HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Submit a response Purchase Article View Shopping Cart Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kilonzo, A. Articles by McCarthy, B. J. Search for Related Content PubMed PubMed Citation Articles by Kilonzo, A. Articles by McCarthy, B. J.

Andrew Kilonzo, MBBS, MMed, Michelle Kouletio, MPH, Sara J. Whitehead, MD, MPH, Kathryn M. Curtis, PhD and Brian J. McCarthy, MD, MSc

Andrew Kilonzo is with Bugando Medical Centre, Mwanza, Tanzania. Michelle Kouletio is with the Cooperative for Assistance and Relief Everywhere (CARE), Dar es Salaam, Tanzania. Sara J. Whitehead, Kathryn M. Curtis, and Brian J. McCarthy are with the World Health Organization Collaborating Center in Reproductive Health and the Division of Reproductive Health, Centers for Disease Control and Prevention, Atlanta, Ga.

Correspondence: Requests for reprints should be sent to Michelle Kouletio, MPH, CARE Tanzania, PO Box 10242, Dar es Salaam, Tanzania (e-mail: care-tzhq{at}care.or.tz).

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION References

 

Objectives. As part of a community-based reproductive health project in rural Tanzania, a maternal and perinatal health care surveillance system was established to monitor pregnancy outcomes. This report presents preliminary results.

Methods. Village health workers were trained to collect data during health education visits to pregnant and postpartum women. Maternal and fetal or infant survival or deaths were tracked on a community monitoring board.

Results. Among 904 pregnancies, the fetoneonatal mortality rate was 69.4 deaths per 1000 live births and fetal deaths; 4 maternal deaths occurred. Intrapartum and early neonatal deaths of infants with birthweights of 1500 g or greater represented a large proportion of deaths.

Conclusions. These preliminary results will be used to prioritize project interventions, including increasing access to skilled delivery care.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION References

 
Of the more than half million women who die each year from pregnancy-related causes, 99% reside in developing countries.¹ Despite the renewed focus on maternal health brought by the launch of the Safe Motherhood Initiative at an international conference in Nairobi, Kenya, in 1987, maternal mortality ratios in Africa are still alarmingly high, estimated at more than 1000 maternal deaths per 100 000 live births for 1995.² At the same time, recognition of the need to consider maternal and newborn care together as a "package" is increasing because many of the same underlying factors lead to both maternal and perinatal deaths.³ Addressing maternal and newborn health in rural parts of Africa offers particular challenges because of the inadequate public health infrastructure and the distance of communities from facilities providing skilled care. Also, program managers lack access to local data for basic indicators such as maternal and perinatal deaths, cesarean delivery rates, and deliveries by skilled attendants. Available data are often facility based rather than population based, and vital registration systems cover rural areas very poorly.

Cooperative for Assistance and Relief Everywhere (CARE)–Tanzania's Community-Based Reproductive Health Project works to address both health care service and management information and surveillance data needs. The project operates in 2 districts of Mwanza Region in Tanzania, with a goal of improving maternal health, maternal and newborn care, family planning, and HIV and other sexually transmitted disease prevention services.⁴ In these districts, with a population of approximately 150 000 women of reproductive age, 50% of births occur at home, often alone or attended by a relative or traditional birth attendant, and the travel time to a facility with emergency obstetric care capacity⁵ is often 6 hours or more. Vital registration of births and deaths outside of main towns is virtually nonexistent.

Through the CARE–Centers for Disease Control and Prevention (CDC) Health Initiative, a maternal and perinatal health care surveillance system was established in the project area. This village-based system includes the monitoring of pregnancy outcomes and enables local and district-level health officials to determine baseline estimates of perinatal, infant, and maternal mortality and other obstetric care indicators.

The Maternal and Perinatal Health Care Surveillance System incorporates a community monitoring board, displayed in a prominent location in each village, which tracks maternal, fetal, and infant deaths on a table charting birthweight by age at death (see Table 2 later in this article), referred to as "BABIES" (Birthweight by Age-at-Death Boxes for Intervention and Evaluation System).⁶ The community monitoring boards are color-coded to match birthweight group and age-at-death "cells" (i.e., square boxes on the board) with clusters of underlying causes and their associated prevention intervention packages. The BABIES model proposes that examining the distribution of perinatal deaths by birthweight and age at death can help to direct intervention planning and monitor the effectiveness of interventions. For example, if the intermediate- and normal-birthweight boxes for the intrapartum period indicate high rates of mortality, these deaths, which are primarily caused by asphyxia, could be prevented by strategies that increase access to cesarean deliveries.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION References

 
The maternal and perinatal health care surveillance system was implemented in 2 clusters of villages in Kwimba and Missungwi districts. The clusters included all segments of the villages and health facilities in a fixed geographic area; 22 of the 32 villages in the clusters were covered in the first phase of implementation reported here.

In each village, 2 to 3 village health workers received training in implementing the surveillance system in conjunction with broader training in reproductive health education. The village health workers identified and registered all pregnant women within their assigned area, generally during the fifth month of pregnancy. They interviewed these women during birth planning and educational visits; up to 4 times during pregnancy; immediately after delivery; and at 1 month, 6 weeks, 6 months, and 1 year after delivery. They collected information about sociodemographics, medical and obstetric history, behavioral factors, and pregnancy outcome, including place of delivery, type of delivery, delivery attendant, birthweight group, and maternal and fetal or neonatal survival or mortality. Birthweight was measured in grams when village health workers had access to scales; when scales were unavailable, birthweight was estimated based on hand size measurements. Surveillance data collection was integrated with counseling on antenatal nutrition, clean delivery, birth planning, warning signs of obstetric complications requiring medical intervention, and identification of the nearest facility with the capacity to manage complications.

Maternal and perinatal survival and mortality information was routinely displayed on a community monitoring board on which pregnancy outcomes were organized by birthweight group and age at death based on the BABIES model.⁶ Community meetings with village leadership and health staff were held periodically to conduct detailed fetal, infant, and maternal mortality reviews, which included opportunities for prevention and intervention planning.

Data from the village health workers' pregnancy registers were entered into Epi Info 6.04b⁷ by CARE project staff and analyzed in collaboration with CDC technical advisors. Only singleton births were included in this analysis. The fetoneonatal mortality rate was defined as fetal or neonatal deaths (up to 28 days postpartum) per 1000 live births and fetal deaths. The maternal mortality ratio was defined as the number of maternal deaths per 100 000 live births.

The BABIES model was used to divide outcomes into 3 birthweight groups corresponding to the level of technology associated with care for each group: normal (2500 g), intermediate (1500–2499 g), and very low (<1500 g). Outcomes were divided into time periods: antepartum (macerated stillbirths), intrapartum (fresh stillbirths), 0 to 7 days, 8 to 28 days, and further periods up to 1 year, which are not reported here. In assessing mortality rates by birthweight and age-at-death categories, birthweight-proportionate mortality rates were calculated (i.e., number of deaths in specific birthweight by age-at-death cells divided by the total number of fetal deaths and live births).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION References

 
A total of 904 pregnant women were followed up through delivery, with 858 (94.9%) followed up to at least 28 days and 828 (91.6%) to at least 42 days postpartum. The rate of low-birthweight infants (<2500 g) was 19%, including 16% intermediate-birthweight (1500–2499 g) and 3% very-low-birthweight (<1500 g) infants. The median age of the women was 25 years; few had more than a primary education (Table 1). Of the women aged 20 to 24 years, 16% had a parity of 3 or more. Slightly more than a quarter of the women had received any iron–folate supplementation during their pregnancy. A majority of the women delivered their babies at home, often alone or with a relative. Although 43% delivered at a health care facility, only 5.8% delivered at a hospital, the only type of facility currently providing emergency obstetric care.⁵ The rates for vacuum-assisted vaginal delivery and for cesarean delivery were each 0.5%.

The number of maternal, fetal, and neonatal deaths is shown, by birthweight and time of death boxes in Table 2. Figure 1 shows birthweight proportionate mortality rates, categorized according to the BABIES model, as they appeared on the community monitoring boards used in the villages.

View larger version (16K): [in this window] [in a new window]   FIGURE 1— Birthweight-proportionate mortality rates (using total live births and fetal deaths as denominator).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION References

The organization of the data in Figure 1 highlights the high mortality for birthweights of 1500 g or greater in the intrapartum and early neonatal periods. Generally, these deaths are directly related to intrapartum complications, such as asphyxia and birth trauma.¹³ One indicator of the lack of access to essential obstetric care for some of these complications is the strikingly low cesarean delivery rate of 0.4%. When we examined these deaths more closely (Table 3), again by using simple variables of birthweight and place of delivery, we found that the deaths occurred at a higher rate at home or at dispensaries, which provide a very minimal level of care. Interventions aimed at improving maternal care in labor, including the presence of skilled attendants at labor and delivery, use of a partograph to monitor labor, early recognition of obstetric complications, and transport infrastructure, would be effective strategies for preventing these deaths.^3,14 In fact, intrapartum deaths of normal-birthweight babies can be considered sentinel events, as are "near-miss" maternal deaths, and could possibly be used as one proxy outcome measure for maternal care in labor. Detailed case review of intrapartum deaths will help to identify the sources of delay in access to essential obstetric care.

The BABIES model suggests further clusters of underlying causes with the birthweight by age-at-death cells. For example, for antepartum deaths of babies with birthweights of 1500 g or greater, the major causes of death are likely to be maternal infections (including syphilis), hypertension, and diabetes.^6,13 In this setting, strategies focusing on prevention, antenatal screening, and treatment of sexually transmitted diseases are likely to be an effective intervention for reducing mortality in this group. Causes of late neonatal deaths of babies with birthweights of 1500 g or greater are likely to be neonatal infections, such as tetanus, sepsis, and acute diarrheal illnesses.^6,13 Potentially successful interventions include increasing the tetanus immunization coverage among pregnant women, ensuring clean delivery practices, and encouraging exclusive breastfeeding. This study population had a relatively low rate of adequate tetanus immunization; however, these coverage rates do not fully capture vaccinations that women received during the index pregnancy, and this aspect of the surveillance system is being improved.

Preventing deaths of very low birthweight babies requires the use of technologically advanced treatment methods associated with neonatal intensive care. This level of care is clearly not feasible for this population, and interventions therefore could focus on decreasing low birthweight by improving maternal nutrition, presumptively treating malaria in holoendemic areas, increasing the use of family planning, and extending birth spacing.^6,15,16

Further analyses of surveillance system data are planned to examine pregnancy outcomes for subpopulations (e.g., by geographic cluster or sociodemographic risk) to target interventions more precisely. Comparing project outcomes with a reference population will clarify gaps in achievable mortality reductions. In addition, CARE is implementing multiple strategies to reduce fetoinfant and maternal mortality, including development of emergency obstetric protocols, neonatal resuscitation training, training in total quality management to improve quality of clinical care, and significant work mobilizing communities around emergency transportation.¹⁷ Ongoing surveillance data will provide some of the outcome indicators for evaluation of interventions. Process evaluation, detailing the implementation steps, also will be conducted.

The simple surveillance system described here can be implemented and used at the community and district health management levels to determine trends in reproductive health outcomes and prioritize interventions in a setting of scarce resources. However, sustainability of the system will depend on the value that community members, mothers, and community leaders attribute to it and on the incentives and supportive supervision that community members provide to village health workers. The very act of recording and accounting for all pregnancy outcomes in a community is a powerful tool for focusing attention on maternal and neonatal health.

	Acknowledgments

 
All of the authors contributed to the conceptual development, analysis, and writing of the paper.

Portions of this work were supported by a grant from the R. W. Woodruff Foundation to CDC and CARE, as part of the CARE–CDC Health Initiative. Salary support for CDC technical assistance was provided by the National Center for Chronic Disease Prevention and Health Promotion.

This project has been reviewed and approved by human subjects officials (exempt; public health practice).

	Footnotes

 
Peer Reviewed

Accepted for publication June 5, 2001.

	References

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION References

 
1. World Health Organization. Revised 1990 Estimates of Maternal Mortality: A New Approach by WHO and UNICEF. Geneva, Switzerland: World Health Organization; 1996.

2. Hill K, AbouZahr C, Wardlaw T. Estimates of maternal mortality for 1995. Bull World Health Organ. 2001;79:182–193.[Medline]

3. World Health Organization. Mother–Baby Package: Implementing Safe Motherhood in Countries. Geneva, Switzerland: World Health Organization; 1998. Available at: http://www.who.int/rht/documents/MSM94-11/9411.htm. Accessed May 7, 2001.

4. Ahluwalia I, Kouletio M, Curtis K, Schmid T. Community empowerment: CDC collaboration with the CARE Community-Based Reproductive Health Project in two districts in Tanzania. J Womens Health Gender-Based Med. 1999;8:1015–1019.[Medline]

5. MotherCare Policy Brief 1. Essential obstetric care and subsets—basic and emergency obstetric care: what's the difference? Available at: http://www.jsi.com/intl/mothercare/PUBS/Policy%20Briefs/policy_brief1.htm. Accessed May 7, 2000.

6. Lawn J, McCarthy B, Ross SR. The Healthy Newborn: A Reference Manual for Program Managers. Atlanta, Ga: Cooperative for Assistance and Relief Everywhere and Centers for Disease Control and Prevention. In press.

7. Dean AG, Dean JA, Coulombier D, et al. Epi Info, Version 6: A Word Processing, Database, and Statistics Program for Public Health on IBM-Compatible Microcomputers. Atlanta, Ga: Centers for Disease Control and Prevention; 1996.

8. Walraven GE, Mkanje RJ, van Roosmalen J, van Dongen PW, Dolmans WM. Comparison of perinatal outcome in rural Tanzania as obtained from a prospective community-based survey and hospital data. Trop Geogr Med. 1994;46:11–13.[Medline]

9. Walraven GE, Mkanje RJ, van Roosmalen J, van Dongen PW, Dolmans WM. Perinatal mortality in home births in rural Tanzania. Eur J Obstet Gynecol Reprod Biol. 1995;58:131–134.[Medline]

10. Kitange H, Swai AB, Masuki G, Kilima PM, Alberti KG, McLarty DG. Perinatal mortality in rural Tanzania. World Health Forum. 1994;15:82–84.[Medline]

11. Olsen BE, Hinderaker SG, Kazaura M, et al. Estimates of maternal mortality by the sisterhood method in rural northern Tanzania: a household sample and an antenatal clinic sample. Br J Obstet Gynecol. 2000;107:1290–1297.

12. Tanzania Bureau of Statistics, Macro International. Tanzania Reproductive and Child Health Survey 1999: Preliminary Report. Dar es Salaam, Tanzania: Bureau of Statistics; 2000.

13. Wigglesworth JS. Monitoring perinatal mortality: a pathophysiological approach. Lancet. 1980;2(8196):684–686.[Medline]

14. Maine D, Rosenfeld A. The Safe Motherhood Initiative: why has it stalled? Am J Public Health. 1999;89:480–482.[Abstract/Free Full Text]

15. Steketee RW, Mutabingwa KT. Malaria in pregnant women: research, epidemiology, policy and practice. Ann Trop Med Parasitol. 1999;93:S7–S9.

16. Bouvier P. Seasonality, malaria, and impact of prophylaxis in a West African village, II: effect on birth weight. Am J Trop Med Hygiene. 1997;56:384–389.

17. Schmid T, Kanenda O, Ahluwalia I, Kouletio M. Transportation for maternal emergencies in Tanzania: empowering communities through participatory problem solving. Am J Public Health. 2001;91:1589–1590.[Abstract/Free Full Text]

This article has been cited by other articles:

				T. Wuhib, B. J. McCarthy, T. L. Chorba, T. A. Sinitsina, I. V. Ivasiv, and S. J.N. McNabb Underestimation of Infant Mortality Rates in One Republic of the Former Soviet Union Pediatrics, May 1, 2003; 111(5): e596 - 600. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Submit a response Purchase Article View Shopping Cart Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kilonzo, A. Articles by McCarthy, B. J. Search for Related Content PubMed PubMed Citation Articles by Kilonzo, A. Articles by McCarthy, B. J.