The Structure of Attitudes towards Traditional Birth Attendants in Health Facility: A Catalysing Factor for Institutional Deliveries

Background: Skilled attendance at birth (SBA) is a critical intervention for maternal and infant survival, yet women in low income settings deliver outside of health facilities, with no assistance. A successful approach to the scarcity of health workers has been the incorporation of alternative health worker cadres into formal health systems. There is evidence that Traditional Birth Attendants (TBA) can contribute to the survival of mothers and new-borns however, they cannot be a substitute for skilled providers. This paper analyse the structure and attitude towards TBAs and how we can mainstream TBAs into healthcare delivery. This research reports on qualitative research exploration of practices, acceptability, the structure of attitudes towards TBAs, and how to potentially mainstream TBAs into health service delivery. Methods: Exploratory qualitative study design using focus group discussions and semi structured interviews was conducted representing policy makers, health workers, local leaders, primary beneficiaries (pregnant and postnatal women and their male partners), non-direct beneficiaries- male and female in reproductive age group, pre-conception women, other male partners, and household members. Analysis of data was done using qualitative content analysis. The study documented the practices of TBAs in Busia district of Uganda. Results: There is evidence of mainstreaming of TBA into health service delivery. The influence of TBAs at community level is still very strong. TBAs meet vital community needs in supporting women throughout pregnancy, childbirth and the post-partum periods. There is also evidence that, TBAs can work closely with health planners, health professionals and other members of the formal health system as a strong link between the community and the health services if the community and health workers attitudes are addressed. TBAs should be included in community education and mobilisation efforts. Conclusion: The new traffic of TBAs in health facilities has drawn attention of research, ethics, and reflection on the future initiatives. These practices can be mainstreamed into health service delivery and their roles restricted to non-clinical aspects. It will also inform program implementer and policy makers to develop socially and culturally accepted interventions that can improve institutional delivery and contribute to the reduction of maternal and child mortality.

improving child health. Despite this improvement, the progress is still slow and Uganda. To promote health and achieve universal health coverage and access to quality health care, including sexual and reproductive health care services [5]. Maternal and child health conditions carry the highest total burden of diseases with perinatal maternal conditions accounting for 20 percent of the total disease burden in Uganda [6]. Approximately 6,000 women die during pregnancy and childbirth each year, and vast numbers experience severe adverse consequences [7].
Skilled attendance at birth (SBA) is a critical intervention for maternal and infant survival, yet women in low income settings deliver outside of health facilities, with no assistance. Research shows that SBA in rural and fragile regions is persistently low [2]. In seeking for alternatives, women often find themselves in the hands of community owed resource persons such as the Traditional Birth Attendants (TBA). Birth is a critical time for maternal and newborn survival and the day of birth is the greatest risk of death and disability [8]. Two thirds of maternal and newborn death occurs at delivery and in the immediate postpartum period [8]. Unfortunately a significant proportion of maternal and newborn death is often linked to delivery with TBAs [9]. Anecdote findings from our recent study about rational for choice of delivery with the TBA or Health facility indicate various reasons for women's decisions including women having limited understanding of the importance of health facility delivery, a finding also noted elsewhere [10]. Delay in seeking care could be due to the fact that women chose to deliver at home or at the TBA [11].
One of the strongest determinants of infant, under-five and maternal mortality rates across countries is the number of health workers available per head of population. The World Health Organisation and UNICEF strongly believe that TBAs can contribute to the survival of mothers and new-borns by facilitating access to needed information, clinical services and support however, they cannot be substitutes for skilled providers. In other words, TBAs, like other community health workers, can effectively convey vital information to families and communities in culturally appropriate ways that will help them to recognise danger signs during pregnancy and know where to go for help or referral [12].
Health facility limitations such as limited coverage by skilled providers, poor quality of services, ineffective referral systems, limited geographical access, health care delivery system failures, lack of drugs, supplies and equipment grossly limits access to care for rural women, further pushing them to TBAs [13]. The persistent shortage of health worker in health facilities standing at 5 health staff (Doctors, nurses and midwives) per 10,000 people, and the high turnover of health staff explains the ongoing practice of TBAs. The reported 51 percent deliveries with TBAs in rural Uganda are partly attributable to health care disparities between the rich and poor across the urban and rural areas [7]. The cost of delivering from private units which in some cases are in close proximity to the women, are equally limiting. In the public facilities there are no user fees but the out of pocket expenses are considered to be high in rural areas [14]. Despite favourable policies and guidelines to strengthen skilled attendance at birth, poor rural women have limited access to formal health services so they explore traditional alternatives.
Attempts have been made in Uganda to ban the activities of TBAs. A case in point; the national MOH distances it's self from their practice [15]. There is neither guidance nor regulation of their practice or reinforcement of their banned practice. As a result TBAs continue to practice, attending to 18 percent of the women, despite their limited skills to avert the most common complications seen during delivery, such as haemorrhage, prolonged labour, and hypertensive disorders of pregnancy.
In the presence of HIV and other blood born diseases the questionable hygienic practices of TBAs is neither safe for them nor the women who they attend.
Proponents of TBA's continued practice condone the banning of their practice, urging that in some places where women cannot access health care facilities, TBAs are the only alternative [16]. Further, traditional and cultural beliefs deterring women from delivering at Health Centers are still strong. TBAs learn by apprenticeship from the previous generation, it is not surprising that their services are valued by the community. The attitudes of the health care providers draw women away from the health facility [17]. Mothers feel comfortable delivering with TBAs with whom they share culture and social beliefs [18]. Previous studies have recommended the integration of TBAs into the formal health systems. Taking into account into the MOH position it is a sensitive issue to have the TBAs participating in care at a health facility without proper compliance to the structure. This research reports the qualitative research exploration of practices, acceptability, the structure of, and attitudes towards TBAs, and how to potentially mainstream TBAs into health service delivery. This study will inform how we can better structure the delivery of service by TBAs at the community level in order to address maternal, newborn and child death in Uganda. Mechanisms to mitigate the risks shall also be addressed.

Study design
An exploratory qualitative study design using focus group discussions, key informants interviews with various stakeholders-policy makers, practitioners, local leaders, partners, other house hold members, and beneficiaries was done.

Study site
The study was conducted in three World Vision Area Development Program (ADP) sites, Busitema, Lunyo and North Rukiga in Uganda, where the AIM-Health [Access to Infant and Maternal Health] is implemented. The program is based on health and nutrition 7-11 strategies and delivered at the household by CHW, using the Timed and Targeted Counselling

Global Journal of Reproductive Medicine
(TTC) approach. The CHWs makes three timed visits to households during a woman's pregnancy, one immediately after she has given birth and six more throughout the first 24 months of the child's life. During the household visits, the CHWs deliver message about health promotion, illness prevention and enhanced health seeking behaviour using the TTC as part of a health and community systems strengthening approach. Some TBAs were trained as CHWs. The Busitema ADP is located in Busia District in Eastern Uganda benefiting 25,000 people. Most of the area (90% of the population) depends on agriculture, with 88 percent of the households living on less than a dollar per day. The district has a total of 14 sub counties, 55 parishes and 2 municipal divisions with a total of 8 wards. Busia District has a total of 27 health facilities, including one General hospital, 18 HC II, 8 HC III and 2 HC IV. Ten health centres offer maternity services, only 20 percent of women delivering from the health facilities. North Rukiga ADP is located in Kabale District in South Western Uganda, with a total population of 580,600 in 2010. It is essentially a highland district dominated by the Bakiga tribe. Uptake of preventative practices such as diarrhoea management, immunisation, exclusive breastfeeding and long-lasting insecticide net utilisation was low. The district has 1 government hospital and 48 Health centres (II, III, and IV) these facilities provide health services for both the urban and rural population. Overall district health service coverage is about 68 percent. Over 66 percent of mothers attend antenatal care, but less than a third delivers from health facilities (UBOS, Macro International 2010). The health facilities have only 2 percent of the required personnel (midwives and doctors).

Study population
Pregnant women and women with children less than 18 month participating in the TTC program as well as various stakeholders were purposively selected for the qualitative study. In addition non-beneficiaries such as male partners, preconception women, women in the reproductive age group, and key stakeholders within the catchment areas of the ADPs were also purposively selected. Health care professionals at public, private not-for-profit and private for-profit health facilities providing maternal child related services in the catchments areas of each ADP were also purposively selected.

Data collection
A stakeholder analysis for identification of the key stakeholders was done by investigators in consultation with national World Vision staff and district ADP staff. A prioritized list of stakeholders at national, district, and community level was generated. A total of 15 Key informants interviews were done in each of the ADP with community level beneficiary representatives (2), implementers (2 ADP staff, 4 CHW), Nonbeneficiaries (3-male partner, preconception women, woman of reproductive age, other men), other stakeholders-community leaders (2). Further in each district representatives of the district health team (2) were interviewed, and national Ministry of Health key informants (2) were interviewed. A total of 47 Key informants' interviews were done. Individual interviews were conducted in households or convenient locations agreed upon with the participants. The interviews lasted 30 -45 minutes. Further 15 Focus group discussions were conducted, 5 in each ADP, 5 with primary beneficiaries including; FGDs were conducted in suitable community locations. Interviews were audio-recorded following consent from the participants. The FGDs were moderated, transcribed and independently analysed by the first author and two of the coauthors (EA, PE). The participants in all the FGDs were purposively selected. The groups were homogenous in composition. A FGD guide was used. In all the FGDs the moderator was supported by a note taker. Both FGDs and KII were conducted in English, or one of the local languages; Rukiga, Luganda or Samya according to the participants' ability to understand and communicate. All the transcripts were translated directly into English before analysis was done. Semi-structured interviews with health care providers in health facilities offering maternal child health services, were done. Each interview lasted for 30-45 minutes. An interview guide with open and close-ended questions was used. Open-ended questions were used to explore concepts such as access to MCH services, experiences, views, and attitudes of the providers towards the 7-11 TTC. The guide was designed to investigate providers' experiences about the influence of CHW TTC on referral pathways. The interview guide used was pretested and changes made before data collection. Additional FGD with the TBAs was done; additional topics were included related to structure and their attitudes, motivation, relationships with health care workers and health facilities and their influence in the community.

Data management and analysis
The first author read through all the KI and FGD transcripts several times while making notes on the transcripts to make sense of the texts. Each of the other three co-authors read some of the transcripts. Coding of the data and the analysis was done manually. The unit of analysis was the focus group or individuals. Latent content analysis technique that involves in-depth interpretation of the underlying meanings of the text and condensing data without losing its quality was used. Events, activities, perceptions and explanations were identified and coded by the first author. The codes were grouped into categories, and sub-categories. The analysis was discussed among the research team members and discrepancies on coding Global Journal of Reproductive Medicine and other issues that required clarity were settled by discussion. Quotes that best described the various categories and expressed what was aired frequently in several groups were chosen.
Ethical approval was obtained from Makerere University School of Public Health Higher Degrees, Research, and Ethics Committee and the Uganda National Council for Science and Technology. Permission was sought and obtained from Busia and Kabale Districts where the ADPS are located. Written and verbal consent was secured from all research subjects. Appropriate measures were undertaken to ensure confidentiality. The interviews and FGDs involving the pregnant women were done by a nurse/ midwife research assistants on the team who attended the pregnant participants in case of discomfort as a result of the interviews.

TBAs have learnt from the midwives
TBAs reported previously being ignorant of HIV codes and their implications.
"We were delivering women from the villages not knowing about HIV codes, midwives taught us these codes and advised us to encourage women to deliver from a good environment" (FGD TBAs, Lunyo)

Buffering roles of TBAs
TBAs revealed that they are trained to escort patients and are accepted to stay with them during delivery at the health facility. They also reported being overwhelmed by the numbers of women who were coming to them seeking delivery service at their homes before their training. The TBAs reported that women no longer deliver from home in their communities, that they move at night sometimes to escort mothers to the health facility. They reported having a high work load. "Women used to be afraid that the midwives will beat them but we sensitize them about delivery at the health center, we are also here for women, we keep with the women when labor starts and reassure them" (TBA Leader, Lunyo)

Pre-conception women, men and postnatal women
The TBAs noted that they tell young people the disadvantages of early pregnancy, danger of getting HIV and the risk of uterine rupture. They reported that they rarely encourage condom use. They also reported that men have started to bring their partners to the Health facility for labor and delivery working with the midwives on what is needed, understanding their responsibility to buy requirements for the mothers. TBAs also stated that they advise postnatal women about immunization, nutrition, hygiene and how to prevent malaria. The TBAs working as VHTs reported that during home visits they teach about preparation for birth, importance of health facility delivery and attending antenatal care. They reported observing some changes such as spacing births, reduced use of herbs to hasten labor and using herbs on the cord by women.

Perceived benefits/values to TBA and needs
TBA s reported the main benefit is to strong relationships with the community, stating that they help women and they say thank you, this re-assures the TBAs and brings friendship.
"The TBA has helped, if not for them my child would have died…" (KI Postnatal woman) "We pray for women to go thru delivery well and this has brought about a difference. There was one woman I prayed for and delivered one live baby and a dead twin" (TBA, Lunyo) The TBAs reported challenges with transport and limited financial payment limited. They work as volunteers. Once in a while the women may give them a token of appreciation, but most women know that this is a government health centre and are not inclined to make any payments. The TBAs based at Lunyo HC said that for income, they rely on farming (but have no time to do it) allowances from working as VHTs, or when they are given transport refund for attending district or NGO meetings. They reported that the in charge of the facility gave them the opportunity to participate in outreach immunization and they were able to receive some allowances or food. The health facility does not give them official payment for their work? "The patient may have a bicycle for the pregnant woman in labour the TBA follows on foot… we work for God." (FGD with TBAs at Lunyo HC)

Comparing attitudes, trust and intentions of TBA
In Lunyo sub-county in Busia district, the majority of the participants reported that a number of women were significantly less likely to seek health facility based services without TBA involvement because they were less trusting of health care provider's ability to protect their privacy and providers negative attitudes during birthing. The mothers indicated that, they did not have close interactions with the formal health worker from the time they conceived to the time they deliver. The mothers assumed that the health workers did not know much about them and it was difficult to trust them during delivery. The health workers were most often considered as strangers. The TBAs in contrast were considered to be part of the community because they grew up and lived in the community making it easy to for mothers to trust them.
"I should rather trust my life with a TBA than someone I have never known or lived with. It is not easy as people may think and that is why I have always decided to deliver with a TBA… by the fact that we share the same culture, live in the same community and speak the same language makes it easy for me to trust them" Woman Beneficiary.

TBAs role at the health facility
The TBAs highlighted that they participate in delivery activities at the health facility; they check women and deliver with the midwives. They noted that some women want to immediately leave after delivery but they encourage them to stay. They converse with women after delivery and check on them. They also reported that after the midwife discharges them, they Global Journal of Reproductive Medicine visit the women in their homes for follow up and give advice on immunization.
"We talk with the women in labour and counsel them…We examine them if ready to push we help to prepare the forceps etc. for the midwives….if the midwife is busy we deliver. The midwife told us what to do, we give injection after delivery, deliver the placenta, clean the woman up and take them to postnatal ward….tell the women to breastfeed" (FGD TBAs at HC). The TBAs reported that working at the health facility has helped to improve their status, and they are increasing in number. They also reported giving women customer care and having improved relations with the women, and that woman are happy and come from far because of the good handling.

Attitude of the family members
Family members report that speaking the same language makes it easy for them to communicate with TBAs unlike when they go to the health facility where there could be a mix of languages most especially in Busia which is a border town with many different language groups living there.

Perception of health workers towards TBAs
Traditional Birth attendants were perceived as uneducated by some health workers. According to the health workers, the low level of education of TBAs makes them un-trainable to provide emergency obstetric care or recognise complications during pregnancy and delivery. The health workers perceived the TBA as a major cause of maternal and child mortality in Busia District and that maternal mortality would not be reduced if women continued seeking the services of TBAs. Health workers also reported some delays by mothers in reaching the health facility for delivery caused by the influence of TBAs. The TBAs most often tried to deliver mothers even when they have already noted potential danger. The health workers also noted that it was not easy to change the influence of the TBAs in the community. They stated the best approach was to establish how to incorporate them into the formal health systems. This study also revealed that the TBAs especially those in the AIM program areas were very supportive in making referrals of mothers who they thought were likely to develop complication to the health facility. Some mothers however still refused to be referred to the health facility.
"According to the TBAs they try as much as possible to refer mothers but some of them would refuse to go" (Implementer AIM Health Program) Some of TBAs also actively participated in encouraging mother to deliver at the facility and also offered some nutritional advice during pregnancy. This was mainly noted in Busitema Sub country. In Lunyo HC III, because the community had so much confidence in the TBAs, the TBAs moved to the health facility and worked with the health workers to deliver mothers. Their roles were limited to non-clinical aspects of care. The number of deliveries greatly increased at this facility. The mothers received care from the TBA that they would have received from the TBAs if they delivered at home but the TBA was working with the health workers in a safer environment.

Perceived quality of care at health facilities
The results of the study reveal that most mothers were attending antenatal care at the facility but majority were still delivering with the TBAs. This was noted especially in Lunyo ADP. The mothers delivering with the TBAs were frustrated with the quality of MCH services offered at the health facility. A number of factors contributed to this. The women, who were interviewed, stated that the health workers were rude and did not attend to them in a friendly way during delivery compared to the TBAs who handled them in a very friendly way. There were also very few workers at the health facility to care for the number of mothers. One of the implementers interviewed pointed out that "from what the mothers say at the health centre, the servicesthere is one midwife, sometime you reach there the midwife has many mothers you have to sit and wait, the midwife is very busy so some of them think I might go there, the midwife is very busy and they will not attend to me in time I rather go to a TBA who will be there, I will be alone and she will give me all the attention" Implementer-Busia.
The study results also showed that, because of the constant sensitization under the AIM health program, some women now preferred to attend ANC and deliver at the health facility. The TBAs under this same program were encouraged to refer and accompany mothers for antenatal care and delivery at the facility. Where referrals were made, the deliveries were far higher than where referrals were not made. It was also seen that some mothers had more knowledge about the consequences of delivering with TBAs and therefore preferred to deliver at the health facility. The FGD in Busitema Sub County established that mothers feared to deliver with the TBAs because of the risk of infection OR mothers did not deliver with TBAs because they feared infection.
"The TBA's don't use gloves and they do not know how to handle the babies. They carry them like one carries offal's… They don't know what to do." (Busitema-MNB)

Influence of TBA at community level
Other findings in the study noted that the influence of TBAs was quite strong at community level most especially in the rural areas compared to the urban areas. Most mothers in the rural areas preferred to deliver with the TBAs than at the health facility.
"There are TBAs who are very influential and community rely believes in them and you cannot tell them to go to the health center when their TBA is just near here" (Implementer-Busia)

Strong cultural attachment to TBAs
The challenge to convince people to withdraw from traditional practices in the communities was reported by some local leaders.

Global Journal of Reproductive Medicine
"For mothers who were used to delivering in the villages, being attended by TBAs, not easy to go to the health facility. This will take time to fade" Local leader

Discussion
The results of this qualitative study provided an understanding on the structure and attitude and how to mainstream TBAs into health service delivery. The present study revealed that, the influence of TBAs was quite stronger at community level. The TBAs wield a great deal of power and influence in determining health seeking behaviours. Their recommendations and advice were taken very seriously and the community believed in them. The influence was stronger in rural areas compared to the urban areas, consistent with other previous studies [15,19]. Most mothers interviewed in the rural area preferred to deliver with a TBA than at the health facility. In Uganda over 80% of the women are estimated to be living in rural areas which are poorly facilitated in terms of health facilities. About 40% are within a 5 km distance from a facility that provides ANC, delivery and immunisation services [14,20]. From the study, it was established that pregnant mothers tended to seek the services the TBAs due to their proximity and accessibility within the community. It is also likely that this kind of influence can affect institutional delivery, creating a need for health service providers to utilise the existing influence of TBAs to increase the number of women willing to deliver from a health facility. Meaning that, important and relevant synergies are developed with TBA's role limited to non-clinical practice but to utilize their influence within the community to change mother's attitudes toward going to a health facility to deliver their babies.
The Uganda Ministry of Health tried to ban TBAs' services in 2009 but this strategy failed to work. The poor and marginalized women in rural areas still needed their services and had trust and confidence in them. This is an indication of the strong influence of TBAs in the community. The weaknesses in the healthcare system contributed to the increase in the influence of TBAs. The continuous challenges in the health system to provide adequate health care workers at health facilities, absence of ambulance services, and distance to the facility and transport costs affected the delivery of maternal and child health services [21]. Given that the TBAs were in close proximity were willing to offer their services, mother do not incur any transport expenses, mothers went to them to be delivered.
Despite the weaknesses in the health system, the WHO is advocating for institutional deliveries. The government of Uganda is implementing a policy to ban all TBAs' services which are often looked at by the rural mothers who cannot access the formal health facilities as the only alternative to seek assistance from during delivery [11]. The implementation of this policy might not be effective if the government does not address the underlying health systems issues identified. What seems feasible is to incorporate the TBAs into the health service delivery systems. Training of TBAs to spot early signs of complication and be able to refer mothers appropriately would make a significant impact on maternal health outcomes. However, this training should also address the issues of attitude and team work that has affected TBA/health worker relations.
The TBAs were valued because of their emotional and social closeness to the community which created loyalty and mothers were able to resort to them in the event that formal health services were not available and accessible. It is therefore important for health service providers to utilise this existing influence to improve the experience of mothers during delivery at health facilities. TBAs should be encouraged to send expectant mother to the health facility. This would be an important mechanism for improving referral pathways. In the some areas the TBAs were provided with referral forms and were encouraged to refer mothers with complications to the health facility. This improved referral and delivery at the health facility. Connecting pregnant and postnatal women and their children to skilled care providers for ANC, birth, PMTCT, Immunization requires working together with community owned resource persons including TBAs. Government, medical workers and researchers need to come together to define the role of TBAs in tackling maternal, newborn and child health in fragile rural communities. It is imperative to encourage mother to have a skilled birth attendant during delivery in order to reduce maternal and child death. Equally, a supportive environment should be created to promote the participation of TBAs in the mainstream health service delivery systems. A network of TBAs and health facility workers should be strengthened. Of equal importance is the need for more research on the contexts of TBA participation in care at health facilities in order to acquire and establish an evidence base that supports the possible benefits of establishing TBAs roles within health facilities. The health sector should provide TBAs with knowledge in how mobilise mothers to make use of various MCH services. Integration of TBAs activities within the formal heath systems should be emphasized.

Limitations
The design of the study necessitated relying on self-reported information gathered in using qualitative methods. Interviewer bias might have arisen, since qualitative methods have a degree of subjectivity. Research assistants were trained and supervised to ensure quality and truthfulness data. Triangulation of data collection methods and researchers helped to overcome potential bias.

Conclusion
The increasing presence of TBAs in health facilities requires research into their roles/influence. The ethics of unskilled personnel working on labouring women at health facility level need further reflection on the future initiatives. These practices can be mainstreamed into health service delivery and their roles restricted to non-clinical aspects. The positive attitude by women towards TBAs should be utilized to strengthen and support the institutional delivery and referral system. The mothers perceived Global Journal of Reproductive Medicine the TBA to be very kind and understanding and this prevented them from utilizing skilled facility based delivery services.
This experience should be utilized to support future initiatives to strengthen referral systems and promote institutional delivery. Since most of the mothers listened to the TBA, this should provide starting point to help mothers perceive the benefit of delivering at the facility. TBAs would be advocates for positive change and would influence mothers and caretakers decision making autonomy regarding where to deliver from. The results of this study show a weakness in the health system as a major factor promoting home based delivery. Future intervention should focus on improving the health system. The human resource gaps should be addressed by providing additional work force and training of TBAs to perform non clinical functions. This study however suggests that, the training of TBAs would provide a sustained impact on maternal and child health outcome. Health workers should also be motivated to stay in the rural areas. This could be done through provision of housing and training incentives.

Authors' contribution
GN developed and lead the study, conducted the field work, transcription, analysis and writing of the manuscript. PE, EA GM contributed data collection, analysis, EM contributed to writing of the manuscript to the development of qualitative research components. The final manuscript was read and approved by all the authors.