Elsevier

Social Science & Medicine

Volume 46, Issue 8, 15 April 1998, Pages 981-993
Social Science & Medicine

The “three delays” as a framework for examining maternal mortality in Haiti

https://doi.org/10.1016/S0277-9536(97)10018-1Get rights and content

Abstract

Haiti has one of the highest rates of maternal mortality in the Caribbean. The “Three Delays” model proposes that pregnancy-related mortality is overwhelmingly due to delays in: (1) deciding to seek appropriate medical help for an obstetric emergency; (2) reaching an appropriate obstetric facility; and (3) receiving adequate care when a facility is reached. This framework was used to analyze a sample of 12 maternal deaths that occurred in a longitudinal cohort of pregnant Haitian women. Because of political upheavals in Haiti during the survey, these deaths are an underestimate of all deaths that occurred in the cohort. Family and friend interviews were used to obtain details about the medical and social circumstances surrounding each death. A delayed decision to see medical care was noted in eight of the 12 cases, whereas delays in transportation only appeared to be significant in two. Inadequate care at a medical facility was a factor in seven cases. Multiple delays were relevant in the deaths of three women. Family and friend interviews suggest that a lack of confidence in available medical options was a crucial factor in delayed or never made decisions to seek care. Expanding the coverage of existing referral networks, improving community recognition of obstetric emergencies, and improving the ability of existing medical institutions to deliver quality obstetric care, are all necessary. However, services will continue to be under-utilized if they are perceived negatively by pregnant women and their families. The current data thus suggest that improvements to Haiti’s maternity care system which focus on reducing the third delay—that is, improving the quality and scope of care available at existing medical facilities—will have the greatest impact in reducing needless maternal deaths.

Introduction

Death from pregnancy-related causes represents one of the most preventable causes of female deaths worldwide (Herz and Measham, 1987). Pregnancy-related mortality in developing countries can approach several hundred times that observed in developed countries, underscoring the essentially preventable nature of these deaths. Although a relatively rare event in absolute terms, maternal mortality is among the leading causes of death for women aged 15–49 years in developing countries (Boerma, 1987; World Health Organization (WHO, 1991)). Deaths due to childbearing are not only personal tragedies, but can be catastrophic to the family, particularly to children, and to communities.

Systematic analyses of maternal deaths in developing countries are hindered by a scarcity of data, particularly in countries with incomplete or non-existent civil registration systems. Many women do not have contact with the health care or civil systems that do exist. Often, when women do have regular contact with medical facilities, there is no structure in place to systematically accumulate and evaluate their records. Finally, women often die at home and their deaths may go ignored by medical or civil authorities, with surviving family members choosing not to report the event if there is a fee for doing so. Even in developed countries with functioning vital records systems, maternal deaths are notoriously under-reported (Rochat, 1981; Koonin, 1988; Bouvier-Colle et al., 1991).

To compensate for the lack of formal data, non-traditional methodologies have been developed to study the magnitude and causes of maternal deaths. The sisterhood method, for example, is an indirect demographic technique for estimating the extent of maternal mortality in an area by measuring deaths to adult sisters of survey respondents (Graham et al., 1989). Cause of death information can be obtained through “verbal autopsies”—interviews with family members and friends about medical symptoms when the death occurred outside of the formal medical system (Fortney et al., 1986; Gray et al., 1990). While such methods have lingering problems, including concerns regarding their validity (Snow et al., 1992; Chandramohan et al., 1994), they represent useful and practical approaches to obtaining important information on numbers and medical causes of death when more formal procedures and data are not available. The analysis of why deaths occurred is less straightforward.

Complications that can lead to death occur in nearly 15% of all pregnancies worldwide (Royston and Armstrong, 1989; Koblinsky et al., 1993). While many women who develop complications have one or more detectable risk factors, the majority of women who share these risk factors do not go on to have serious problems (Maine, 1991). Screening procedures are thus of limited value for predicting complications. However, while the risk of developing complications is similar around the world, the risk of death once a complication occurs is not—nearly 99% of all maternal deaths occur to women in developing countries (WHO, 1996). Thaddeus and Maine (1990)have argued that not getting adequate care in time is the overwhelming factor leading to deaths of Third World women.

This lack of care, they contend, can be related to three factors; (1) a delay in making the decision to seek care when experiencing an obstetric emergency; (2) a delay in reaching an appropriate obstetric facility once the decision has been made to go; and (3) a delay in receiving adequate and appropriate care once the facility has been reached. Clearly these three steps are not independent: expectations of transport delays or of low quality care at the nearest facility influence the initial decision to seek care. However, the approach presents a comprehensive method for evaluating the system of maternity care, with an emphasis on identifying where improvements can best be made to save lives. We use this “Three Delays” framework to examine a sample of maternal deaths in Haiti.

Section snippets

Background

The Republic of Haiti shares the island of Hispaniola with Dominican Republic. Haiti covers a territory of 27,800/km2, much of which consists of hard-to-cultivate, eroded mountains. Projections from the most recent national census in 1982 place the current population as approaching 7 million people (Institut Haitien de Statistiques et de l’Informatique (IHSI), 1988; Population Reference Bureau, 1992). Despite increasing urbanization, 70% of the population still live in rural areas, depending

Methods

Our data are derived from a national prospective study of maternal mortality carried out in Haiti in 1990 and 1991 by the Institut Haitien de l’Enfance. The objectives of the study included obtaining information on medical causes of pregnancy-related deaths, and also examination of the non-medical circumstances that led to a woman dying from what was most likely a treatable medical condition. The study consisted of a nationwide, two-stage probability sample based on two strata: the metropolitan

Results

The 12 mortality cases came from 10 of the 58 rural study sites. The women ranged in age from 17 to 47, with a mean of 34.8 years (Table 1). Only one of the 12 deceased women reported having received any formal schooling. Compared to the entire rural sample, women who died were older, of lower socio-economic status, and had more children and adult sisters than surviving women (Table 1). Deceased women also appeared to have very different health-related behaviors, and alcohol and tobacco use

Discussion

The two summaries, along with the data of Table 2, show that multiple factors influenced obstetric care utilization by these Haitian women. While most of the medical conditions that the study women experienced were likely to have been treatable given adequate medical resources, non-medical factors conspired to prevent appropriate care from being received. Studies from many parts of the world, most of them institution based, have concluded that inadequate care is a significant factor in maternal

Conclusions

The desire to use modern medicine is irrevocably shaped by perceptions of the care that will be received: whether it is appropriate for the problem as perceived by the people involved, whether it will be efficacious, and whether it will be reachable and affordable (Myntti, 1988). The current data show that the women were actively taking steps to insure their health, and were making choices about medical care that were rational within their own system of beliefs and knowledge. When symptoms are

Acknowledgements

The authors would like to thank Wendy Graham and Deborah Maine for helpful comments on the manuscript.

This research was supported by the National Research Council, BOSTID grant no. REA-HT-2424-184. The contents of this manuscript are solely the responsibility of the authors and do not necessarily represent the official views of the National Research Council.

References (54)

  • Bolton, P., Kendall, C., Leoutsini, E. and Whitaker, C. (1990) The Impact of Health Technologies on Women in Developing...
  • M.H. Bouvier-Colle et al.

    Reasons for the underreporting of maternal mortality in France, as indicated by a survey of all deaths among women of childbearing age

    International Journal of Epidemiology

    (1991)
  • J. Bruce

    Fundamental elements of the quality of care: a simple framework

    Studies in Family Planning

    (1990)
  • Campbell, O. M. R., Koblinsky, M. A. and Taylor, P. (1993) Off to a Rapid Start: Appraising Maternal Health and...
  • D. Chandramohan et al.

    Verbal autopsies for adult deaths: issues in their development and validation

    International Journal of Epidemiology

    (1994)
  • J. Coreil

    Traditional and western responses to an anthrax epidemic in rural Haiti

    Medical Anthropology

    (1980)
  • J. Coreil

    Parallel structures in professional and folk health care: a model applied to rural Haiti

    Cultural Medicine and Psychology

    (1983)
  • Deming, M. (1993) Centers for Disease Control and Prevention, Atlanta, GA. Personal communication regarding...
  • Départment de Santéet de la Population (1984) Programme d’Entrainement des Matrones. Division d’Hygiène Familiale et de...
  • Farmer, P. E. (1994) The Uses of Haiti. Common Courage Press, Monroe,...
  • J.A. Fortney et al.

    Reproductive mortality in two developing countries

    American Journal of Public Health

    (1986)
  • Freedman, L. P. and Maine, D. (1993) Women’s mortality: a legacy of neglect. In The Health of Women: A Global...
  • W. Graham et al.

    Estimating maternal mortality: the sisterhood method

    Studies in Family Planning

    (1989)
  • Gray, R., Smith, G. and Barss, O. (1990) The Use of Verbal Autopsy Methods to Determine Selected Causes of Death in...
  • Grossman, J. (1990) Health education: the leading edge of primary health care and maternal and child health. In Health...
  • Harvard Center for Population and Development Studies (1993) Sanctions in Haiti: Crisis in Humanitarian Action. Harvard...
  • Herz, B. and Measham, A. R. (1987) Programme pour la maternitésans danger. World Bank Discussion Papers, No. 9,...
  • Cited by (0)

    View full text