search
for
 About Bioline  All Journals  Testimonials  Membership  News


East African Journal of Public Health
East African Public Health Association
ISSN: 0856-8960
Vol. 3, Num. 2, 2006, pp. 10-13

East African Journal of Public Heath, Vol. 3, No. 2, October 2006, pp. 10-13

LEVELS, TRENDS AND RISK FOR EARLY NEONATAL MORTALITY AT MUHIMBILI NATIONAL HOSPITAL, TANZANIA, 1999 - 2005

Method R. Kazaura1, HL Kidanto2 and Siriel N. Massawe3

Corresponding author:  Method R. Kazaura, Department of Epidemiology/ Biostatistics, P. O. Box 65015, Dar es Salaam. E-mail:mrkazaura@muchs.ac.tz 

1Epidemiology/Biostatistics Department, School of Public Health and Social Sciences, Muhimbili University College of Health Sciences, Dar es Salaam, Tanzania
2Maternity Unit, Department of Obstetrics and Gynaecology, Muhimbili National Hospital, Dar es Salaam, Tanzania
3Dept. of Obstetrics & Gynaecology, Muhimbili University College of Health Sciences, Dar es Salaam, Tanzania

Code Number: lp06009

Abstract

Objective: To determine the magnitude, trend and to assess risk factors for early neonatal mortality in one of the referral hospitals in Tanzania
Methods: We used logistic regression analyses with data from the Maternity Unit of Muhimbili National Hospital, Tanzania, adjusting for possible confounding factors.
Results: We found early neonatal mortality rate of 20 per 1000 live births (95%CI, 19/1000 – 21/1000). Results indicated the reduced risk of 0.8 (95% CI, 0.7 – 0.9) per 10 years increase of maternal age at delivery. We also found a reduced risk of neonatal mortality by increase in birth weight of the infant (OR = 0.87: 95%CI, 0.87-0.88 per 100 grams increase). Male born babies were found to have an elevated risk (OR = 1.4, 95%CI, 1.3 – 1.5) of early neonatal mortality as compared to females and the risk of neonatal mortality among offspring of women who have history of neonatal death was 1.9 times (95%CI, 1.1 – 3.1) as compared to those without a history of neonatal death.
Conclusions: Hospital-based data understate the magnitude of early neonatal mortality but maternal age and history of previous neonatal death should be used as markers for such undesired birth outcome.

Key words: Early neonatal mortality; Risk; Tanzania

Introduction

The early neonatal period, hence neonatal deaths, commences at birth and ends seven completed days after births (1). Since the distributions of infant and mortality rates are normally skewed to the left, specifically during the first few weeks or month, it is important to put some emphasis on deaths during these early periods of life.

It is estimated that about four out of the 130 million infants’ worldwide die during the first four weeks of life and more than three million are stillborn (2-5). In developing countries, there is always lack of information on vital events that also include mortality for under-fives. Nevertheless, recent estimates indicate that about 98% of the world reported neonatal deaths and almost an equal proportion of stillbirths are from developing countries; with highest proportions from sub-Saharan Africa (3, 5, 6).

Although data on birth outcomes are important to plan maternal and child health care services, accurate indices especially from developing countries are quite difficult to obtain. For example, there are big variations in estimates of perinatal mortality in Tanzania ranging from a minimum estimate of 27 to as high as 125 per 1000 births across different regions in the country (7-12). A hospital based study in Kilimanjaro found neonatal mortality to account for two thirds of infant mortality (13). While some studies about neonatal mortality in the country have the rates ranging from 19 per 1000 births to 69 per 1000 live births (8, 9, 14), a recent community based study in northern rural area of Tanzania reported early neonatal mortality rate of 12 per

Population- and community-based under-one-year mortality studies in most of developing countries are either few or lacking because of unavailability of vital registration or due to some limitations in resources. Available studies are either too small to make inference for the whole country or based on hospital data that tend to under-estimate the magnitude of the problem. Yet, such studies maybe considered tracers of what might be happening in other areas of the country.

Several risk factors have been suggested to influence neonatal mortality. They include parity, maternal age, race, marital status, smoking, birth weight and gestation age, labour complications, frequency of antenatal visits, previous unfavourable outcomes like stillbirth and neonatal deaths and various socio-economic factors (7, 15-19). However, it has been discussed elsewhere whether parity and maternal age each has an independent effect to pregnancy outcome or one factor acts as a proxy for the other (20). Some of the available cross-sectional studies have dismissed parity to have a separate effect from that of maternal age (21). Maternal morbidity during pregnancy, for example malaria and maternal HIV has been suggested to have strong association with undesired pregnancy outcomes (22-24).

We used the complete and available routine-hospital-based data systematically collected since January 1999 to December 2005 at The Muhimbili National Hospital (MNH) in Tanzania. The hospital has a high attendance emergency and emergency obstetric care for mothers referred from neighbouring health units and also it serves as a teaching hospital for medical students. The main objective is to estimate the period prevalence of early neonatal mortality, time trend and some associated risk factors.

Materials and methods

The Muhimbili National Hospital (MNH) is one of the four big consultant hospitals in the United Republic of Tanzania situated in Dar es Salaam which is the country’s largest city. According to the 2002 National Population Census, the city had a total population of about 2.5 million (45.9% females) with annual growth rate of 4.3% (25).

The Maternity Unit belongs to the Department of Obstetrics and Gynaecology in the MNH structure. The Unit runs routine antenatal clinics and receives ‘patients’ from districts and private hospitals, clinics and dispensaries from either the city’s vicinity or upcountry. The MNH serves a teaching hospital with a range of qualified personnel, for example in the maternity and delivery, operative services and in handling complicated cases is all available in the Unit. On average, the Unit registers over 30 deliveries each day. There are early discharges for mothers who do not need post-natal care. This is one of great limitations as it likely to under-estimate the risk of early neonatal mortality.

We used data from the Unit that included all births in the hospital from January 1999 to December 2005. Delivery records are carefully registered in the book, approved by the day-in-charge and thereafter entered into the computer. Each month, a report is generated from these records and approved by the Head of the Department (Obstetrics and Gynaecology) for information and decision making.

Regular information is recorded routinely when pregnant women attend antenatal visits. Referred obstetric cases from other heath units, come with cards containing these information. Information for mothers who come directly from home is provided on arrival or after delivery by the mother or close relative subject to the health condition of the mother. Routine information includes clinical and obstetric data like weight, serostatus, gestational age, maternal previous experience (parity, previous stillbirths, neonatal deaths,) etc. Other information, like, maternal and birth outcome, mode of delivery, sex of the child, birth weight (kilograms), apgar scores, blood loss, etc are obtained during and/or after delivery.

Since multiple pregnancies maybe directly associated with neonatal mortality (26, 27), we only used singleton during the analyses. We furthermore excluded outcomes with very low birth weight (less than 1000 grams) (28, 29). Of the total 99,329 live births, we excluded 3,600 (3.6%) multiple pregnancies and 960 (1.0%) with birth weight of less than one kilogram (26.7% of multiple pregnancies had very low birth weight).

We used logistic regression modeling in our analyses of effects of neonatal mortality. Some of the selected factors and covariates as predictors of neonatal mortality include maternal age, number of antenatal visits, previous neonatal deaths and stillbirths, gestation age, birth weight, sex of the infant and parity and year of delivery. At first, maternal age at delivery and birth weight of the infant were treated as continuous variables in order to detect a linear trend by neonatal mortality risk. Then, they were all categorized into smaller intervals to minimize dangers of confounding effect within wider intervals (30). Although data were entered into the computer using Epi Info, statistical analyses were performed using SPSS for Windows version 11.0 (31, 32). Odds ratios (OR) with 95 percent confidence intervals (CI) were used as measures of strength of association. Given the numerically low prevalence of early neonatal mortality, these odds ratios are close approximations of relative risks.

Results

Between 1999 and 2005, the Maternity Unit at Muhimbili National Hospital registered 1,907 early neonatal deaths among 94,769 singleton live births of at least 1000 grams. This makes a total early neonatal mortality rate of 20 per 1000 live births (95%CI, 19/1000 – 21/1000).

The risk for early neonatal mortality among singleton increased from 20 per 1000 live births in 1999 to the peak of 26 per 1000 live births in the year 2001. It then decreased significantly to a lowest mortality level of 14 per 1000 live births in 2004 (p < 0.001) (figure 1).

In table 1, we present results from the logistic regression analyses of effects of selected factors on the risk of early neonatal mortality among offspring of women delivering at MNH. We found strong association between early neonatal mortality and maternal age at delivery, sex of the infant, previous neonatal death and birth weight of the infant. Maternal age estimated by odds ratios as a linear function of early neonatal risk per year was negatively associated with the risk of early neonatal mortality. We found highly significant estimated reduced risk of 0.8-fold (95% CI, 0.7 – 0.9) per ten years increase in maternal age at delivery. This is further exemplified by categorical maternal age groups and infant’s birth weights. For example, mothers aged less than 20 years have a 1.3-fold (95%CI, 1.1 – 1.4) significantly increased risk; whereas mothers aged 35 to 39 years have a significantly reduced risk of 0.7 (95%CI, 0.6 – 0.9) as compared to mothers aged 20 to 24 years (table 1).

In table 1, we present results from the logistic regression analyses of effects of selected factors on the risk of early neonatal mortality among offspring of women delivering at MNH. We found strong association between early neonatal mortality and maternal age at delivery, sex of the infant, previous neonatal death and birth weight of the infant. Maternal age estimated by odds ratios as a linear function of early neonatal risk per year was negatively associated with the risk of early neonatal mortality. We found highly significant estimated reduced risk of 0.8-fold (95% CI, 0.7 – 0.9) per ten years increase in maternal age at delivery. This is further exemplified by categorical maternal age groups and infant’s birth weights. For example, mothers aged less than 20 years have a 1.3-fold (95%CI, 1.1 – 1.4) significantly increased risk; whereas mothers aged 35 to 39 years have a significantly reduced risk of 0.7 (95%CI, 0.6 – 0.9) as compared to mothers aged 20 to 24 years (table 1).

Table 1.  Risks of early neonatal death among offspring of women delivering at MNH, Tanzania, 1999-2005

Risk factor

n* (Rate†)

Unadjusted OR (95%CI)

Adjusted OR (95%CI)

Maternal age§

 

Maternal age group

1850(20.1)

0.96 (0.95- 0.97)

0.98 (0.97 – 0.99)

   < 20

477 (28.6)

4 (1.2 – 1.5) 1

1.3 (1.1 – 1.4)

20 – 24

614 (21.4)

Reference

Reference

25 – 29

426 (17.8)

0.8 (0.7 – 0.9)

0.8 (0.7 – 0.9)

30 – 34

215 (15.0)

0.7 (0.6 – 0.8)

0.7 (0.6 – 0.9)

35 – 39

92 (13.5)

0.6 (0.5 – 0.8)

0.7 (0.6 – 0.9)

40 – 44

24 (17.2)

0.8 (0.5 – 1.2)

1.2 (0.8 – 2.0)

45 – 50

 (10.2)

0.5 (0.1 – 1.9)

0.7 (0.2 – 2.8)

History of stillbirth

 

 

Yes

192 (20.0)

1.0 (0.9 – 1.2)

0.9 (0.8 – 1.1)

Never

1681 (20.1)

Reference

Reference

History of neonatal death

 

 

Never

1863 (20.0)

Reference

Reference

Sex of  infant

 

 

Male

1076 (22.2)

1.3 (1.2 – 1.4)

1.4 (1.3 – 1.5)

Female

801 (17.7)

Reference

Reference

Parity

1 (26.3)

1.3 (0.2 – 9.4)

2.0 (0.3 – 15.6)

0

1837 (20.6)

Reference

Reference

1 – 5

 53 (11.7)

0.6 (0.4 – 0.7)

0.6 (0.5 – 0.9)

  6+

1858 (19.9)

0.27 (0.25 – 0.29)

0.26 (0.24 – 0.28)

Birth weight§Birth weight categories

1.0 – 1.4

332 184.3)

17.2 (15.0 – 19.7)

18.4 (15.8 – 21.4)

1.5 – 1.9

87 (90.9)

7.6 (6.6 – 8.7)

7.7 (6.6 – 8.9)

2.0 – 2.4

207 (23.0)

1.8 (1.5 – 2.1)

1.8 (1.5 – 2.1)

    2.5+

1032 (13.0)

Reference

Reference

Infant’s birth period

 

 

 

1999

343 (20.0)

Reference

Reference

2000

362 (21.0)

1.1 (0.9 – 1.2)

1.1 (0.9 – 1.3)

2001

382 (25.9)

1.3 (1.1 – 1.5)

1.4 (1.2 – 1.7)

2002

286 (21.7)

1.1 (0.9 – 1.3)

1.1 (0.9 – 1.4)

2003

217 (18.8)

0.9 (0.8 – 1.1)

1.1 (0.9 – 1.3)

2004

151 (14.1)

0.7 (0.6 – 0.9)

0.8 (0.6 – 1.0)

2005

166 (16.3)

0.8 (0.7 – 1.0)

1.1 (0.9 – 1.3)

Antenatal visits

 

 

0 – 3

625 (30.6)

1.9 (1.8 – 2.1)

1.0 (0.9 – 1.2)

   4

295 (21.3)

.3 (1.2 – 1.5)

1.0 (0.9 – 1.2)

   5+

952 (16.0)

Reference

Reference

* Number of early neonatal deaths. Some of the denominators vary due to missing information

† Per 1000 live births,     ‡ Confidence interval,  § Included in the model as continuous variables

Similarly we found a highly significant reduced risk by increased birth weight of infant (OR = 0.87: 95%CI, 0.87-0.88 per 100 grams increase). When we used categorical birth weight in the model, infants of between 1 and 1.4 kilogram had a highly significant increased risk of early neonatal mortality by 18.4 (95%CI, 15.8 – 21.4) times as compared to infants of at least 2.5 kilograms. Likewise, infants weighing between 1.5 and 1.9 kilograms were 7.7-fold (95%CI, 6.6 – 8.9) dying before four weeks as compared to infants weighing 2.5 kilograms and above. Male born babies were found to have an elevated risk (OR = 1.4, 95%CI, 1.3 – 1.5) of early neonatal mortality as compared to females. Furthermore, we found a significant indication of recurrence of neonatal mortality. The risk of early neonatal mortality among offspring of women who have history of neonatal death was almost twice as compared to those without a history of neonatal death (OR = 1.9, 95%CI, 1.1 – 3.1).

Discussion

The actual magnitude of both infant and neonatal mortality rate, especially in developing countries are either unknown or inaccurate. However, available studies from the region many of which are hospital-based (33), largely concentrate on perinatal and neonatal mortality. Using an urban and large referral hospital in Tanzania, we estimated the early neonatal mortality rate of 20 per 1000 live birth. Because of the high potentials of selection bias, this estimate maybe on the higher side as compared from a recent community-based study in northern Tanzania (7).

Our data also indicated a significant decrease in early neonatal mortality rates especially between 2001 and 2004. To the best of our knowledge, there are no recent available studies that have assessed trends in early neonatal mortality and even perinatal mortality in the region. However, a report of levels and trends of mortality in East Africa some few decades ago suggested substantial declines in infant and child mortality (34).

 There are reports suggesting strong association between neonatal mortality and demographic characteristics, obstetric factors and maternal health factors (15-19). If these pre-stated risk factors also would influence early neonatal mortality in the same direction, we only found strong association with maternal age, birth weight, sex of the child and mothers reporting of previous neonatal death. Maternal age and infant’s birth weight have been widely reported (7, 16, 18, 19). With similar direction but different estimates, one study found an increased risk of early neonatal mortality over 8 times for birth weight less than 2.5 kilograms (35), we found a risk of 4.8 (95%CI, 4.4 – 5.3) (table not indicated).

Contrary to some literature on perinatal and neonatal mortality (17), we did not find independent effect of parity and antenatal visit on increased risk of early neonatal mortality. Our findings are consistent with a study conducted in Israel on perinatal mortality (36)

Although the current study is based on a huge data set, one should be careful in generalization and comparison of results from this hospital-based study with other community-based studies because the former maybe source of potential selection bias. Furthermore, in our study, early discharge must have understated the number of early neonatal deaths. Although this factor has a direct effect on the magnitude of early neonatal mortality rate and a slight effect of the power of the study, we don not feel, however, it would bias estimated odds ratios of early neonatal mortality.

Data from this study show that although we may use hospital data help inform and design strategies for child survival, there maybe serious overstating the magnitude of neonatal or/and perinatal mortality rate. Low maternal age and a reported history of neonatal death should be considered a market for early neonatal mortality. Strategies suggested elsewhere to improve birth weight maybe revisited (37).

References

  1. World Health Organization. International statistical classification of diseases and related health problems, 10th revision. Geneva: WHO, 1992-1994.
  2. Lawn JE, Cousens S, Zupan J. 4 million neonatal deaths: when? Where? Why? Lancet 2005; 365: 891-900
  3. Zupan, J. Perinatal mortality in Developing countries. N Eng J Med, 2005; 352:2047-8
  4. World Health Organization. World Health Report 2001. Geneva: World Health Organization, 2001: 144
  5. World Health Organization. Essential newborn care. Geneva: WHO, 1996.
  6. Adam T, Lim SS, Mehta S, Bhutta ZA, Fogstad H, Mathai M, Zupan J, Darmstadt GL. Cost effectiveness analysis of strategies for maternal and neonatal health in developing countries. BMJ. 2005; 331:1107
  7. Hinderaker SG, Olsen BE, Bergsjo PB, Gasheka P, Lie RT, Kvåle G. Perinatal mortality in northern rural Tanzania. J Health Popul Nutr. 2003; 21:8-17
  8. Kilonzo A, Kouletio M, Whitehead SJ, Curtis KM, McCarthy BJ. Improving surveillance for maternal and perinatal health in 2 districts of rural Tanzania. Am J Public Health 2001; 91:1636-40
  9. 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; 46:11-3
  10. Kitange H, Swai AB, Masuki G, Kilima PM, Alberti KG, McLarty DG. Perinatal mortality in rural Tanzania. World Health Forum. 1994;15:82-4
  11. Tanner M, de Savigny D, Mayombana, C. et al. Morbidity and mortality at Kilombero, Tanzania, 1982-88: In: Disease and Mortality in sub-Saharan Africa. Washington, DC: World Bank, 1991, 286-305
  12. Moller B, Lushino, O, Kabukoba, J. et al. A prospective area-based study of the outcome of pregnancy in rural Tanzania. Ups J Med Sci. 1989; 94:101-9.
  13. Klingenberg C, Olomi R, Oneko M, Sam N, Langeland N. Neonatal morbidity and mortality in a Tanzanian tertiary care referral hospital. Ann Trop Paediatr. 2003; 23:293-9
  14. Jamieson DJ, Meikle SF, Hillis SD, Mtsuko D, Mawji S, Duerr A. An evaluation of poor pregnancy outcomes among Burundian refugees in Tanzania. JAMA. 2000; 283:397-402
  15. Weiner R, Ronsmans C, Dorman E, Jilo H, Muhoro A, Shulman C. Labour complications remain the most important risk factors for perinatal mortality in rural Kenya. Bull World Health Organ. 2003; 81:561-6
  16. Shah NM, Shah MA, Khalaf AA, Mustafa MM, Al-Sayed A. Searching for socioeconomic risk factors in perinatal mortality in Kuwait: a case control study. Soc Sci Med. 2000; 51:539-50
  17. Zeitlin J, Combier E, De Caunes F, Papiernik E. Socio-demographic risk factors for perinatal mortality. A study of perinatal mortality in the French district of Seine-Saint-Denis. Acta Obstet Gynecol Scand. 1998; 77:826-35
  18. Ibrahim SA, Babiker AG, Amin IK, Omer MI, Rushwan H. Factors associated with high risk of perinatal and neonatal mortality: an interim report on a prospective community-based study in rural Sudan. Paediatr Perinat Epidemiol. 1994; 8:193-204
  19. Meda N, Traore GS, Meda HA, Curtis V, Cousens SN, Mertens TE. Perinatal mortality in Burkina Faso: risk factors in an urban environment of Bobo-Dioulasso Ann Soc Belg Med Trop. 1991; 71:307-16
  20. Skjaerven, R. Dependencies in perinatal outcome between successive siblings. Thesis. Bergen. University of Bergen, 1989
  21. Bross DS, Shapiro S. Direct and indirect associations of five factors with infant mortality. Am J Epidemiol 1982; 115:78-91
  22. Verhoeff FH, Brabin BJ, Chimsuku L, Kazembe P, Broadhead RL. Malaria in pregnancy and its consequences for the infant in rural Malawi. Ann Trop Med Parasitol. 1999;93:S25-33
  23. Brocklehurst P, French R. The association between maternal HIV infection and perinatal outcome: a systematic review of the literature and meta-analysis. Br J Obstet Gynaecol. 1998; 105:836-48
  24. Sukwa TY, Bakketeig L, Kanyama I, Samdal HH. Maternal human immunodeficiency virus infection and pregnancy outcome. Cent Afr J Med. 1996; 42:233-5
  25. The 2002 Tanzania Population Census. Bureau of Statistics Planning Commission Tanzania. Dar-es-Salaam: Government Printers; 2002
  26. Salihu HM, Bekan B, Aliyu MH, Rouse DJ, Kirby RS, Alexander GR. Perinatal mortality associated with abruption placenta in singletons and multiples.
    Am J Obstet Gynecol. 2005; 193:198-203
  27. Misra DP, Ananth CV. Infant mortality among singletons and twins in the United States during 2 decades: effects of maternal age.
    Pediatrics. 2002; 110:1163-8
  28. Karro H, Rahu M, Gornoi K, Baburin A. Estonian medical birth registry 1992-1994: association of risk factors with perinatal mortality. Eur J Obstet Gyenecol Reprod Biol. 1998; 80: 151-6
  29. Flores Nava G, Joachin Roy H, Lopez Padilla MG, Corzo Pineda JA, Ramirez Garcia LA. Perinatal risk factors for mortality in the high-risk newborn.
    Ginecol Obstet Mex. 1998; 66:440-3.
  30. Weinberg CR. How bad is categorization? Epidemiology. 1995; 6: 345-7
  31. Epi Info 6 (Version 6.04d). Centers for Disease Control and Prevention, USA. World Health Organization. Geneva, Switzerland. 2001.
  32. Norusis MJ. SPSS for Windows Advanced Statistics, Version 11.0 (computer program). Chicago, SPSS Inc., 2001
  33. van Roosmalen J. Multiple pregnancy as a risk factor in rural Tanzania. Trop Geogr Med. 1988; 40:196-200.
  34. Ewbank DC. A re-evaluation of levels and trends of mortality in east Africa. Popul Bull UN. 1985; 18:41-
  35. Salins J, Semenciw RM, Morrison HI, Lindsay J, Sherman Gj, Mao Y, Wigle DT. Risk factors for perinatal mortality in Canada. CMJ. 1985; 133:1214-9
  36. Samueloff A, Mor-Yosef S, Seidman DS, Adler I, Persitz E, Schenker JG. Ranking risks factors for perinatal mortality. Analysis of a national-wide study. Acta Obstet Gyecol Scand. 1989; 68:677-82
  37. Kambarami RA, Matibe P, Pirie D. Risk factors for neonatal mortality: Harare Central Hospital Neonatal Unit –Zimbabwe. Cent Afr J Med. 1999; 45:169-73

© Copyright 2006 - East African Journal of Public Heath


The following images related to this document are available:

Photo images

[lp06009f1.jpg]
Home Faq Resources Email Bioline
© Bioline International, 1989 - 2024, Site last up-dated on 01-Sep-2022.
Site created and maintained by the Reference Center on Environmental Information, CRIA, Brazil
System hosted by the Google Cloud Platform, GCP, Brazil