic characteristics of service users (pregnant women or mothers).

https://doi.org/10.1371/journal.pone.0217070.t003

Table 2 shows FCHVs represented diverse ethnic groups with a majority being Brahmin (n = 17) followed by Tamang and other ethnic groups. There was no FCHV from the Dalit community.

Table 3 shows a majority of service users represented ethnic minority groups. In the hill villages, users were from Dalit, or indigenous communities such as Chepang, Tamang or Bhujel. In the Terai, they represented either the upper caste (Brahmin, Kshatri) or minority ethnic groups (Madhesi and Muslim), who were often illiterate. Service users were chosen if they were pregnant or mothers of children under two and living in the same area as the volunteers. Local health workers were also interviewed to reflect their views on volunteers’ services, and included individuals (6 males and 5 females) from both public health centres and Non-Governmental Organisations (NGOs) working in the study areas.

All the subjects (71 individuals) the first author approached participated in the study. While there were times when the subjects were a bit sceptical about the study, they agreed to participate as soon as the primary author explained to them about the research objectives and why she was undertaking the study. In total, only 15 women volunteers, who had to commute to participate in interviews, were paid each with NRs500 (US $3.46). The volunteers were reimbursed because the first author was able to secure a travel grant of US $1543 (£1180) from the University of Sheffield to cover the cost of local transportation for data collection within Nepal. Remaining interviews with health workers and service users including FGDs with volunteers were arranged at participants’ convenience and were not paid. They were held mostly at respondents’ homes but they were also undertaken in gardens, nearby fields or local teahouses. When possible, some snacks and refreshments (tea, biscuits and cold drinks) were arranged for the study participants.

All interviews and FGDs were conducted by SP in Nepali and ranged in duration from 15 minutes to about one hour. During the FGDs, a local assistant helped with note taking. A local assistant was hired in each study region to accompany the researcher and to assist her to locate study participants. Table 4 shows the demographic characteristics of health workers.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Demographic characteristics of health workers.

https://doi.org/10.1371/journal.pone.0217070.t004

Table 4 shows health workers represented a diverse ethnic group.

During the data collection, SP tried to ensure that her position, as an upper caste educated female from a foreign country, would not influence the participants’ complete disclosure of events, as social background is often reported to influence data collection and interpretations [25]. The researcher’s gender and nursing background might have given participants a feeling of being at ease, because both volunteers and service users openly discussed the reasons behind the underutilisation of healthcare services. In addition, her familiarity with the study environment and her previous research in the country might have helped her to approach the participants in a friendly manner, as all people approached to take part agreed to participate.

Data analysis

The interviews were transcribed with help from local assistants. SP listened to the recordings and translated them. The data set was stored in NVivo10 for analysis. Data was analysed using thematic analysis [26, 27]. The data used in this paper is a part of the doctoral thesis that explored the role of volunteers in maternal health service provision in rural Nepal [13]. A key finding of that doctoral study was that the ethnic minority groups often did not use volunteers’ services despite their attempts to engage with them. Hence, this paper focuses on the perceived barriers to use healthcare by ethnic minority groups.

In order to increase the study’s rigor, multiple coding was applied to part of the data [28]. The third author coded a selection of the early transcripts in English independently and the codes were compared with those of SP and any discrepancies were discussed and resolved. In order to ensure internal validity, the data were triangulated across research methods (interviews and FGDs), respondents (FCHVs, their services users, and the local health workers) and study sites (hill/Terai). The convergence of themes across different groups of participants, places or methods was meant to enhance the validity of the research [29], but we also included divergent views. Our aim was to make the data analysis more comprehensive [25] thus providing a broader understanding of the use of FCHVs’ services. The study findings are illustrated with appropriate anonymous quotes using the following identifiers: HW (Health Worker), D (Dhading) and S (Sarlahi) each with the number of the participant.

a) Lack of knowledge among service users

While a majority of service users approached biomedical services (hereafter simply referred as healthcare services), many service users did not use such services despite their availability through volunteers or healthcare centres. Some volunteers were concerned that their service users particularly from indigenous ethnic minority groups such as Chepang and Tamang in the hill villages, and Madhesi and Muslims in Terai did not understand the importance of healthcare services and hence did not access them. Many Dalit and indigenous mothers (6 out of 14) in the hills delivered at home, compared to only one Muslim mother in the Terai (Table 3). This happened despite referral to health centre by the volunteers. Also a volunteer reported that some parents were reluctant to immunise their children:

“I walked two to three hours up hill to provide polio drops. The woman did not accept the medicine. She said, ‘Our children do not need medicine, they will survive’. It took me three hours to reach them. I did not go there through my self-interest, yet I could not do anything about it.” FCHVD7.

This was consistent with views of some service users for whom the concept of modern medicine was new and hence they did not see its value. A Tamang mother reported that she had not vaccinated her children:

“People in the village say, “why do you need to vaccinate? People from the past are surviving without a single injection.” For this reason, my two kids are not vaccinated.” WomanD10.

Another Tamang woman reported that she did not need any contraception, as she had learned from her previous pregnancies that she would not get pregnant for the next two years:

“I won’t menstruate till the baby is two years. I menstruate exactly after 2 years then I conceive. It has always been like this. I have four kids and was the same for all. Now, I am on the 11 months.” WomanD10.

Health workers also revealed a similar attitude toward ethnic minority groups. They reported difficulties in communicating and dealing with Chepang, who reportedly had substantially lower level of awareness of hygiene and healthcare compared to the general Nepali population. A health worker commented:

“It is mainly associated with the level of awareness of healthcare services among people. In this ward number 4, hygiene is very poor among the Chepang community. The lack of enthusiasm for their work among the Chepang leads to low participation in health related activities by Chepang women.” HW5.

b) Lack of trust in the volunteers

Lack of trust on volunteers’ services was observed. This was related to volunteer’s inability to communicate health messages. Some people, including those reasonably well educated, strongly rejected the volunteers’ offers of medicine. This was reported in the Terai, where volunteers’ distribution of medicine to prevent filariasis infection from mosquito bites was seen as unnecessary or detrimental:

“Some people ordered us, ‘you go away! you go away! you go away! we don’t want to take your tablets. You brought this medicine to kill us.’ The people, who reacted like this were the ones who were better educated. They didn’t take the medicine for filariasis.” FGD3 Participant4.

One of the reasons for such negative reactions could be due to perceived unknown side effects of the filariasis medicine, which might or might not have been true.

“Some children became unconscious, some had fever and some developed typhoid due to that medicine. So people came to us several times to complain about the matter.” VolunteerS9.

Another reason could be due to the volunteers’ inability to communicate the importance of medicine as many volunteers were illiterate (Table 2) and had received only two weeks of training at the beginning of their volunteering. This is evident when some volunteers acknowledged that the current generation of women are more educated and confident than the FCHVs themselves, thus often not needing their help. A FCHV commented:

“There are educated girls in the mothers’ group now [a group of women gathered from a village to gain health information]. I ask one of them to read and ask the rest of the group to listen to her and learn from that. We have become old and forgotten the training. …. If there is anything wrong, then they correct us.” FGD2Participant2.

We also found that volunteers’ inability to communicate the health messages could additionally be related to the quality of training they receive. A health worker admitted that sometimes volunteers were not adequately prepared for their work, as they were simply trained and paid for a day instead of the actual training duration of three to four days:

“The training duration for FCHVs is reduced contrary to the given guidelines, so as to save money from giving allowances to them. If there is training for three or four days then, it would be reduced to one day.”HW11.

The above findings show that the use of health services among pregnant women and mothers is affected by their awareness of healthcare services as well as skills among health workers and volunteers in communicating the importance of healthcare services.

c) Traditional beliefs and healthcare practices

In the hill villages, visiting a faith healer was a common practice among all ethnic groups. In addition, many Dalits in the hill villages instead of visiting health centres for their illnesses, they believed that praying to God would heal them. In the Terai communities, Muslim families believed that use of modern healthcare services would anger their God, so they did not use them.

Almost half of service users reported that they went to faith healers for treatment before they sought care from a health centre. A woman from Chepang community commented:

“I usually go to a faith healer for treatment. If that does not work, then I go to a hospital for further treatment.” Woman3D.

A Dalit woman reported how her baby recovered after her visit to a faith healer:

“My baby was too hot. I took him to a faith healer for treatment. Then he was okay. I didn’t take him to a hospital.” WomanD1.

While some service users believed in faith healers, some Dalits believed in the healing power of the Bible, as they had recently been converted to Christianity.

“Some people told us that if we believe in the Bible, we don’t need to go to faith healers. What to do, I don’t know. They told us we need to have faith on Bible, then our children do not become ill. Or, even if they become ill, if we pray then it will heal.” As we don’t know how to read, they (Christians) read for us.” WomanD5.

In one case, according to the mother from a Dalit community, family members kept praying for the life of the baby who was ill instead of taking him to hospital. Eventually, the baby died.

“The baby gasped once. His body became blue. He abnormally dragged the clothes on his body that was kept for praying, and then he rolled his eyes upwards. Then they (relatives) said that it was not necessary to take him to a hospital. He died in my brother-in-laws house.” WomanD8.

While the Dalit mother regretted that she did not take her baby to a hospital, she appeared to have accepted the untimely death of her son as his fate:

“Even if we had him taken to a hospital he could not have reached to the Thati (close by public place) on time. His time had arrived. He might not have lived even if we had taken him to a hospital. However, he was not taken (to the hospital), and that was a mistake.” WomanD8.

Similarly, in the Terai, the volunteers reported that service users from Muslim families were reluctant to accept immunisation, as they believed that immunisation would anger their god, and this belief was reinforced by mothers-in-laws.

“Some service users say, ‘the wound from immunization makes our God angry. I don’t want to have any injections. Why do you ask me to have an injection? If my mother-in law or father-in-law knew about this, they would be annoyed.’” (FGD2 Participant5).

Another case was that of a pregnant Muslim woman with four children, all delivered at home without any assistance from health workers. She reported that her mother-in-law would move her pregnant abdomen to correct the baby’s position, unaware that such practice could be harmful:

“I never felt a need to go to a health centre. The health workers might look at me and rotate my baby. That much can be done by my mother-in-law at home so I don’t go.” WomanS2.

While the above evidence shows how traditional beliefs among service users and their family members discourage them from using healthcare services, the findings also demonstrate that how women have to rely on approval from their family members to access healthcare services. Below we present how family influences women’s use of healthcare services.

d) Low decision-making power of women

Some women complained of their inability to access healthcare when they needed care. Husbands, mothers-in-laws or relatives frequently perceived healthcare services as irrelevant or unnecessary to them and their families. In addition, traditional beliefs such as sons are better than daughters had forced women to repeatedly become pregnant in order to give birth to a boy.

A volunteer reported how mothers-in-laws were reluctant to accept their advice to support their pregnant daughters:

“The government tells us ‘new mothers should take rest for a month, do not wash clothes.’ We say the same thing in the village. However, some say, ‘this woman (FCHV) has come to us as an advocate to provoke our daughters-in-law so that they would not have to work.’” (FCHVD3).

Similarly, despite awareness about possible complications related to frequent pregnancies, some women kept having children. A pregnant woman, already a mother of four children, reported that she was having another child in order to fulfil her husband’s wishes:

“I got pregnant because my husband wanted another son. I have only one son, and you can’t rely on one…I am thinking to take an injection after the delivery of this child (to prevent further pregnancies). Let’s see what my husband wishes.” WomanS2.

In one case, a husband agreed to abort a pregnancy but the mother-in-law resisted, who wanted an additional male child in the family.

“My husband told me that I am very ill, so if I could not carry on with the pregnancy, then I could choose to abort. When I came home, I asked my mother-in-law, who said, ‘people have 6–7 sons, you have only one son, why do you want to abort? Your body is ill, if you throw the baby then you become more ill or you may die, so it is not necessary to do that.’” WomanD10.

Despite the resistance from family members, volunteers encouraged women to use healthcare services, as illustrated by the flowing quote:

“She reassured me whenever I had any difficulties. She is my relative and she used to repeatedly tell me about health visits and their importance.” WomanS15.

After crossing the above hurdles, when some service users finally managed to present at health centres, they found that they had to face another obstacle, which is presented below.

e) Perceived indignities experienced when using health centres

We found that negative experiences at health facilities among service users discouraged them from visiting health centres. Both volunteers and their service users reported how health workers’ rough attitude towards pregnant women and mothers dissuaded their use of healthcare services. Sometimes health workers were verbally abusive to users as they instructed them what to do and did not pay sufficient attention to their concerns. The following example demonstrates, how volunteers had to accompany mothers at health centres to ensure that they were at a low risk of being treated badly:

“I go with pregnant women to hospital; while the staff scold the women I close their (women’s) mouths. Then I say to the health workers, ‘why do you scold them, these are poor women, they cannot go anywhere.”‘ FCHVD3.

A mother of two children echoed this sentiment. She revealed that she had not taken her iron tablets as prescribed due to their bad taste, but she was scared that the health worker would shout at her if she revealed this. Therefore, she lied to a health worker:

“I did not take those iron tablets. I didn’t like the taste. I had a stomach-ache, so I didn’t use them. I was worried that the health worker would scold me, so I told them that I had taken them. But, I threw them away.” WomanD7.

However, health workers did not reveal anything negative regarding the volunteers or users’ experiences at health centres. The above examples show how the interaction of health workers, volunteers and service users shape the use of healthcare services by pregnant women.