Sampling of health facilities.

In each district, we, together with the district health officer, purposefully selected one or two facilities to represent the three levels of health centres. Level II health centres provide antenatal care services, level III facilities provide antenatal and normal delivery services, while level IV facilities provide all maternity services including surgery [8]. We were primarily interested in the quality of care at health centres rather than hospitals because health centres, especially in rural Uganda, are often the nearest facility for women seeking skilled care during pregnancy and childbirth. We also sampled public and private-not-for-profit (PNFP) facilities that provided maternity services. Table 2 summarises other characteristics of the facilities included in this study.

Download:

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

Table 2. Characteristics of participating health centres.

https://doi.org/10.1371/journal.pone.0213511.t002

Individual interviews with managers and health workers.

One day prior to the interviews, the research team visited each facility to introduce ourselves, request permission to conduct the study, identify participants and schedule interviews. We purposefully selected two or three health workers that worked in the maternity unit, and were available to participate in the study on the day of the visit and their manager. By selecting different cadres of health workers, we aimed to achieve variation in the data. All the twenty-eight health workers that were invited agreed to participate in the study. We conducted in-depth interviews using semi-structured questionnaires that had been pre-tested in Wakiso, a district that did not participate in this study (S1 File: Data collection tools). Interviews were carried out individually so as not to disrupt work, but also to ensure that the health workers were able to speak freely. Questions focused on health workers perceptions of good quality antenatal, intrapartum and postnatal care; factors that influence the provision of this care; and their suggestions to improve the quality of care. Interviews were conducted by one of the authors (SMB, HN), and/or a research assistant, who also took notes and audio recorded the interview. In a few instances, only one of the researchers or a research assistant conducted the interview. Interviews were conducted in English, in a quiet area at the health facility, and lasted 1–1.5 hours. When new ideas were introduced at an interview, we probed for additional information at subsequent interviews, until no new ideas were generated.

Focus group discussions with mothers.

Each of the participating health centres collaborated with a village health team. A health worker contacted one member of the village health team, and requested their help to identify 6–10 mothers that had delivered a baby six months to one year before our visit. We were unable to ascertain the number of mothers that were invited but declined to participate in the study. We chose to conduct focus group discussions because this was practically easier than meeting each mother at home. In order to encourage unrestricted conversations, we held discussions with only mothers, and meetings were held outside a church or in a community member’s compound.

In each district, we held two focus group discussions (FGD), one for mothers that delivered in a health facility, and another for mothers who had delivered elsewhere, for instance at home with the help of a traditional birth attendant or relative. We included mothers who had delivered elsewhere in order to understand how previous experiences with the health system might have influenced the decision not to have a facility birth. FGDs were conducted in the local language, moderated by one interviewer (SMB or a research assistant) and a research assistant took notes and audio recorded the discussions. Discussions were guided by a selection of open-ended questions that were similar to those used in the interview with health workers. Questions focused on mothers’ experiences of antenatal, intrapartum and postpartum care during their last pregnancy, their perceptions of the available quality of maternity care and how it could be improved. Each FGD lasted approximately 90 minutes.

The audio records of the interviews and FGDs were transcribed in English by the research assistants. We did not return transcripts from the interviews or the FGDs to participants for correction.

Clinical care process.

Health workers referred to a number of technical processes and clinical services when discussing what they saw as good quality antenatal, intrapartum and postnatal care. For antenatal care, this included health education to help women prepare for childbirth and encourage facility attendance; illness prevention, for instance intermittent presumptive treatment against malaria; screening for infections; and examinations to determine the condition of the mother and baby and identify high risk pregnancies for referral. For intrapartum care, health workers also screened and assessed mothers to determine their capability of a normal delivery, identify the stage of labour, and detect complications in need of referral. Health workers discussed using the partograph to closely monitor the condition of the mother and baby, and viewed it as a useful tool to alert them to intervene. They also described the importance of conducting safe deliveries in clean and sterile environments and the use of techniques such as controlled cord traction for delivery of the placenta. For postnatal care, health workers emphasised immediate baby care, including skin-to-skin contact, keeping baby warm, initiating breastfeeding within the first hour, monitoring baby’s breathing and ensuring that the cord is securely tied; and the close monitoring and assessment of mothers, for instance for bleeding or hypoglycemia. In addition, health workers described health education to promote good cord care, breastfeeding and vaccination (Additional information in S2 File).

When asked to describe good quality maternity care, mothers broadly described the importance of health workers with the skills to examine, advise and treat women.

Standards of care.

Some health workers also described formalised standards of care when discussing good quality care, including the policy recommendation that women attend at least four antenatal care visits, the postnatal follow-up, and the vaccination schedule. While a few health workers reported using Uganda’s general clinical guidelines [30] which also include maternal health, others reported that they did not have guidelines and relied on knowledge acquired during training and through experience:

‘For sure we don’t refer to the guidelines. We use the knowledge we got from school and experience we are having. We don’t have written guidelines on obstetric care.‘ (Midwife, HCIV)

Timely care.

A few health workers described the importance of providing timely care by minimizing mothers’ waiting time or acting promptly to intervene or refer mothers to higher levels of care:

‘[Good] quality care should start when a mother comes, there shouldn’t be a lot of waiting time. When a mother comes, she should be checked immediately, but in most cases this is compromised due to the shortage of human resources.’ (Midwife, HCIV)

Care that allows women’s choice.

One health worker described the importance of allowing women to choose their labouring position:

‘The mother is positioned, made to lie on back or any other comfortable position she [wants]’. (Nurse Midwife, HCIV)

Availability of medicines and supplies.

When asked to describe the challenges they face when providing maternity care, health workers at all health centres reported that medicines and supplies, including mama kits, were sometimes out of stock. Health workers suggested the supplies received were not aligned to numbers served, leading to stock-outs. In Uganda, the quantity of supplies for health centres at level II and III is determined centrally, also referred to as the ‘push system’, and managers cannot request more according to their need. This is a political decision that is beyond the control of health workers or health system managers.

Another reason for the lack of medicine and supplies was that deliveries take place at Level II facilities even though these facilities are not designated to conduct normal deliveries, but are expected to refer mothers to health centre level III, including stabilising mothers with complications.

‘One of the biggest challenges is shortage of supplies because for us we are on the push system as Health Centre II ….we don’t order what we require. During emergencies, essential drugs like magnesium sulphate are inadequate all the time. We run for months without such supplies. [Our request] for these supplies, like magnesium, disinfectant like jik, are not honored [by National Medical Stores].’ (Clinical officer, HCII)

Faced with the lack of supplies, informal solutions were developed. For instance, health facility managers used primary health care funds to buy supplies, or health workers borrowed supplies from other facilities or asked mothers to purchase their own.

Mothers described the availability of medicines and supplies among the aspects of good quality care that attracted them to health facilities. Some mothers expected to find supplies at health facilities, and did not trust health workers’ intentions when asked to bring their own supplies or for money to purchase them. Mothers’ limited financial ability to purchase these supplies sometimes led to non-facility deliveries and could worsen inequity in access to care.

‘These health workers, why do they ask us for gloves and cotton? It is a government facility where there are midwives, drugs and it is a free health facility. We are poor, we cannot go to private clinics.’ (FGD Facility).

‘The reason I refused to deliver from a health centre is [because] I did not have the supplies required [by the] health facility…… Yes, I had planned to deliver at home because I could not afford requirements of a health facility…. There are many women who cannot afford to buy gloves, polythene, and cotton wool.’(FGD Non-facility)

The availability of skilled and competent health workers.

Both health workers and mothers described the need for sufficient numbers of skilled health workers as part of good quality care. Several health workers described the challenges associated with human resource shortages, for example when doctors or theatre nurses were not available at level IV facilities to provide emergency obstetric care, or when health workers were not able to closely monitor mothers in labour using the partograph:

‘There is no problem in using the partograph but we don’t have time to [complete] it every 30 minutes … So you find we record after one hour or after one and a half hours. We don’t concentrate only on the mother who is in the labour ward’ (Midwife, HCIV).

When asked how they manage to provide 24-hour care given the staff shortages, health workers reported how they cooperated to cover all shifts. However this resulted in ten to 12-hour, or weeklong shifts; the use of unqualified nursing assistants to provide care; or unstaffed health facilities. The long shifts were sometimes informal arrangements by health workers to allow them extended time off with their families:

‘Being a hard to reach area … we are three midwives and all of us [have families in town], so it is difficult to stay here for a long time. So we have agreed to work in shifts. We work weekly, covering both day and night. But this is confidential, we have agreed with only the “in-charge”, and the district health officer is not aware. Midwives have accommodation/ houses with in the health facility, the conditions are very difficult because we don’t have water.’ (Midwife, HC IV)

Some mothers described how health workers were commonly absent on public holidays, and weekends and this sometimes led to non-facility deliveries:

‘When labour pains started I went to the health centre but I did not find any nurse because it was a public holiday. …I went back home as I did not have transport means to go to hospital. I went to a village TBA’s home and she delivered me.’ (FGD Non-facility)

Apart from the shortage of health workers, some health workers indicated how a lack of skills limits their ability to provide good quality maternity care. For instance, some clinical officers, midwives that trained a long time ago, and nursing assistants reported how they lacked skills to use the partograph. Others reported a lack of skills and a need for additional training in managing complicated pregnancies and emergencies such as eclampsia, manual evacuation for post-abortion care, and in quality improvement. Health workers at lower level facilities are expected to recognise symptoms and provide first line management for complicated pregnancies, before referring mothers to higher levels of care. Some mothers reported instances where they believed they were wrongly diagnosed, or had heard about bad experiences they attributed to a lack of skills among health workers:

‘No, she did not [examine me well], and she concluded my baby had no head and she called the hospital to come and pick me because of the assumed complication. ….I think she did not know much because I produced my baby with a head.’ (FGD-Non facility)

Motivated health workers.

When we asked health workers to describe conditions they considered important for retaining their job, many referred to sufficient, regular and timely payment of salaries. They considered their present salary insufficient for the work done, for qualifications acquired after employment, or for length of service. Several health workers also discussed how accommodation at the health facility was important, but insufficient for the number of health workers; or in poor condition, with leaking roofs, lacking electricity or water. Other important conditions mentioned were opportunities for career progression, for example through study leave; workshops; professional support through regular supervision, rewards and timely appraisal; as well as a good work environment with sufficient drugs, supplies and human resources.

Communication and emotional support.

Both health workers and mothers described the importance of good relationships between them as an important part of good quality care. Health workers used terms such as ‘greet her’ and ‘handle her in a good manner’, to describe what they do when interacting with mothers. They described allowing mothers to drink and to have a companion during labour.

Health workers described how they ‘inform mothers’ or tell mothers what to do as part of good quality care. They tended to emphasise a one-way exchange of information and seldom referred to two-way discussions between health providers and mothers:

‘She should (be) informed also about labour progress because when you don’t talk to a mother she thinks maybe the midwife is proud, but when she is informed she sees that the midwife cares about her. Also the relatives need to be informed.’ (Nurse Midwife, HCIV).

Some mothers reported positive experiences and felt cared for by health workers who told them what to do, remained close to the mother during labour, or showed acts of kindness, for instance by making the mother’s bed after delivery. However, some mothers described how they received insufficient information, for instance on reasons for referral. Others felt ignored by health workers who were unavailable and needed to be called repeatedly before they responded to the mothers’ needs. Several mothers described health workers’ communication style as harsh or rude and leaving them feeling uncared for, and this sometimes influenced their willingness to return to the clinic or have a facility delivery:

‘Some time back I delivered from a government facility, but the way they treat us is not good. When labour pains have started, they shout at you and tell you to shut up. Because of that mistreatment, I decided to deliver from home the following delivery. Not all of the health workers mistreat us but some do.’ (FGD Non-facility)

Some women felt discriminated by health workers because they were considered too young or old for childbearing, unmarried or had few resources:

‘It is just by chance that you find the security guard at the health centre at night, sometimes he is not there. And when he tries to wake up the midwife, she will first ask him what kind of mother has come. If they tell her the mother is not financially okay, and she is alone, the nurse refuses to come out. She knows that mothers who come alone cannot afford to give her money. …..if she knows there is a man, he may give her money. They [look down upon] unmarried women.’ (FGD Non- facility)

A few women also described physical abuse by health workers who slapped them:

‘If they talk and you talk back, they beat you…. They slap and abuse you….They can even spit on you….’ (FGD Non- facility)

When probed about mothers’ reports of rude communication, some health workers attributed this behaviour to an individual health worker’s personality, large workloads or norms about accepted ways of communicating in particular communities. Some mothers attributed poor communication to health workers’ workload and lack of training in providing emotional support to mothers.

‘I think all health providers should be forced to attend a course on handling pregnant mothers, because the moment for delivery is almost unbearable, and at that time you need to be taken care of, but you reach there and nobody cares about you. If these health workers cared for us we wouldn’t feel much pain during delivery.’ (FGD non-Facility)

Teamwork, supervision and mentoring.

Some health workers described teamwork and collaboration as an important aspect of good quality care. They appreciated receiving professional support from other health workers at the facility and from higher levels of care, especially when emergencies arose. In addition, some health workers viewed supervision and mentoring as important for ensuring good quality care. While some health workers reported lack of supervision, others reported that they received sufficient supervision and support from district and sub-district managers.

‘To have good quality care, external and internal supervision should be conducted regularly. Mentors, who are senior staff, should help [health workers] adhere to standards’ (Nurse, HCIII).

Collaboration with village health workers.

Health workers reported collaborating with village health workers who provide health information, refer pregnant mothers to health facilities for skilled care, and follow up mothers in the community. However mothers did not consider the village health workers as a channel through which their grievances with the health facility could be addressed, and felt ignored or unaware of any actions taken to address their complaints.

Funding.

The Ministry of Finance provides limited funds to the Ministry of Health to implement primary health care activities such as community outreach for vaccination. However, primary health care funds were reported to be insufficient for all of the needs that arise. Health workers therefore felt obliged to develop informal solutions, including the use of their own personal funds, and called for increased funding to improve the quality of maternity care:

‘PHC [funds] are for immunization, outreaches, school visits, home visits. (However,) PHC is not enough and the money comes after three months… ….so I [sometimes] use my own money’ (Midwife, HCII).

Alignment between planned policy and practice.

Some level II health centres included in this study were conducting normal deliveries even though they are not mandated to, and so lacked the equipment and commodities they needed, and struggled to provide maternity care. Health workers described how staffing levels, infrastructure and supplies available at health facilities were insufficient for the large population served, or the increasing range of activities health workers were expected to provide:

‘Government should change policy of staffing norms [to consider the range of services at health facilities].The system should be reviewed so that they consider space, human resources and equipment [in relation to] the number of patients or deliveries’ (midwife, HCIV).

Geographical location of health facilities.

We observed that the functionality of health facilities at the same level differed, suggesting that structural issues related to the location of the facility may influence quality of care available at the facility. For instance a health centre level IV in one district had a medical doctor, four midwives, an anaesthetist, a functioning theatre where a caesarean section operation had recently been conducted, and an ambulance. This facility was located close to a tarmacked highway leading to the main urban town, and was only a short distance from the main district hospital. The other health centre level IV in the same district did not have a doctor, but had a clinical officer, midwife and nursing officer, had a theatre that was not in use, and mothers were referred to hospital for caesarean sections. This second facility was located a long distance away from the main urban town, in a hilly area, with difficult access. The location of the first health facility along the highway and related infrastructure such as electricity or easy access to the district hospital, may have made it an attractive place to work and retain health workers.