Summary
- An effective maternity referral system is an essential component of district primary healthcare for women in reducing maternal mortality. It is a dynamic process whereby a health-seeking individual is referred – either through a self-presentation or a professional referral process – to a higher level of healthcare when she requires it: before, during or after delivery.
- Maternity referral systems are specific to each region or district and are created in the context of existing local structures.
- The referral system can be conceptualised by the “sender-transport-receiver” model: the “sender” is often a locally based health professional referring to a major hospital; the transport is often individually organised; the receiver is often the district hospital. However, in reality, maternity referral systems are extremely complex and there are a huge amount of factors influencing these three components of the model.
- Different levels of referral facilities are:
- Individual/community,
- Local healthcare facility,
- Regional or district Hospital.
- Maternity referral systems have been evaluated by the following measures:
- The spread of births within a district,
- The utilization of resourced hospitals by women with obstetric complications,
- The progress towards reduction of maternal mortality at referral facility level,
- Inappropriate use of EOC level facility,
- Perinatal outcomes at peripheral-level facility,
- Cost-effectiveness analysis.
- Key barriers to and determinants of maternity referral systems include cost, geographic and transport, political focus, local education and health system communication between all levels of care. Access to an appropriate level of care is influenced by geography, transport, culture and perceived quality of care at the service provider.
- Current interventions in developing countries focus on education at a community level, reducing geographical and financial barriers and on improving transport and communication.
- Inequity is still a prolific issue in the implementation of and access to maternity healthcare in developing regions, including effective referral. Poor-rich inequity, rural-urban inequity and inequity in marginalised communities are examples.
- There is a wealth of knowledge in the area of implementing effective maternal health programs in developing countries. Wide-scale international awareness of maternal mortality in developing countries has been around for over 25 years, with the instigation of the Safe Motherhood Initiative.
Background
Approximately 26% of the world’s maternal deaths occur in India annually (Shah, Divakar and Meghal 2006). Of the total 529 000 annual global maternal deaths, 136 000 occur in India (Shah, Divakar and Meghal 2006). India’s maternal mortality ratio (MMR) – defined as the number of maternal deaths per 100 000 livebirths – was 437 per 100,000 live births in 1991 (Shah, Divakar and Meghal 2006). According to the Indian Sample Registration System, the MMR has declined further from 301 during 2001-2003 to 254 in 2004-2006 (UNICEF http://www.unicef.org/india/health.html). However, for every 1 maternal death, 20 additional women are reported to experience serious harm relating to delivery (The UK All Party Parliamentary Group on Population, Development and Reproductive Health, 2009).
Post partum haemorrhage – excessive loss of blood from the reproductive tract after delivery – is the leading cause of death, contributing to 29.6% of all maternal deaths in India (Shah, Divakar and Meghal 2006). Other leading causes of maternal death include infections, eclampsia, and unsafe abortions.
India is a large country. 70% of the population lives in rural areas (Shah, Divakar and Meghal 2006). Of the 25 million annual deliveries in India, 18 million occur in peripheral areas, which, through its nature, translates into less ready access to healthcare and services (Shah, Divakar and Meghal 2006).
Strategies employed to prevent maternal deaths include risk identification, increasing the proportion of births attended by skilled health professionals, improving access to emergency obstetric care, and ensuring a functional referral system. These strategies have various associated historical levels of success and evidence. In India, the proportion of births attended by a skilled attendant increased 25.5% in 1992 to 39.8% in 2002. Although previously a large focus, training of Traditional Birth Attendants has not been shown to be effective in reducing maternal mortality. Likewise, risk prediction – attempting to screen all pregnant women and categorise those most at risk of an adverse pregnancy outcome – has proven inaccurate, with little effect on maternal mortality (Shah, Divakar and Meghal 2006). It is difficult to predict those who will suffer from a post partum complication through risk factor identification alone(Shah, Divakar and Meghal 2006). This has led to the recommendation of “every woman is a at risk.” (WHO 1991). Shah, Divakar and Meghal recommend three key elements of obstetrics care that have the ability to reduce maternal mortality:
(1) A skilled attendant at delivery;
(2) Access to emergency obstetric care (EOC);
(3) A functional referral system.
(Shah, Divakar and Meghal 2006).
Of course, appropriate human utilization of limited resources in a healthcare system are essential to ensure optimal health outcomes for women and children. Delays in care may occur at any level and may be attributed to both material and non-material factors. A model proposed by Shah, Divakar and Meghal (Shah, Divikar, Meghal 2006) suggests that delays in healthcare opportunities that lead to maternal death can be summised by three factors:
(1) Delay in deciding to seek care;
(2) Delay in reaching care;
(3) Delay in getting treatment at the facility.
Strengthening the referral system is important to reduce delays. A strong referral system equates to women being referred in a timely, proactive and efficient way. Especially, in the case of maternal mortality, time is of the essence due to the nature of rapidly evolving medical emergencies. As Shah puts it, inefficiency is deadly.
Unfortunately, it has been reported that widespread ‘failures’ in maternity referral systems are often present in developing countries. (Shah, Divakar and Meghal 2006). Moreover, these failures detrimentally affect women from a poor and marginalized background. There appears to be a chasm between theory and reality when it comes to a functional and effective referral system. A review by Murray and Pearson found that academic conceptualization of an ideal referral system may have a dangerously tenuous relationship to realities on the ground. (Murray and Pearson 2006)
What is a maternity referral system?
The referral system is an essential component of district health systems. It is a dynamic process whereby a health-seeking individual (in this case, the pregnant woman) mobilises directionally through a healthcare system to realise optimal healthcare for her condition. Jahn and de Brouwere define referral as: “any upwards movement of health care seeking individuals in the health system” and categorise maternity referrals as (Jahn and de Brouwere 2001):
(1) institutional or self-referral, depending on the involvement of first line services;
(2) antenatal, delivery or postnatal referral; and
(3) elective or emergency referral
Effective primary healthcare requires adequate referral systems in place to facilitate access to emergency healthcare for those who need it. An effective referral system allows primary healthcare to be provided by local professionals close to the patient’s home whilst serving as a back-up in case of emergency.
Each referral system operates within the broader context of the regional healthcare system and is appropriated to this system. In general, health systems are defined by the World Health Organisation (WHO) as: “all the activities whose primary purpose is to promote, restore, or maintain health” (WHO 2000). An individual healthcare system, therefore, is extremely complex, encapsulating healthcare facilities, health education, health system financing, health regulatory systems and non-government organizations.
Levels of referral
Regarding regional maternity services, the following levels of healthcare have been identified (WHO 1996):
- District or sub-district hospital, to which a woman at high risk is referred prenatally or sent for emergency obstetric care, and where the following essential services should be available:
- surgical obstetrics,
- anaesthesia,
- medical treatment,
- blood replacement,
- manual procedures and monitoring labour,
- management of women at high risk,
- family planning support and
- neonatal special care (WHO 1991).
- Local health centre
- Family and community
These levels of care are dynamically linked by a system of referral from one level to another. In the case of obstetric emergencies, women who are unwell would be referred by their local health centre to a hospital where they could receive adequate medical, surgical and neonatal care.
Linking levels of care through referral
The dynamic process of referrals is an essential element of primary healthcare systems. It ensures that healthcare centres can function locally and can link into more advanced care if needed (King 1966). The back-up function of a referral system plays an integral role in pregnancy and childbirth, where a range of life-threatening complications may occur rapidly and require more advanced medical care. (Jahn and de Brouwere 2001).
Elements of the referral system
A referral system involves both physical and process components. Jahn and Brouwere conceptualized the sender-transport-receiver model of healthcare referrals. It identifies the main actors and actions in the system linking the locations and processes of referral between levels of healthcare. (Jahn and De Brouwere 2001) For example, having communication with a referral hospital will not be effective if there is no form of transport to get from A to B. Likewise, despite efficient transport, if local stigma prevents women from attending hospitals, this system will fail.
With such a complex system, each part is integral to the whole. In the sender-transport-receiver models, many different elements play an important role, including individuals, policy makers, health workers, transport systems and finances. While a simplified model, it is apparent how complex this would become in real life.
It is important to support and educate the sender/referrers in this system to ensure only adequate referrals are obtained. In the context of most district hospitals being overburdened already, this is an important mechanism to reduce this workload and streamline patients appropriately. Moreover, it has been demonstrated that adequate outreach services in the community have the capacity to reduce the burden on hospital systems. Bypassing by healthy women of local clinics in preference for a district hospital continues to be a problem and may be representative of lack of confidence in quality of care or of perceived or actual inefficiency in the local referral services (Murray and Pearson 2006)
Elements of an effective maternity referral system include:
- Adequately resourced referral centre
- Communication and feedback systems
- Designated transport
- Agreed setting-specific protocols for the identification of complications
- Personnel trained in their use
- Teamwork between referral levels
- Unified record system
- Meachanisms to ensure patients do not bypass a level of the referral system eg patient education, information and a structured fee/exemption systems (Murray et al 2001)
Types of maternity referrals
Many different referrals are made during pregnancy. Studies from a rural Tanzanian hospital indicate that the most common referral pattern is self-referral, whereby referral to a hospital for delivery is made by the individual based on individual thought processes and perceptions of the pregnancy (Jahn 1998). In the above hospital, 70% of all admissions for delivery were self-referral, compared with only 30% of women being referred through a local clinic or institution. This pattern has been reflected in Burkina Faso, and Nepal (Jahn and de Brouwere 2001). This represents a skewed pattern of referral, with many women by-passing primary care options altogether (Jahn and de Brouwere 2001). Of note also, is that emergency referrals only comprised 1.2% of all referrals to high-level obstetric care, the remainder being elective referrals.
Benefits of the referral system
Full access to and utilization of all proven, effective interventions in pregnancy and childbirth would avert two-thirds of child deaths and three-quarters of maternal deaths (Wagstaff and Claeson (2004) (Jones and others 2003). Half of the maternal deaths could be averted through access to essential emergency care for a) haemorrhage, b) sepsis, c) eclampsia and d) obstructed labour. (Murray and Pearson 2006). The referral system is what links women from the community into appropriate emergency care.
Monitoring referral systems
To thoroughly monitor the effectiveness of a regional health centre’s maternity referral system, Murray et al (2001) proposed a measurement tool consisting of five different elements:
- The distribution of births within a district, across levels of different facilities and the population coverage of maternity services achieved
- The spread and number of women who deliver at home, in a local health centre, or in hospital
- How many deliveries are covered by health services
- The utilization of resourced hospitals by women with obstetric complications
- Pregnancy related referrals to hospitals
- “Met need” for hospital-based pregnancy; it is usually estimated that around 15% of women will require essential hospital care during pregnancy or delivery
- Intrapartum referral rates; are women who have acute labour problems referred in a timely way
- Caesarean section rates, a surrogate measure for complicated deliveries
- The progress towards reduction of maternal mortality at referral facility level
- Case fatality rates
- Maternal mortality reviews
- Inappropriate use of EOC level facility
- The proportion of births who self-referr who have “simple” deliveries
- Perinatal outcomes at peripheral-level facility
- Intra-partum stillbirths
- Apgar scores of newborns, a measure of neonatal wellbeing and is a inaccurate indicator of birth asphyxia (Murray et al 2001)
A cost effectiveness analysis of strategies for maternal and neonatal health was performed by Adam et al (Adam, 2005). The most effective interventions, measured by the cost per disability adjusted life year averted in the year 2000 in two poor global regions in Africa and South-east Asia, include community-based newborn care package, maternity care programs (involving tetanus immunization, screening for pre-eclampsia, screening and treatment of syphilis), skilled attendance at birth, and emergency obstetric care around birth. Although not measured directly, the referral system plays an important role in the delivery of such interventions.
The determinants of a maternity referral system
Barriers to accessing maternity care through a referral system include issues of distance and geography, access to transport, financial constraints of families, cultural constraints, and individual perceptions of quality of care provided at the local hospital. Other determinants of a maternity referral system include political willpower, money, communication between levels of care, and education of local individuals and staff to know when to refer.
Issues of funding play a prominent role in any health service or health intervention. However, as Freedman et al argue, “the financial costs of meeting the maternal and child health Goals are dwarfed by what the world spends on preparing for and waging war. Indeed, they are dwarfed by the enormous sums already spent on interventions that do not reach those who need them—and by the terrible price being paid in human lives as a result.” (Freedman et al. 2007)
On an individual level, costs of procuring transport to present to hospital, and costs involved of hospital care can sometimes be exorbitantly expensive for individual families. In areas such as cost of transport to get from the community to the hospital may be as much as 20 times the mean daily salary. 18% of all deaths in a South Africa-wide maternal deaths enquiry were attributed to lack of access to transport (National Committee on Confidential Enquiries into Maternal death 2000). If geographic factors present a barrier to ready-access to a hospital, then it often becomes an unachievable individual option to hire or access a mode of transport.
Political will at both regional and local levels also is a powerful force in the creation and implementation of maternity referral systems, and in creation of health and social policies that will facilitate this. The Millennium Development Goals (MDGs), aspirational targets set at the United National Millennium Summit, serve as powerful and measurable international targets to which national governments are accountable. The MDG 5, reducing maternal mortality, puts maternal health and delivery healthcare systems at the forefront of polical view. Unfortunately, of the eight targets, maternal health is one that is lagging behind the most in achieving the targets.
Gender and cultural issues present unique challenges to healthcare systems, even more so in the area of childbirth, where often cultural normal and traditions play an enormous role. Jahn and De Brouwere argue that rural women in particular fear stigmatization and discrimination; they are concerned of loss of dignity, an unfamiliar environment, and loss of emotional support, to name a few (Jahn and De Brouwere 2001). With limited resources, it is often difficult to provide a spectrum of services to cater for individual specific needs. However, cultural training and appropriate mix of medical professionals are strategies employed to overcome these.
Likewise, perceived quality of care by an individual accessing hospital care is an important yet difficult-to-measure barrier in the decision making process of individuals in accessing hospital services. It is interesting to note that, in a referral hospital in Karachi in 1993, the 118 women who died during childbirth lived within an 8km radius of the hospital (Jafarey & Korejo 1993). The lack of easily measurable physical barriers means that important cultural, communication and quality aspects of care are less readily identified and measured.
Communication between health professionals and levels of healthcare, not only during emergencies, but also for education and prevention, is a key concern. Whilst – reported in a South African maternal death inquiry – 18% of deaths were attributable to lack of transport, a huge 57% were due to in-service problems. Communication is essential to overcome this. (National Committee on Confidential Enquiries into Maternal death 2000).
Recognition of obstetric complications at a local level will allow timely referral to appropriate care facilities. To ensure swift recognition of these issues, investing in education and communication between levels of care plays a vital role.
The Current Situation
It has been demonstrated that hospital-based obstetrics care referrals predominantly consist of self-referrals with a low level of emergency referrals. It has also been demonstrated that to reduce maternal mortality, it is imperative to improve emergency referral processes. (Jahn and de Brouwere 2001).
The end-point of a patient’s journey through the referral system in most instances is the secondary or tertiary hospital. In order to realize the full potential of the maternity referral process, hospital-based care must be of an adequate standard. Unfortunately, it is often the case that in developing regions, hospital care is of a sub-standard quality and may contribute to maternal mortality. (Jahn and de Brouwere 2001).
Current interventions in developing countries focus on education at a community level, reducing geographical and financial barriers and on improving transport and communication.
In the field of community education, Sierra Leone (Kandeh et al. 1997) and Nigeria (Nwakoby et al. 1997) report promotion of utilisation of obstetric care through community activists, educating local communities in referral of obstetrics emergencies. A Ugandan project focused on improving multiple areas of the referral system including linking trained birth attendants to hospitals through walkie-talkies and radio systems. This has increased obstetric referrals by threefold. (Inter-agency group for safe motherhood 1997)
Improving accessibility to healthcare centres has been a focus of many past and current programs. For example, the Safe Motherhood program in Tanzania demonstrated difficulty in keeping vehicles on the road due to lack of maintenance and fuel (Jahn and de Brouwere 2001). Olaniran (Olaniran et al. 1997) describe a loan scheme to fund local car owners who act as on-call transport in the case of local obstetric emergencies. Likewise, in Mali, referral loan funds are available and are managed by local health committees. (Jahn and De Bouwere 2001)
The global Safe Motherhood Initiative was implemented in 1987. This project encouraged international developmental focus on maternal mortality in pregnancy and childbirth. The main killers of women during childbirth were identified and the barriers to accessing care were explored. Aspects of the Safe Motherhood Initiative have been incorporated into much international and national health policy. (Inter-agency group for Safe Motherhood 2012)
There is still a long way to go in strengthening maternity referral systems in developing countries. The current situational failures of maternal healthcare in developing countries can be summarised by the “Three Delays” model. This model identifies three crucial points that may critically affect access to maternal health services:
- Phase 1 delay. Delay in decision to seek care
- Failure to recognise complications
- Acceptance of maternal death
- Low status of women
- Socio-cultural barriers to seeking care: women’s mobility, ability to command resources, decision-making abilities, beliefs and practices surrounding childbirth and delivery, nutrition and education
- Phase 2 delay. Delay in reaching care
- Poor roads, mountains, islands, rivers – poor organisation
- Phase 3 delay. Delay in receiving care
- Inadequate facilities, supplies, personnel
- Poor training and demotivation of personnel
- Lack of finances
(Maternity Worldwide, 2012)
Inequities and Failures
Effective strategies to reduce maternal mortality in developing countries such as a maternity referral system have been written about for over 25 years (Say and Raine 2007). The concept of a maternity referral system is not a new idea. Despite a large volume of academic literature on the subject, it seems that frustratingly there has been slow, if any, actual progress. Key health programs and interventions continue to have limited uptake in developing countries.
Within countries, poor-rich inequities in access to maternal healthcare remains a huge problem. In particular, there is a large gap in accessing professional care for delivery between the top socioeconomic quintile and the bottom socioeconomic quintile of women in a region. Even delivery care provided by nurses and midwives favours the rich (Howelling 2007).
Marginalized women, especially religious or ethnic minorities, and women from a rural background may find accessing healthcare services difficult. It is recognized that women often fear stigmatization and discrimination; they are concerned about loss of dignity, an unfamiliar environment, and loss of emotional support (Jahn and De Brouwere 2001). Religious, language, and financial barriers all culminate in an often-insurmountable barrier to accessing care through the maternity referral system.
What can be done?
The ideal scenario is this: as part of an integrated primary healthcare system, every birth, whether it takes place at home or in a facility, is attended by a skilled birth attendant, backed up by facilities that can provide emergency obstetric care and essential newborn care and by a functioning referral system that ensures timely access to the appropriate level of services in case of a life-threatening complication. (Freedman et al. 2007)
Strategies for tackling maternal and neonatal mortality should focus on delivery and the immediate postpartum period. This requires an effective referral system to ensure access to care when clinically indicated. The following recommendations have been collated from a review of the current literature. To ensure an effective referral system the following must occur:
- Ensure referral hospital facility is of an adequate standard of care, including a 24/7 service, adequate resources, and adequate staff training. This can be audited, monitored, improved through targeted goals.
- Raising awareness of complications and danger signs of obstetric emergencies so that community members are able to identify and refer women in a timely manner
- Ensure adequate community resources for emergency communication and transport
- Engagement of local community members and health workers (Murray and Pearson 2006)
- A referral strategy informed by assessment of population needs and health system capabilities
- Active collaboration between referral levels and across sectors
- Formalized communication and transport arrangements
- Agreed setting-specific protcols for referrer and receiver
- Accountability for provider’s performance and supportive supervision
- Pro-poor protection against the costs of emergency referral
- Capacity to monitor effectiveness
- Policy support (Murray and Pearson 2006), preventing excessive segmentation of the health system and increasing the power of the poor and other marginalized groups. (Freedman et al. 2007)
- Focus on building capacity at a local level through solid primary healthcare systems and a focus on district-level care (Freedman et al. 2007)
- Health systems must be valued as core social institutions that will assist in the reduction of poverty and inequity and for advancing human rights.
- Increase funding in areas of need through bilateral donors and international financial institutions
- User fees should be abolished at a local level so it does not become another financial burden on an already impoverished population
- Future capacity planning to create a workforce of local skilled birth attendants—the health workers key to reducing maternal deaths.
REFERENCES
Adam et al Cost effectiveness analysis of strategies for maternal and neonatal health in developing countries. BMJ 2005;331:1107
Gareth Jones, Richard W Steketee, Robert E Black, Zulfiqar A Bhutta, Saul S Morris, and the Bellagio Child Survival Study Group. How many child deaths can we prevent this year? The Lancet. 2003;362
Houweling TAJ, Ronsmans C, Campbell OMR & Kunst AE. Maternity care in developing countries: a study of inequalities. Bulletin of the World Health Organization .2007;85:745–754
Jahn A, De Brouwere V. Referral in pregnancy and childbirth: Concepts and strategies. Stud Health Service Organisation Policy 2001; 17: 229-246.
Jahn A, Kowalewski M & Kimatta SS (1998). Obstetric care in Southern Tanzania: Does it reach those in need? Tropical Medicine and International Health 3,926-932.
Jafarey SN & Korejo R (1993). Mothers brought dead: An inquiry into causes of delay. Social Science and Medicine 36,371-372.
Kandeh et al. Community motivators promote use of emergency obstetric services in rural Sierra Leone. International Journal of Gynecology & Obstetrics. 1997;59(2):S209-S218
Kowalewski M, Jahn A & Kimatta SS (2000). Why do at-risk mothers fail to reach referral level care? Barriers beyond distance and cost. African Journal of Reproductive Health 4,100-109.
Lynn P, Freedman RJ, Waldman H de Pinho, Wirth ME. Who’s got the power? Transforming health systems for women and children. UN Millenium Project Task Force on Child Health & Maternal Health, 2005:77–95
Maternity Worldwide. The three delays model. Visited 2012
http://www.maternityworldwide.org/about-us/information-and-resources/the-3-delays-model/
Murray SF, Buller AM, Bewley S, Sandall J. Metrics for monitoring local inequalities in access to maternity care: developing a basket of markers from routinely available data. Qual Safe Health Care 2010;19:e39.
Murray SF, Pearson SC. Maternity referral systems in developing countries: Current knowledge and future research needs. Social Science and Medicine. 2006:62;2205-2215
Murray SF, Pearson S. 2004. “Maternity Referral Systems in Developing Countries: Challenges and Next Steps. A Scoping Review of Current Knowledge.” Background paper commissioned by the UN Millennium Project Task Force on Child Health and Maternal Health and the World Health Organization. New York.
National Committee on Confidential Enquiries into Maternal death (2000). A review of maternal deaths in South Africa during 1998. National Committee on Confidential Enquiries into Maternal Deaths. S.Afr.Med J. 90,367-373.
Nwakoby et al. Community contact persons promote utilization of obstetric services, Anambra State, Nigeria. International Journal of Gynecology & Obstetrics. 1997;59(2):S219-S224
Safe Motherhood Group. Safe Motherhood Group Website. Visited 2012
http://www.safemotherhood.org/priorities/initiatives.html
Say L, Raine R. A systematic review of inequalities in the use of maternal health care in developing countries: examining the scale of the problem and the importance of context. Bulletin of the World Health Organization .2007;85:812–819
Shah, Divakar and Meghal. Combating Postpartum Hemorrhage In India: Moving Forward. In B-Lynch C, Keith L, Lalonde A, Karoshi M. A Textbook of Postpartum Hemorrhage. A comprehensive guide to evaluation, management and surgical intervention. Jaypee Brothers Publishers. 2006
The UK All Party Parliamentary Group on Population, Development and Reproductive Health, 2009
UNICEF. Indian Health Information. Visited 2012. http://www.unicef.org/india/health.html
Wagstaff and Claeson. The millennium development goals for health : rising to the challenges. The World Bank, 2004.
WHO. Essential elements of obstetric care at first referral level, Macmillan Press Ltd., London, 1991.
WHO. Care in Normal Birth: a practical guide, Maternal and Newborn Health/Safe Motherhood Unit, Family and Reproductive Health, WHO, Geneva, 1996
WHO Mother-Baby-Package: Implementing safe motherhood in countries, Maternal Health and Safe Motherhood Programme, WHO, Geneva, 1994
Women Deliver. Homepage. Visited 2012.
http://www.womendeliver.org/about/the-issue/three-core-strategies-to-save-lives/