About Bioline  All Journals  Testimonials  Support Bioline  News

African Journal of Reproductive Health
Women's Health and Action Research Centre
ISSN: 1118-4841
Vol. 7, Num. 1, 2003, pp. 92-102

African Journal of Reproductive Health, Vol. 7, No. 1, April, 2003 pp. 92-102

Zambian Women's Experiences of Urban Maternity Care: Results from a Community Survey in Lusaka

MacKeith N1, Chinganya OJM2, Ahmed Y3 and Murray SF4

1Researcher, Institute of Child Health, University College, London. 2Deputy Director, Central Statistical Office, Lusaka. 3Senior Lecturer, Department of Obstetrics and Gynaecology, University Teaching Hospital and School of Medicine, University of Zambia. 4Senior Research Fellow, King's College, London
Correspondence: Dr Susan F. Murray, Women's Health Academic Unit, 10th Floor, North Wing, St Thomas' Hospital, London SE1 7EF, England. E-mail:

Code Number: rh03012


Urban African maternity care systems face problems, as rapid population growth puts them under increasing pressure. In 1983 a decentralised system with midwife-run maternity units at health centres was initiated in Lusaka. A community-based survey of 1210 women conducted in 1999 examined access, coverage and quality of care in these maternity services. Results were generally positive: 99% of respondents received some antenatal check-ups and three quarters had five or more. Institutional delivery rate was 89.5%. Home birth was associated with belonging to a "very poor" household. Sixty three per cent of births were in the decentralised units. Eighty nine per cent reported care as "good" or ``very good", but 21% remembered someone who had treated them badly during labour, principally by shouting or scolding. One fifth of women reported having been left alone for "too long" in labour. Less than half of the women said they would like a lay labour companion and three quarters would prefer a companion at the delivery. (Afr J Reprod Health 2003; 7[1]: 92-102)


Les expériences des femmes zambiennes à l'égard des soins de maternité urbains: résultats d'une enquête dans une communauté à Lusaka. Les systèmes de soins de maternité urbains ont beaucoup de problèmes car une croissance démographique rapide fait de plus en plus pression sur eux. En 1983, on a introduit à Lusaka un système décentralisé avec des centres de maternité dirigés par les sages-femmes dans des centres médicaux.  Une enquête de 1210 femmes basée sur la communauté et menée en 1999 a examiné l'accès, la couverture et la qualité de soin de ces services de maternité. En général, les résultats ont été positifs: 99% des interrogés ont subi quelques examens médicaux prénatals alors que 75% en ont subi cinq on plus.  Le taux d'accouchement était 89,5%. L'accouchement à domicile était lié à l'appartenance à une famille "très pauvre". Soixante-trois pourcent des naissances ont eu lieu dans les centres décentralisées.  89% ont déclaré que les soins étaient "bien" ou "très bien", mais 21% se souvenaient que quelqu'un les avait mal-traitées pendant le travail, surtout en criant ou en engeulant. Un cinquième des femmes ont déclaré qu'elles ont été délaissées pour "trop longtemps" pendant le travail. Moins de la moitié des femmes ont dit qu'elles préféreraient un compagnon à l'accouchement. (Rev Afr Santé Reprod 2003; 7[1]: 92-102)

KEY WORDS: Zambia, maternity services, labour companions, home birth


Most publications on safe motherhood have focused on the enormous challenges faced in providing essential obstetric care to women in rural communities. Less research has focused on the specific problems faced by urban maternity care systems, or on women's views of the services currently provided by them, yet they do impact on significant sectors of the population in many developing countries. Rural to urban migration, combined with natural population increase, is leading to high rates of urban growth in Africa. More than one in three Africans now live in an urban setting, compared to one in seven in 1950, and it has been estimated that by 2020 more than half of the populations of sub-Saharan Africa will live in cities.1 Many of the assumed advantages of urban living such as better access to education and health care services are under threat from this rapid urbanisation.

In large conurbations with rapidly growing populations, overflowing outpatient clinics, congested and chaotic labour wards, and two-to-a-bed postnatal wards are all too familiar sights. In the early 1980s this was the picture of maternity care in Lusaka.  A decentralised maternity care model of `satellite' maternity units run by nurse-midwives at local health centres with a central referral hospital was then initiated. This paper reports findings from a community-based survey conducted in 1999, which explored issues of access, coverage and the quality of care from the perspective of women users of Lusaka's public maternity care services.


Zambia is one of the most urbanised countries in sub-Saharan Africa and its capital city, Lusaka, is home to one fifth of its population.2 It is also an extremely poor country. An estimated 84.6% of the population live below the international poverty line of one US dollar per day3 and there are serious resource constraints. 

The city of Lusaka, in south central Zambia, covers an area of about 360 square kilometres. It is Zambia's chief administrative centre.  Industries include food processing, motor vehicle assembly and the manufacture of clothing and electronic equipment. Its current population is about two million people, over half of whom reside in peripheral `compounds', largely unplanned previously informal or squatter settlements. Planning for Lusaka since the 1970s has largely revolved around programmes to upgrade the `compounds' in terms of provision of improved housing, roads, social services, water and sanitation infrastructure.

The maternal mortality ratio in Lusaka is currently estimated to be between 300 and 500 deaths per 100,000 live births and as such is typical of many African urban centres. High HIV seroprevalence rates in this population are an important underlying factor, the seropositive rate in antenatal attendees at sentinel sites in Lusaka in 1998 was 27.4%.4

At the beginning of the 1980s the city's hospital service was under severe pressure. Of the estimated 34,000 births a year at that time, 70% were taking place at the university teaching hospital although its maternity facilities were only designed to cater for 8,000 deliveries per year. Nearly a quarter of births were taking place at home, and only 6% in three small peripheral units.5 In 1983 the Lusaka Urban Maternity Care Project was initiated with capital funds provided by the Government of Ireland and running costs met by the Ministry of Health, Zambia. The project continued until 1996. It aimed to reduce overcrowding at the teaching hospital by increasing the number of deliveries at urban health centres. Women in the townships were to be provided with efficient, safe and convenient services close to their homes so that only pregnant women with complications would need to go to the hospital.6 Key elements of the project included upgrading and extension of existing health centres to provide delivery services, introduction of the partograph to monitor progress in labour, provision of emergency transport and radio communications, and staffing and running of the satellite maternity units at the health centres by nurse-midwives.

As new centres came into operation, the number of deliveries at the teaching hospital declined. By 1988 the number of deliveries at the satellite clinics had risen from 2,000 to over 15,000, with an in-labour transfer to hospital rate of 10%.5 By 1998, there were three times as many births taking place in the city's health centres (32,000) as at the teaching hospital (10,500).6

At the present time, in eight health care zones of the city administered by the Lusaka District Health Management Board (LDHMB), midwives provide routine antenatal care locally at twenty four health centres (locally called `clinics') and round the clock labour and delivery care at ten of these centres. These health centre maternity units provide care for between 100 and 400 deliveries per month each.

The maternity units are staffed and led by nurse-midwives. There are usually two nurse-midwives per shift in the MCH clinic and two to three in the labour ward. With a three-year nurse training followed by a post basic midwifery qualification, these midwives have considerable autonomy in their clinical practice, providing care in uncomplicated cases and making referral decisions when and if they consider it necessary. Supervision is provided by the senior midwife of the Lusaka District Health Management Team through regular visits to clinics and audit of unit activities, birth outcomes, referrals and adverse outcomes (such as fresh and macerated stillbirth). 

The university teaching hospital, which is the designated national referral hospital for complicated maternity cases, functions as the central district referral hospital for these satellite maternity units. The obstetrics and gynaecology department at the teaching hospital comprises four teams of medical staff (seven consultants, residents from a four-year M. Med. programme and interns) as well as nurse-midwifery staff. There is 24-hour theatre and anaesthetic cover.

User fees were introduced in health facilities in Zambia in August 1993. Women either pay their fees in cash or they contribute to a card `scheme' that covers their pregnancy and delivery care in the public sector. Because the teaching hospital provides care primarily to complex obstetric cases and emergencies, user fees are normally charged to women with uncomplicated labours who choose to bypass the clinics and opt for hospital delivery. This arrangement aims to discourage women from overusing the hospital. Women identified by the health centre midwives as requiring hospital level care are, however, referred from the clinic for hospital care at no extra charge. Such are the shortages of supplies that women may be asked to bring their own materials such as gloves, cord clamps, syringes or equipment for intravenous infusion when there is a deficit. The referral system includes radio contact between the clinics, the LDHMB offices, and a 24-hour centralised ambulance service. There are plans for the hospital also to be included in the radio communications loop but this had not occurred at the time of the survey.

The reduction in the proportion of institutional births taking place at UTH attests to the project's success in convincing many women to use the local clinic services, however, women's views of these services are yet to be documented in any detail.  Furthermore, global figures still tell us little about how well the services reach poorer or more marginalised sectors of the population. This paper reports the results of a community survey conducted in February and March 1999. Its aims were to produce estimates of the coverage and use of these urban public maternity services in Lusaka, to generate quantifiable information on user experiences of the services now on offer, and to solicit their views on topical issues such as companions in labour.


Questionnaire Design

Ethical approval was obtained from the research ethics committee of the University of Zambia. Following a review of relevant literature on user views of maternity services, focus groups were held with local people in order to identify issues that were important at the community level and to ascertain the usual ways of talking about pregnancy and birth. Four focus groups were conducted in local languages by an experienced local interviewer. Groups were composed of women with one child, women with more than one child and male partners of women with children. The tapes were transcribed and translated into English by the interviewer. The survey questionnaire was then drafted and piloted in two areas. Minor modifications were made to clarify the questions.  The questionnaire included questions on basic demography, outcomes of recent pregnancies, self-reported morbidities, knowledge and use of maternity services in Lusaka, reasons for non-use of services, and views on `human environment' aspects of quality of care within Lusaka health facilities. Information on unwanted pregnancy and induced abortion was also collected and will be reported elsewhere.

The Interviewers

The survey was carried out by a team from the Central Statistical Office in Lusaka in liaison with the Women Friendly Services Project (WFSP). WFSP is a collaboration between the Department of Obstetrics and Gynaecology at UTH, the Lusaka District Health Management Board, and members of the Safer Motherhood and Newborn Care group at the Institute of Child Health (University College, London).  Thirteen female interviewers unconnected with the health services but with previous experience of survey work were selected. They underwent a two-day intensive training on the contents of the questionnaire and the sampling methods. The multiplicity of local languages presented particular challenges and the 62 questions in the questionnaire were kept as simple as possible to aid translation. The questionnaire was originally drafted in English, the language used in the formal education system. The schedule was then translated into one of the most widely used local languages, Nyanja, and re-translated to check its clarity. Responses from the face to face interviews were recorded in English/Nyanja.

Subjects and Sampling

The largest clinics with maternity beds were identified within each of the eight health zones. For each clinic townships that constitute the catchment area of a clinic were identified. In each zone a list of townships/compounds were made. A compound with a high density population and a compound with a low density population were selected.

Within each of the eight selected pairs of townships or compounds, four census supervisory areas were selected by linear systematic sampling. In each of the thirty two selected CSAs, a standard enumeration area (SEA) was selected again by linear systematic sampling. All households in each of the thirty two selected SEAs were then enumerated. The data collectors went from house to house listing all households and all women who had been pregnant in the previous two calendar years. Linear systematic sampling from the lists of eligible women was used in each area until a quota of 38 eligible women had been interviewed in each SEA. Although SEAs are supposed to be of a standard population size, they in fact differed widely. Where the SEA had more than one hundred eligible women, every third woman was interviewed. If there were fewer than thirty eight eligible women in the enumerated households, the next SEA was also included. In one case the quota of 38 was not achieved even when this was done, and the total number of women surveyed was therefore 1,210 instead of the 1,216 projected.

Data Quality Control Entry and Analysis

Two supervisors checked all returned questionnaires for quality and consistency and made spot check visits to the field.  Data entry was carried out at the CSO using double entry. Data was analysed using EPI INFO software. Simple chi-square tests were performed in order to ascertain whether there were any significant associations between non-use of services, or reports of poor treatment, and membership of specific demographic, tribal or socioeconomic groupings.

It is customary for some women to go to their natal home for birth particularly in the first pregnancy. Others were away from the city at the time of delivery for other reasons. In total, 14% of the interviewees had gone out of Lusaka at the time of the birth of their most recent baby. Those sections of the questionnaire designed to obtain information on maternity services in the city are reported only for the 946 women who had had their last delivery in Lusaka.


Socio-Demographic Characteristics of Respondents

Majority (85%) of the 1,210 respondents were in the 20-39 age group, with 11% aged 15-19 years and three respondents younger than 15 years old. Seventy nine per cent lived with their male partners, 12% lived in households headed by females. In keeping with the high literacy rates in urban Zambia, all but 5% had at least primary education and nearly half (49%) had also attended secondary school. Nyanja (37%) and Bemba (30%) were the most strongly represented tribal groups, but the sample included Lozi, Tonga, Lunda, Kaonde, Luvale and others. A self-assessment of household economic status commonly used by the Central Statistics Office in Zambia was used for broad classification of the survey population. Respondents were asked to describe the economic status of their household according to three categories. Twenty six per cent described themselves as very poor, 61% as moderately poor and only 13% described themselves as not poor. These self ratings are likely to be on the conservative side. When the responses to this question by heads of households were compared against money metric poverty measures based on household income or expenditure in Zambia, the respondents perceived themselves to be poorer than the objective poverty status would indicate.7

Pregnancy histories ranged from one pregnancy (3% of respondents) to twelve (0.2%). Just below one third of the sample (32%) had had three to four pregnancies, and a quarter had had between seven and ten pregnancies by the time of the survey. One thousand two hundred and ten women had 1,235 pregnancies in the two years prior to the survey. Five per cent had two or more births during the same period and 11% had experienced one or more miscarriages.

Birth Outcomes

Stillbirth or early neonatal death had occurred in 2% of the women's most recent deliveries (n = 1097) in the previous two years. A further 7% reported that their babies were sickly in the days after their birth. Data on maternal deaths was not obtainable due to the nature of the survey.  Of the 946 women who had had their last delivery in Lusaka, sixty two (6.5%) had a caesarean section.

Knowledge, Accessibility and Use of Maternity Services

Use rates of public health care services in Lusaka are generally high. Only two women out of the 1,210 could not name a maternity clinic in their area. Eighty per cent of the women usually walked to their local clinic and 14% got there by public transport. 

Pregnancy Care

Only five women out of the Lusaka births sub-sample did not have any antenatal check-ups at a health facility. This corroborates the high rates of antenatal care for urban women found in the 1996 Demographic Health Survey (DHS) for Zambia.8 Almost three quarters of this group of respondents (n = 685) had five or more check-ups during the pregnancy.

Postnatal Care

Eighty four per cent of the women reported having postnatal check-up at a health facility sometime within the six weeks after birth. The women went for postnatal check-up "because I wanted to make sure I was back to normal" (364 responses), "because the midwife had told me I should" (363 responses) and "because the baby needed care and I had a check-up at the same time" (332 responses). More than one response was permitted. Half (n = 78) of the women who had no postnatal check-up said it was because they "felt well".

Self-Reported Morbidity and Access to Medical Care

Thirty five per cent of the women said they had suffered serious problems at some time during their last pregnancy or in the month after delivery. Some reported more than one health problem: 9% mentioned problems related to blood pressure, 8% malaria and 4% reported ``massive bleeding'' or ``haemorrhage''. Eighty five per cent of the women who reported having serious problems said they had been able to get medical treatment as soon as they felt they needed it. Two thirds of them went to the local clinic while most of the others went to a hospital. The reasons given for not receiving sufficient medical treatment quickly were diverse. The largest single category was non-availability of drugs (reported by 12 women), while seven women said it was too expensive.

Care after Miscarriage

One hundred and twelve women sought medical aid after a miscarriage, 64 went to a hospital, 25 went to a clinic, 11 visited a traditional birth attendant and 10 consulted with a private medical practitioner.  The care received was reported to be "good" or "very good" by all but seven women. Five of the complaints about poor care were related to treatment at a hospital.

Place of Delivery

Nearly two thirds of the 946 women who had their babies in Lusaka received delivery care at the public sector health centres (n = 597) and a quarter at UTH (n = 232). One hundred and one delivered outside health facilities, giving a home birth rate of 10.5%, while less than 2% delivered in a private facility (n = 16).

The Domiciliary Births

Eighty four per cent of those who did not give birth in a health facility already had at least one child. Female relatives assisted at 59% of these births and 13 women gave birth alone. Traditional midwives seem to play a minor role, assisting in only ten of the births in the sub-sample of 101 who delivered outside the health facilities. Two third of women who delivered at home said the labour advanced too quickly for them to do anything else (n = 68). However, there was a significant association between belonging to a "very poor" household and giving birth at home (X2 = 13.13 at one degree of freedom; p = 0.0003).

Costs of Birth Care


Approximately two thirds (n = 606) of the 946 women who gave birth in Lusaka said they had to pay for their care. Payment was usually in cash (373 women) or through a savings card scheme (221 women). At the time of the survey 2,500 Kwacha was the equivalent of about one US dollar. Out of the 606 women who paid towards the costs of their care, majority (63%) were charged between 1,000 and 5,000 Kwacha, 17% were charged between 5,000 and 10,000 Kwacha, and 7% between 10,000 and 20,000 Kwacha. The costs of health care were mentioned as a deterrent to the use of services by 13% of the women who delivered at home.

Purchase of Delivery Equipment and Items for the Baby

Ninety five per cent of women who gave birth in Lusaka reported that they had to buy things for the birth, the most common items being baby clothes, protective gloves for staff, pads or cotton wool and umbilical cord clamps. When asked if it was difficult to find money for the equipment, one third of the women found it ``not difficult'', one third found it ``quite difficult'' and another one third found it ``very difficult''.

User Perspectives of Aspects of Quality of Maternity Care

The sub-sample of 845 women who had given birth to their most recent babies within a health facility in Lusaka were asked questions about specific aspects of the care they received and were asked to evaluate their personal treatment overall. The problematic aspects of assessing satisfaction with child- birth care have been recognised for some time.9-12 Two of the common inhibitors to criticism of medical care, ``captive patient populations'' and ``captive researchers''2, were avoided by conducting the survey in the community with interviewers who were not health workers. A mixture of fixed response and open-ended questions was then used, covering different aspects of quality of service.

Professional Attendance at Delivery

The 783 interviewees who had vaginal births were asked who was in attendance at their birth. Most of them had appropriate professional attendance (midwives or doctors) but there were incidents at health centres where this did not happen. Three births were attended by cleaners, one by another unqualified person, and 19 women delivered alone with no assistance.

Care and Companionship in Labour

A series of randomised controlled trials3 indicating that labour companions can have a positive effect on labour experiences and outcomes has produced increased interest in the possibilities of permitting lay companions in health facilities. This is not current practice in Zambian public health facilities although a study of the impact of companions has been conducted at the University Teaching Hospital. A series of questions designed to explore women's preferences on this topic were included. When asked about their last labour experience, more than one fifth (n = 171) of the women reported that they were left alone in labour for too long, and this was more likely to happen at a health centre than in the hospital. There is a significant association between receiving labour care at a health centre and being left alone for too long (χ2 = 19.36 at one degree of freedom; p = 0.00001).

Views on the Desirability of "Lay" Labour Companions

All the women were asked if they would like someone other than a staff member to accompany them during their first stage of labour. Forty four per cent said they would indeed like this while 55% said they would not. Mothers were the most commonly preferred labour companions, followed by male partners, sisters and friends as shown in Table 1. The 518 women who did not want a labour companion gave a range of reasons as shown in table 2 (more than one response was permitted).  

Views on a "Lay" Companion at Delivery

The idea of a companion being present at actual birth of a baby was popular among majority of the women (Table 1), although over a quarter said they would not like anyone to be present except the health worker. For those in favour of a delivery companion, mothers were still the most popular choice, with a higher proportion wanting the presence of their male partner at actual birth (22%) than during the labour (9%).  

Views on Episiotomy

Institutions rarely document their episiotomy rates.14 Among the survey population there was a 28% episiotomy rate in women who had their babies vaginally at a health facility. Half of those who had episiotomies done said this had not been a problem for them, they felt that it was normal to have one done or said that there was not enough room for the baby to be born easily without it. Those who found the episiotomy to be a problem reported that it hurt (two thirds) or that it took too long to heal (one third). 

Ratings of Personal Treatment Overall

The 845 respondents who had their deliveries in a health facility were asked to rate overall their pregnancy and birth care on a four point scale of ``very good'' to ``very bad''. Their responses are shown in Figure 1. Almost all the women with higher education (69 out of 71) reported good or very good care experiences. There is a marginally significant association between having no education and reporting care as bad or very bad, as evidenced by the chi-square value (χ2 = 3.86 at one degree of freedom; Fisher's exact one-tailed p = 0.0526). There was no association between belonging to a very poor household and reporting bad care.

Suggestions for Improvements

Seventy four per cent of women who gave birth in a Lusaka facility replied in the affirmative to the following question: "Is there anything you would like to see changed?" and they gave 1,047 suggestions. Concerns were expressed over the need to improve equipment supply (n = 139) and maintenance of better hygiene in toilets and bathrooms at health facilities (n = 154). Over half (550) of the respondents wanted improvements in attitudes of staff members and kinder treatment of women.

Poor Personal Treatment

Twenty one per cent of women remembered someone who had treated them badly. Midwives were the most commonly mentioned group here (n = 150), but this is probably a reflection of their key provider role. Cleaners were mentioned by 21 women and doctors by only five. The commonest complaint was shouting or scolding (56%). Another 16.5% said the person concerned ``did not come'' and suggested that health workers should stay nearby (99 respondents). We found no significant association between an adverse perinatal outcome and accounts of poor treatment. Fourteen women in the study population who delivered in a Lusaka health facility had stillbirth or neonatal death in the first few hours of life. Only one of these women described her care as ``bad''.

Good Personal Treatment

Many women reported good treatment by maternity care providers: 58% of the women remembered someone who had done his/her job particularly well, and a further 29% said everyone did their jobs well.  Midwives were the most commonly mentioned occupational group to receive praise (n = 437), while doctors, whose public sector maternity work is confined to the hospital level, were mentioned by 54 women. The sometimes neglected contribution of ancillary staff to the general "human environment" was highlighted by 44 women who praised the cleaners. Some multiple responses were given (total = 576). Kindness, encouragement and understanding were attributes that the women remembered and valued (making up half of the 821 mentioned). Giving good care (n = 256) was also important. Fifty nine women said "they didn't shout".


Estimates for coverage and use of Lusaka's public sector maternity services produce, on the whole, an encouraging picture. The proportion of births attended by health professionals in the city is high, with an 89.5% institutional delivery rate. Seven out of every eight women had their babies within the public care system. There were high rates of use of services for antenatal as well as postnatal care. Almost all women in the sample received some antenatal care, four fifths had postnatal check-ups within six weeks, and 85% of those women who reported what they felt to be serious morbidities said they had been able to get medical treatment as soon as they felt they needed it.

The figures confirm the success of the decentralisation process in shifting where women have their babies in Lusaka. Nearly two thirds of the women surveyed gave birth at the health centres with professional midwives attending and only one quarter delivered at the hospital. The 10.5% home delivery rate is open to interpretation. It suggests a substantial reduction from the estimated quarter of births occurring at home in the early 1980s5 and may, therefore, be seen as another indicator of the success of the decentralisation process. On the other hand, the association found between belonging to a very poor household and having a home delivery (i.e., without a skilled attendant) suggests an area where special initiatives may still be required. Delivery related costs such as fees, gloves, cord clamps and baby clothes should not be a deterrent to accessing care.

If service use is a measure of acceptability of public health services, then on this criterion the decentralised maternity services in Lusaka performed well. The more detailed questions on labour and delivery experiences did, however, show up areas that need improvement. While 89% of respondents rated the standard of their overall care as good or very good, reports by one fifth of respondents of incidents of bad treatment and the frequency of shouting or scolding by some midwives are still causes for concern. There are indications of a link between such displays of lack of compassion and social status using the marker of educational level. Why such behaviour exists within a ``caring profession'' such as nursing has been the subject of two recent papers based on qualitative studies conducted in neighbouring countries in the Western Cape Province of South Africa5 and in Murehwa district and Harare in Zimbabwe.16 Both of these papers describe accounts of ``reactive'' and ``ritual'' verbal and physical abuse and neglect of patients by nurses, and attributes such poor nurse-patient relationship to the broader infrastructure context: nurses are engaged in an unremitting struggle to claim a status as a middle class profession within environments in which political, professional, historical and personal factors continuously undermine this claim.15

Jewkes et al argue that the deployment of violence has become established as normal in nursing practice because of lack of powerful competing ideologies of patient care and nursing ethics.15 The Lusaka survey, however, does not give a picture of unremittingly poor nurse-patient relationship. Most women (87%) said they remembered maternity staff who did their jobs particularly well, showing kindness, understanding, giving encouragement and providing good care. The challenge for managers, therefore, is to find ways to reinforce the caring ideology that does exist and to support staff in his practice as well as to discipline staff members who are abusive to patients. In Lusaka, as a result of the survey findings, we began to use the Health Workers for Change methodology developed by the WHO to improve the personal communication skill of professional and ancillary staff.17

The issue of some women being left alone for too long in labour is also a fundamental to be addressed particularly given all that is known about the negative influences of fear and anxiety on the progress of labour. 18 The admission of lay companions to labour wards may be one solution; indeed the presence of a companion at birth is one of the five suggested measures in the Bologna scale that is currently being developed by the WHO19 as a possible evaluation tool for appropriate management of normal labour. (The other four are use of partograph, absence of augmentation or emergency caesarean section, use of a non-supine position for birth, and skin to skin contact of mother and baby for at least 30 minutes in the first hour).17 However, responses to the question on companions reveal that this is a more complex issue than is sometimes suggested. Firstly, women's preferences may differ as regards labour and delivery, when they wish to be accompanied and by whom.  Secondly, while some women would clearly like and benefit from having companions, it is by no means a universal wish. The proportions of women who would rather be alone or were concerned that their companions ``might say things afterwards'' in our Lusaka sample indicate that such intervention has to be individually tailored. It also raises some doubts about the applicability of the presence of a companion at birth as a universal indicator of appropriate labour management. There may also be other ways of helping women to ``feel accompanied'', perhaps focusing on staff activities or on the physical layout of the labour ward and midwives' desks.

This paper has reported findings from a survey of user views and experiences of the Lusaka public maternity care system. They cannot, however, tell us much about other aspects of crucial importance such as effectiveness and safety, and we have written elsewhere about the ways in which effectiveness of maternity referral may need to be measured and monitored.20 A post-evaluation of the Lusaka maternity clinics project by an Irish aid team noted that the number of maternal deaths at the Lusaka health centres has been ``very low'' during the duration of the project and in the two years that followed, with a total of five maternal deaths between 1982 and 1998.6 What this survey has thrown up is a number of important quality of care issues, and the Lusaka district health management board and the University Teaching Hospital maternity services have both subsequently been involved in piloting interventions to improve their women-friendliness.21 The findings also indicate that the decentralised maternity care system in Lusaka has many important achievements to its credit.  It has made important advances in coverage and accessibility of care for women in the city as well as contributing to a significant reduction in the pressure on the hospital maternity unit.


This survey was carried out as part of the Women-Friendly Services Project supported by the UK Department for International Development (Project SCF219). However, DFID can accept no responsibility for any information or views expressed.  We would like to express our gratitude to Dr Hilary Standing, consultant to the WFS Project; our focus groups interviewer, late Mr. T. Chabala; the Zambia Central Statistical Office for their key role in sampling and data collection; the two supervisors, Ms Carol Mweemba and Ms Sheila Shimwambwa; the CSO interviewers; mapping and data entry staff; and all the women who gave their time to take part in the survey.


  1. Rossi EE. Meeting the growing demand for quality reproductive health services in urban Africa: partnerships with municipal governments. SEATS Project, John Snow Inc, Arlington, 2000.
  2. Atkinson S, Ngwengwe A, Macwan'gi M, Ngulube TJ, Harpham T and O'Connell A. The referral process and urban health care in sub-Saharan Africa: the case of Lusaka. Zam Soc Sci Med 1999; 49(1): 27-38.
  3. World Resources Institute.  World resources report Table El. 3. Distribution of income and poverty, 2000. data_tables /ei3n_2000.PDF
  4. Medical Research Council of South Africa. Local authority response to HIV/AIDS: an overview of key cities. Urban Health Dev Bull 2000; 3(2). authority.htm
  5. Nasah BTand Tyndall M. Emerging problems of maternity care in urban settings. In: BT Nasah, JKG Mati and JM Kasonde (Eds.). Contemporary Issues in Maternal Health Care in Africa. Luxembourg: Harwood Academic Publishers, 1994, 67-87.
  6. Irish Aid Evaluation and Audit Unit.  Evaluation report, Lusaka urban maternity clinics project, Zambia, 1999.
  7. Central Statistics Office.  Living conditions monitoring survey report, 1996.
  8. Central Statistics Office, Ministry of Health and Macro International Inc. Zambia demographic and health survey 1996.
  9. Shearer BC. How do parents really feel after caesarean birth? Birth 1983; 10: 91-92.
  10. Lumley J. Assessing satisfaction with childbirth. Birth 1985; 12(3): 141-145.
  11. Bennett A. The birth of a first child: do women's reports change over time? Birth 1985; 12: 153-158.
  12. Gilson L, Alilio M and Heggenhougen K. Community satisfaction with primary health care services: an evaluation undertaken in the Morogoro region of Tanzania. Soc Sci Med 1994; 39(6): 767-780.
  13. Hodnett ED. Support from caregivers during childbirth. Coch Rev 2000; 4.
  14. Maduma-Butshe A, Dyall A and Garner P. Routine episiotomy in developing countries. Br Med J 1998; 316: 1179-1180.
  15. Jewkes R, Abrahams N and Mvo Z. Why do nurses abuse patients? Reflections from South African obstetric services. Soc Sci Med 1998; 47(11): 1781-1795.
  16. Bassett M, Bijlmakers L and Sanders D. Professionalism, patient satisfaction and quality of health care experience during Zimbabwe's structural adjustment programme. Soc Sci Med 1997; 4(12): 1845-1852.
  17. WHO. Health Workers for Change: A Manual to Improve Quality of Care. Geneva: WHO, 1995.
  18. Klaus MH, Kennell JH, Robertson SS and Sosa R.  Effects of social support during parturition on maternal and infant morbidity. Br Med J 1986; 293(6547): 585-587.
  19. Chalmers B and Porter R. Assessing effective care in normal labour: the Bologna score. Birth 2001; 28(2): 79-83.
  20. Murray SF, Davies S, Kumwenda Phiri R and Ahmed Y. Tools for monitoring the effectiveness of district maternity referral systems. Health Pol Plann 2001; 16(4): 353-361.
  21. TALC/Quadreto. How to Make Maternal Health Services more Women-friendly. A Practical Guide  Teaching Aids at Low Cost. London, 2001.

Copyright 2003 - Women's Health and Action Research Centre

The following images related to this document are available:

Photo images

[rh03012t2.jpg] [rh03012t1.jpg] [rh03012f1.jpg]
Home Faq Resources Email Bioline
© Bioline International, 1989 - 2014, Site last up-dated on 14-Apr-2014.
Site created and maintained by the Reference Center on Environmental Information, CRIA, Brazil
System hosted by the Internet Data Center of Rede Nacional de Ensino e Pesquisa, RNP, Brazil