Working with Midwives to Improve Maternal Health in Rural Ghana

M. Kay Matthews, M.N., S.C.M.

Honorary Research Professor

Memorial University, St. John’s  NL.

Nursing and Midwifery Consultant

MaterCare International

matthews@mun.ca

Ph. 709-579-6472:  Fax 709-579-6501

Robert. L. Walley, FRCOG; MPH

Honorary Research Professor

Memorial University, St. John’s  NL

Executive Director

MaterCare International

rwalley@mun.ca

Biography

M. Kay Matthews.

168 Patrick St

St. John’s,  NL A1C 5C2

Ms. Matthews is an Honorary Research Professor at the School of Nursing, Memorial University of Newfoundland and Nursing and Midwifery Consultant for MaterCare International, an NGO involved in Safe Motherhood Projects and the treatment and rehabilitation of women with obstetric fistulae. 

ACKNOWLEDGEMENTS

This project was partly funded by the Canadian International Development Agency.  The authors thank all who participated in the project, especially the project midwives Kate Andah and Elizabeth Ankore, project assistants Patrick Frimpong and Efua Turkson, Doctors John Wilson, Harry Tagbor and Stephen Poku and all the midwives, TBAs and mothers who taught us much. 

Abstract

This paper describes the implementation and evaluation of a safe motherhood project in rural Ghana. This project included a partograph and emergency skills program for rural midwives, training and monitoring traditional birth attendants (TBAs), a blood bank and an emergency obstetric transport service.  The midwives roles in caring for women in rural health centres, training and monitoring traditional birth attendants TBAs) and in emergency obstetric transport are described.  In this rural area of Ghana TBAs are responsible for approximately 65% of women in the district during birth.  It is important to include them in projects to improve maternal health.  The results of process evaluation showed that the rural maternal health system was improved by the project activities through a good communication system, better support for rural midwives, better integration of traditional birth attendants into the health care system and more community involvement.  Outcomes included an increase in the number of referrals of women with risk factors and complications to the district hospital, no cases of prolonged obstructed labour in the group of mothers cared for by rural midwives and trained TBAs, a comprehensive recording and reporting system for women cared for by TBAs, and a well established emergency obstetric transport service. 

Integration of the TBAs into the health care system has had positive effects on TBA practice and morale.  However, some TBAs had problems receiving remuneration for their work due to cultural beliefs and practices.  The author concludes that community education and participation are key components of safe motherhood projects.

Safe motherhood; maternal mortality/morbidity; rural midwives; partograph training; referral of women at risk; traditional birth attendants; emergency obstetric transport.

Working with midwives to improve maternal health in rural Ghana.

            Problems in maternal health in developing countries are well known.  The numbers of women who die in childbirth vary across the world, but the reported statistics suggest that for every 100,000 live births in Africa, as many as 979 women die in childbirth¹.  Death and disability rates are further affected by war, famine and other natural disasters, the responses to which are usually beyond the developing country’s resources.  The major causes of death in childbirth are haemorrhage, pregnancy-induced hypertension and infections.  An even higher number of women suffer injury related to childbirth, such as obstetric fistulae.  The physical consequences of obstetric fistulae for the woman include bowel and bladder incontinence, skin excoriation and infertility.  The social consequences of these problems to the woman and her family are devastating. There are three major factors associated with maternal mortality and morbidity.  These are known as “the three delays”, delay in identifying the problem, delay in transportation and delay in intervention in a timely manner at the referral hospital^2.  Strategies to address the delays is a crucial component of any safe motherhood project.. The World Health Organization (WHO) and other national and international agencies have called on governments to support initiatives to address maternal health problems.  Midwives are central to the care of mothers and babies during childbirth.  Therefore, involvement of midwives in any strategies to address maternal health problems is critical to the improvement in maternal and infant health outcomes.

            The purpose of this paper is to describe a collaborative Canada-Ghana safe motherhood project between MaterCare International, a Canadian non-government organization, and the Diocesan Health Committee, Catholic Diocese of Sunyani, Brong Ahafo Region, Ghana.  The project implementation and evaluation for the period 1999 to 2003 will be described and some of the challenges and issues which arise in international projects will be discussed. 

            In 1998, a safe motherhood project was established in a rural district in the Brong Ahafo region in central-west Ghana.  The project covered six rural maternity centres of 11 in the district.  We had hoped to include all health centres, but five of the centres were not staffed by midwives and did not offer maternity care.  The project was carried out in two phases.  Phase One was from mid-1998 to July 2001 and Phase Two August 2001 to July 2003.   The project team in Ghana consisted of a midwife who was also a public health nurse, a project assistant who collected the evaluation data, the local medical director and, on intermittent visits, the Canadian nursing and midwifery consultant and the Ghana and Canada medical directors who were consultant obstetricians. 

The overall goals of the project:

1.        to improve maternal health by lowering maternal mortality and morbidity rates in the district of Nkoranza, Brong Ahafo Region, Ghana.

2.        to develop an integrated maternal health service linking the rural areas with the district hospital. 

The major objectives:

1.        To establish use of the partograph to monitor and record the progress of labour for all women in labour in six rural health centres.

2.        To identify and train all traditional birth attendants in the project area to conduct a clean, safe birth and recognize and refer women with risk conditions and/or complications in childbirth to the local health centre or the district hospital.

3.        To establish an emergency obstetric transport service linking the rural areas with the referral hospital.

4.        To establish a maternity blood bank at the district hospital.

Project Description

The project consisted of four components:  partograph and emergency skills training of rural midwives, training of traditional birth attendants, (TBAs), an emergency obstetric transport service (EOTS) and a blood bank.   The trained midwives were central to the operation and success of the first three components of the project.  

The context:

            Nkoranza district is one of 13 administrative districts in the Brong Ahafo Region.  It is situated in the mid-west part of Ghana and covers an area of 1200 square kilometres.   The people of the district are mainly farmers producing maize, yam, cassava, beans, groundnut and rice.  There are about 191 communities in the district making up a population of about 139, 690. There are no telephones in the villages and transportation from the villages to the town is a problem due to bad roads, especially in the rainy season when they can be impassable.    As well, few villagers have cars. 

As in most developing countries, maternal death is not uncommon and perinatal death even m ore common.  In Ghana generally the maternal mortality ratios in 1993 were estimated to be 214: 100,000 live births, but there were wide regional variations.  A recent survey in Brong Ahafo region showed rates of 212-328.3   In the year prior to the start of the project (1997) there were four recorded maternal deaths in 790 in-hospital births at this hospital, a ratio of 506:100,000 live births.  As well, there were anecdotal reports from the midwives and doctors of women who died having arrived in critical condition at the hospital often too late for help.   The maternal and perinatal mortality rates in the villages for out-of-hospital births at this time were not known as there were no systematic comprehensive data collected in the 191, mostly rural, communities which comprise the district.  Many births took place in the village with some trained, but most untrained traditional birth attendants.  

Midwives in the Health Clinic and Hospital Setting

In Ghana, most midwives are educated as nurses first and then attend midwifery school.  Ghana’s colonial past is linked to Britain and the midwifery education and practice models are similar to the British system in that most programs are hospital diploma based and midwives practice takes place either in the hospital setting, or in the community as government employees.  Community midwifery practice is based in rural health centres.  Each health centre is staffed by a nurse, a midwife and, sometimes, a physician assistant.  Some midwives and nurses have specialized community health qualifications.  The government provides motorbikes to enable the public health nurse or midwife to visit her large district, but frequently there is a shortage of gasoline or vehicle breakdown which makes outreach difficult.   As well, only the bravest midwife will venture forth by motorbike alone to visit the outlying villages in her district because of the bad roads, isolation and distance.  In talking to midwives I have heard that salaries and morale are low.

The hospital midwife works shifts in the small combined pre- intra- and postpartum unit.  The hospital, run by the local diocesan health committee, is better equipped with basic supplies compared with some of the government hospitals, and I have always been impressed with the midwives’ resourcefulness and competence in difficult situations.   

Project Activities

The first step in the project was to provide a course to upgrade the partograph and emergency obstetric skills of the rural midwives.  The World Health Organization (WHO) recommends the use of a partograph (a graph for monitoring and assessment of cervical dilatation and foetal head descent over time) which alerts the midwife to the need for maternal transfer to hospital for prolonged labour.  For some of the midwives it was a refresher course, for some it was new material.  The partograph course was taught by a midwifery lecturer from the regional school of midwifery and emergency obstetric skills by the Ghanaian consultant obstetrician to the project.  The health centres were supplied with partographs and a system of data collection for evaluation of partograph use was established.   Data included the number of births at each maternity centre, the number of partographs used, the number of referrals to the hospital, particularly those referred for prolonged labour, the number of mothers who were in  partograph categories 2 and 3  (those who had reached the alert or action lines) who were not referred, and maternal and perinatal outcomes.  

Training of Traditional Birth Attendants (TBAs)

The second step was to identify and train the TBAs.  The local project team visited the  villages in the project area and asked the TBAs to come forward.  If there was no TBA the village committees were asked to recommend someone from each of their communities  for training.  For the first phase of TBA training 131 TBAs were identified and trained.  A few TBAs  were offered, but declined training.  In the second phase, another 35 were trained.  Each training session took place over a two-week period and was conducted by the project team and a TBA trainer from the Ministry of Health.  Topics included clean, hygienic delivery, care of the newborn and recognition of childbirth risk factors and the signs and symptoms of complications.  For the risk factors and complications, we used a pictograph (a set of drawings of the major complications of pregnancy, labour and postpartum) that we had used for a similar project in Nigeria.  The pictograph drawings included pregnancy induced hypertension, very short stature, the schoolgirl, antepartum hemorrhage, previous cesarean birth (uterine scar), prolonged labour, breech presentation, twin pregnancy, postpartum hemorrhage, retained placenta and grand-multipara.   If the TBA checked one of these pictures for any of her clients she was instructed to refer the woman to the health centre, or if more appropriate, directly to the hospital.  After the TBAs were trained they were supplied with delivery kits and pictographs.  Initially, TBA practice was monitored by the project team, but this was gradually passed over to the health centre midwives to monitor the TBAs in their own district.  A system for data collection was established throughout the TBA project to evaluate process and outcomes.

There were no reliable statistics of maternal deaths in the villages prior to the project. This made a pre and post test evaluation of project outcomes not feasible.  However, useful descriptive statistics and qualitative data for process and outcome evaluation were obtained.   Data collection for this component included the number of TBAs trained and still practising, the number of mothers who gave birth with the TBAs, the use and accuracy of the pictographs, the number of referrals to the hospital or health centre and maternal and perinatal outcomes. 

The Blood Bank

The laboratory at the district hospital was provided with the essential equipment for safe collection and transfusion of blood for obstetric emergencies.  

Emergency Obstetric Transport Service

In January 2000 the emergency obstetric transport service (EOTS) was established.  A land rover ambulance was equipped with emergency supplies, a midwife from the hospital was assigned to the emergency transport system, an instruction manual developed for all personnel and the community and TBAs informed about the service.  Most important, a radio communication system was set up linking each of the health centres with the maternity unit.  The radio system was solar-powered and generally worked well, although problems occurred due to atmospheric conditions at one health centre during the rainy season.

The data collected on the EOTS included the person who initiated the ambulance call, reasons for referral, any treatments required en route, treatments on arrival at the hospital and maternal and perinatal outcomes.  As well, we collected information on the emergency response time from the attempt by the midwife to call for help to arrival of the woman at the hospital.  Instances when the system failed for some reason were analysed and remedial action taken as necessary.    Data were also collected on the cost of gasoline, vehicle maintenance etc.  The midwife assigned to the service was responsible for keeping the ambulance stocked with emergency supplies. 

PROJECT OUTCOMES

Evaluation was done on an on-going basis by the Canadian nursing and midwifery consultant and, at the end of the fourth year of the project, by independent evaluators.4   The system worked well to link the traditional birth attendants and the rural health centres and midwives to the district hospital.  The radio communication system was considered a great benefit as the health centre midwives could contact the maternity unit for advice as well as summon the ambulance as needed.  The TBAs were more closely linked to the professional health care system through improved relationships with the rural midwives and the hospital.   Although the primary aim of the project was to link the TBAs with the health care system through the health centres, the TBAs were encouraged to refer either to the health centre or the hospital directly depending on the circumstances.  For example, in emergency if the TBA considered transfer to the hospital rather than to the health centre was indicated, or if a woman had a previous caesarean birth the TBA would refer to the hospital antenatal clinic or maternity unit.  The TBAs complained that some of the hospital midwives did not welcome them when they brought their pregnant women to the hospital, but relationships improved over time as the hospital midwives became more oriented to the project.  The trained TBAs were highly motivated and if they had problems in the villages they readily consulted with the health centre midwife and referred when necessary.   Most women attended the antenatal clinics at the health centres even if eventually they gave birth in the village with the TBA.  Over the life of the project the number of referrals from the TBAs either to the hospital or health centre increased.  Some of the women referred by the TBA s were cared for at the health centre.  In other cases, the midwife transferred the woman on to the district hospital. 

The Partograph Project

Throughout the project, the consultant reviewed all partographs for use and accuracy and on-going support and advice given as necessary. An independent evaluation of partograph use found that partographs were used for the majority of births (Table 1) and non-completion was usually due to the woman arriving at the health centre fully dilated.  A Midwife Competency Checklist was used to assess the accuracy of 50 randomly selected partographs.  This found that 65% of partographs were completed correctly.  The majority of items not completed or incompletely recorded related to whether drugs were given or not.  However, other than ergometrine for the mother and Vitamin K for the baby, drugs are rarely used in the health centres.  A review of partographs of mothers referred to the hospital for prolonged labour found that all the mothers in the sample had been referred from the health centre before they reached the “action” line on the partograph.4 The evaluators suggested that this indicated the women were transferred too early, but given the difficulties in transportation (the average ambulance response time from reception of the call to arrival at the maternity unit was 72 minutes), it is difficult to criticize the midwife’s judgement in these cases. 

Although referral of primipara to the hospital for birth was recommended by the local health authorities, some primipara gave birth in the health centres particularly those furthest from the hospital.  Other primipara gave birth in the village.  This was usually due to the woman’s choice, or arrival at the health centre or TBA too late for transfer. There were no cases of prolonged obstructed labour in women cared for by health centre midwives or the trained TBAs.    From the table, 16.5% of mothers had a postpartum haemorrhage and one of these mothers died.  The maternal and perinatal mortality ratios calculated for the group for 1999-2002 were 71: 100,000 and 21:1000 respectively. 

Training and Monitoring TBAs

Referrals of women with risk factors or complications in labour to the health centre or hospital by TBAs increased over the life of the project.  For example, 116 mothers were referred to the health centre from 1999-2001.  The midwives were able to care for 65 (56%) of the women at the health centre and transferred 51 (44%) on to the district hospital.  During the second phase, 381 referrals were made, of which 160 were referred to the health centre and treated there.  Of the mothers transferred to the district hospital, 35% received potentially life-saving interventions.  Most of the mothers who gave birth without complications were primipara, and grand multipara referred because they were considered to have risk factors. 

The role of the health centre midwives was crucial in the follow-up, support and monitoring of TBAs after they had been trained.  In the first phase of the project (1999-2001) the project team conducted the monitoring and support of the trained TBAs.  Monitoring sessions included visits to the villages to meet the TBAs, inspect the delivery kits, review the use of the pictograph and correct and complete them as necessary.  The TBAs were also asked about any other births that may have occurred in the village and whether there were any untoward circumstances or adverse outcomes.  Continuing education was an important focus of the visits.  For example, after a newborn died because the cord ligature had slipped, sessions were held to reinforce the method and importance of cord ligature.

Sustainability after the project had ended was an important consideration and for the second phase of the project (2001-2003) monitoring and support was handed over to the health centre midwives.  This was already a role assigned to the rural health centre midwives by the Ministry of Health, but monitoring was sporadic and incomplete because of transportation problems.  Motorbikes were provided for community outreach, but they were often inoperable due to lack of gasolene, lack of repairs and, as well, only the most intrepid midwives would venture out on the roads after two of them had sustained broken limbs.   Following consultations, the midwives felt that market days were good days for centralized monitoring.  Most women travel to the central village for market day and the midwife met the TBAs at a convenient location for monitoring on selected market days.     Generally, this worked well, but some TBAs would forget to bring their kits and pictographs for inspection and others were absent.  In these cases the midwife followed up in the villages.

The Emergency Obstetric Transport Service:

The average response time for the ambulance was 100 minutes during the first phase of the project.  Delays occurred due to problems contacting an ambulance driver until walkie-talkie radios were given the driver-on-call. During the second phase of the project, the response time was reduced to 73 minutes.  However, in some emergency situations, the midwife sent the woman to the hospital by taxi or private transport to save time. 

MATERNAL AND PERINATAL OUTCOMES

We monitored and documented all cases of maternal death in the villages, health centres and mothers referred from the area covered by the project throughout both phases of the project.  The total numbers of births and deaths which occurred within the hospital, the rural health centres and in the villages with TBAs through each phase of the project and the totals are presented in Tables 2, 3 and 4.  Although we focused on maternal health we also documented the perinatal outcomes. 

Health Centre Outcomes

All cases of maternal death in women within the project area were reviewed. The deaths were investigated through interviews with family and community members, traditional birth attendants and midwives when a woman died out of hospital.  We tried to do this in a non-threatening way.   Over the first phase there was one maternal death among the group who gave birth in the health centres.  The woman, who had a history of postpartum haemorrhage, had declined referral to hospital, arrived late in labour and died of postpartum haemorrhage before she could be transferred to the hospital.   This death occurred before the EOTS was available.

During the second phase of the project, 1970 mothers gave birth at the rural health centres. One maternal death occurred at one of the health centres.  The woman, in her second pregnancy, had been in labour in the village for several days.  In this village the trained TBA was away and the woman’s husband, who had successfully helped his wife give birth to the first baby believed he could handle this birth.  Eventually, the woman was brought to the health centre and died while the ambulance was being summoned and the midwife was trying to start an intravenous transfusion.   The cause of her obstructed labour was determined to be locked twins.   Cause of death was related to uterine rupture.

It was not possible to determine the causes of perinatal death as autopsies were not done.  Antepartum hemorrhage, pregnancy induced hypertension and prematurity were leading factors in several deaths.  However, most deaths were unexplained.  One baby died in the village after the cord ligature had slipped.   Some of the perinatal deaths occurred in mothers who had been transferred to the health centre or hospital from the TBAs.

TBA Outcomes

In the district over the course of the project the TBAs were responsible for 58% of all births in the project area.  In the first three years of the project, four maternal deaths occurred in the villages.  Two were women cared for by project-trained TBAs.  One of these was not directly obstetric-related.  Death was thought to be due to cerebral-spinal malaria following an uncomplicated birth in the village.  Her baby was also very ill and both died following transfer to hospital. The other mother died en route to hospital in a taxi from postpartum haemorrhage also before the emergency obstetric transport service was in place.  Of two other deaths reported in the villages, one was related to AIDS, the other was a woman who had been cared for by an untrained TBA and when labour was prolonged she was brought to the house of the trained TBA where she died soon after arrival, before the TBA was able to summon help.  Autopsies are not done, but conclusions as to cause of death can  be made on a careful analysis of the history and signs and symptoms.  In this case, ruptured uterus was thought to be the cause.  Maternal deaths in the hospital numbered eight during the first phase of the project, but these included women from the area not covered by the project, and those who did not receive trained TBA or professional midwifery care.

There were 1588 births supervised by trained TBAs during the first phase of the project and 2863 births by trained TBAs in the villages during phase two for a total of 4451 births.  The TBAs referred 251 (16%) mothers to the nearest health centre or hospital during phase one and 497 mothers of the group of 2863 (17.3%) for risk factors, or complications of pregnancy or labour during phase two.  There were five maternal deaths in the total group, two during phase one and three during phase two.  The deaths occurred in hospital following transfer. Two deaths were related to malaria, a significant problem for African mothers and one from placenta accreta in spite of surgery and blood transfusion.  This mother was referred to the hospital by theTBA during pregnancy because of a previous postpartum haemorrhage.  Two other deaths were reported from the villages.  These occurred in mothers not care for by trained TBAs.  In the last two years of the project, no deaths of mothers cared for by trained TBAs were reported in the village.  However, there was a rise in maternal deaths at the district hospital (Tables 2 and 3). 

Hospital Outcomes

The data suggests that although deaths were reduced in the villages covered by the project, mothers were still dying after transfer to hospital or health centre.   There were eight maternal deaths in the hospital from 1999-2002 and 11 maternal deaths in the hospital from January 2001 to December 2002.  Ten of these deaths were mothers from the area not covered by the project, areas not offering community midwifery care.  This made analysis of the predisposing factors in those deaths difficult due to lack of data and we could only assume that the rise in hospital deaths was likely related to the higher number of referrals and an increased in-hospital birth rate.  From the hospital records we know that a significant factor in the maternal deaths was delay in the mothers reporting to the trained TBA or health facility for prolonged labour, pregnancy induced hypertension, or for risk of haemorrhage and delay in transportation. 

The comparative maternal mortality ratios for hospital, health centre and TBAs are shown in Table 4.  Based on our records, we calculated the district ratio to be 205:100,000. This is slightly lower than the most favorable ratios reported from the neighboring district 3.  However, there may have been maternal deaths in the villages in the area of the district not covered by the project.  

Referrals to hospital from TBAs or health centre:

A major objective of the project was to develop an integrated maternal health service through training of midwives and TBAs to refer women to hospital in appropriate circumstances and a timely manner.   Over the life of the project referrals to the hospital or the health centre increased.  The TBAs referred 251 (16%) mothers to the hospital or health centre during phase One and 497 (17.3%) during Phase Two.  Approximately, 25% of the mothers referred to the health centre by the TBA gave birth at the centre.   The health centre midwives referred 315 of their clients to the hospital (21%).   On reviewing the records, we found that approximately 36% of the referrals needed some type of obstetric intervention following transfer to the hospital.  

Community Involvement

It has been clear that community involvement is critical for the success of safe motherhood projects.  From the beginning, the project team have met with village leaders (usually men) to inform them about the project and seek support.  For example, the Ghanaian team believed strongly that people had to pay for use of the ambulance service to ensure sustainability after the end of the project.  The cost was negotiated with the village health committees and the fee established was based on the average taxi fare and paid directly to the hospital. Meetings, or durbars, have been held throughout the district to promote the importance of good prenatal care and care by either a trained TBA, the local health centre or the hospital.  In spite of our efforts, when the evaluators held focus groups in the community they found that more needs to be done to involve the community and more education is needed about the project4.  As well, the health insurance scheme should be extended to cover health centre clients especially maternity clients and the ambulance service.

DISCUSSION

There have been problems with some of these strategies.  The external evaluators considered the fee for maternal transport by ambulance was too low and should be increased in order for the service to be cost-effective and ensure sustainability.  Unfortunately, poverty is a major issue for the rural population and we are afraid the people will not use the service in emergency and seek less safe methods of transport if the fee in increased.  The diocesan health committee has set up a health insurance scheme for hospital care at the diocesan hospitals.  However, the scheme does not extend to community health centre care.  A review of the fee structure and use of the ambulance service should be done in 2005 to see whether cost, or other factors affect utilisation of the service.  Maintenance of the ambulance and the radio communication system is crucial for the availability, accessibility and efficiency of the service.   Therefore in the long-term the viability of the EOTS remains open to question. 

Monitoring and supporting the TBAs was a challenge.  For example, at the end of Phase One, 97 TBAs of the original 131 (74%) trained were still practicing and, in spite of training another 35, only 102 (61%) were still practising at the end of phase Two.  This is a drop-out rate of about 35%.  There are several possible explanations for this.  The TBAs set a small fee for their services, half of which would go to the village health committee. The committee would then use the fee to supply the items to re-stock the TBA kits.   However, some TBAs had trouble getting paid even the very low fee.  In the village, traditionally the TBA has been paid with a chicken or some other gift and monetary compensation for such services is foreign to the people.  Unfortunately, some money is needed to replace the items in the delivery kits to ensure cleanliness and we have asked the village health committees to address the problem in those villages if possible.  Another interesting issue was raised when the TBA referred a pregnant woman for hospital care due to a risk condition. For example, the TBAs were instructed to refer women with previous caesarean birth to the hospital. However, if the woman subsequently had a normal birth the TBA could lose face in the community as the family had a higher cost for a hospital birth that was then seen by the woman and family to be unnecessary.    As well, four TBAs died and others moved away. 

CONCLUSIONS

Implementing effective strategies to improve maternal mortality and morbidity has been a challenge.  As Lawson et al note “There are no easy solutions and short cuts to the problems of reducing the prevailing high mortality rates in poor countries” (p. 15).   These authorities conclude that death rates are increasing because the underlying problems of extreme poverty and widening inequalities have not been addressed.  As well as poverty, isolation, bad weather, and cultural beliefs and values, factors difficult to control, also played a significant part in the outcomes for this group of mothers.   

In spite of many improvements in the accessibility and availability of services and integration of the traditional birth attendants into the system, the three delays, delay in recognizing the problem, delay in transfer and delay in treatment at the hospital, which are significant factors in maternal mortality, continue to affect the overall effectiveness of the health care system.  Maternal and perinatal mortality rates continue to be unacceptably high in spite of the training programmes, emergency transport and the provision of a blood transfusion service.  It is very disappointing to find that most maternal deaths in this sample of mothers were preventable had intervention occurred earlier.   Nevertheless, safe motherhood projects such as this one have the potential to improve the quality and accessibility of the maternal health services, especially in rural areas.   As well, during the project, rural midwives felt more supported in their role as midwifery care providers in isolated areas and the traditional birth attendants appreciated their relationships with the rural midwives. 

However, such projects must be grounded in the community and reflect community values, attitudes and beliefs.  Time needs to be built in at the beginning of the project development to involve all community members, including the leaders, TBAs, women’s groups, and health professionals in determining what the community wants and how best the objectives can be achieved.   On-going mechanisms for community feedback need to be established from the outset if the community is to take ownership of the project and, more important, to keep it going after overseas funding has been withdrawn.   Durbars (community mobilizations) and media campaigns have been recommended to engage the general public in safe motherhood activities.  An informed community may be the most effective factor in improving maternal mortality rates. 

REFERENCE LIST

1.        Maternity care in developing countries.  JB Lawson,  KA Harrison, S Bergstrom.. London: RCOG Press, 2001.

2.        Thaddeus S, Maine D. Too far to walk: maternal mortality in context.  Soc. Sci. &  Med. 1994   38.  1091-10

3.        Smith JB, Fortney JA, Wong E, Amatya R, Coleman NA, deGraft Johnson J.  Estimates of the maternal mortality ratio in two districts of the Brong Ahafo region, Ghana.  Bull. of the WHO.  2001.  79:  400-8

4.        Sopiimeh WS, Adjei-Mensah E.  Nkoranza safe motherhood project:  Evaluation report.  2003.  Unpublished report

Table 1.  Partograph use and maternal and perinatal outcomes for rural midwives (1999-2002)

Table 2.  Maternal and perinatal deaths in hospital, health centre and with TBAs

Phase One (July 1999- August 2001)

*Data not tabulated in the hospital statistics because deaths occurring in the neonatal period were not recorded accurately.

+The hospital rate covered the whole district, including areas without community maternity services.

Table 3.  Maternal and perinatal deaths in hospital, health centre and with TBAs

Phase Two (September 2001-- April 2003)

Table 4.   Total maternal deaths for hospital, health centre and TBAs (July 1999-April 2003)