| The determinants of infant and child mortality in Tanzania. | |
| | |
MedLine Citation:
|
PMID: 10154361 Owner: NLM Status: MEDLINE |
Abstract/OtherAbstract:
|
This paper investigates the determinants of infant and child mortality in Tanzania using the 1991/92 Tanzania Demographic and Health Survey. A hazards model is used to assess the relative effect of the variables hypothesized to influence under-five mortality. Short birth intervals, teenage pregnancies and previous child deaths are associated with increased risk of death. The Government of the United Republic of Tanzania should therefore maintain its commitment to encouraging women to space their births at least two years apart and delay childbearing beyond the teenage years. Further, this study shows that there is a remarkable lack of infant and child mortality differentials by socioeconomic subgroups of the population, which may reflect post-independence health policy and development strategies. Whilst lack of socioeconomic differentials can be considered an achievement of government policies, mortality remains high so there is still a long way to go before Tanzania achieves its stated goal of 'Health for All'. Although infant and child mortality declined in Tanzania during 1967-93, the levels are viewed as still too high. Analysis of statistics from the 1991-92 Tanzanian Demographic and Health Survey is performed in order to determine the factors related to infant and child morality at the individual level. The nationally representative sample includes live births since January 1, 1986, among 9238 women aged 15-49 years. The quality of data is considered good. Piecewise constant proportional hazards models are used for determining the risk at 0 months, 1-7 months, 8-14 months, and 15-59 months. Significant risk factors identified for neonatal mortality are birth order, preceding birth interval, type of place of residence, and survival status of the preceding child. Risk was 38% lower in rural areas. The relative risk of mortality under 1 month was 1.48 for those with a preceding sibling death. Birth intervals under 2 years were related to an increased neonatal mortality risk, particularly for births of 5th or higher order. Mortality risk at age 1-59 months was significantly associated with maternal age, birth order and preceding birth interval, survival status of the preceding birth, partner's education, and zone of residence. The hazard decreased with increasing maternal age. Teenage mothers and mothers aged over 35 years had higher risks. There was lower risk among children of mothers who never lived with a partner. Risk was lowest in the Northern zone and in Zanzibar. Findings are interpreted as support for policy that encourages longer birth intervals (beyond 2 years). Special care should be provided for nulliparous women, women aged under 20 years or over 35 years, and women with prior deaths. The findings are viewed as a surprise due to the lack of differentiation by socioeconomic status. Increased rural development and low literacy contributing to underreporting in rural areas are viewed as potential explanations for lack of urban advantage. Further declines are expected to be conditioned on improvements in quality of care and economic development. |
| | |
Authors:
|
A J Mturi; S L Curtis |
Publication Detail:
|
Type: Journal Article; Review |
Journal Detail:
|
Title: Health policy and planning Volume: 10 ISSN: 0268-1080 ISO Abbreviation: Health Policy Plan Publication Date: 1995 Dec |
Date Detail:
|
Created Date: 1996-04-04 Completed Date: 1996-04-04 Revised Date: 2005-11-16 |
Medline Journal Info:
|
Nlm Unique ID: 8610614 Medline TA: Health Policy Plan Country: ENGLAND |
Other Details:
|
Languages: eng Pagination: 384-94 Citation Subset: H; J |
Affiliation:
|
Department of Social Statistics, University of Southampton, UK. |
Export Citation:
|
APA/MLA Format Download EndNote Download BibTex |
| MeSH Terms | |
Descriptor/Qualifier:
|
Birth Intervals Birth Order Child Welfare / statistics & numerical data* Child, Preschool Data Collection Demography Health Policy* Health Services Research Health Surveys* Humans Infant Infant Mortality* Infant, Newborn Multivariate Analysis Risk Factors Socioeconomic Factors Tanzania / epidemiology |
From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine
Previous Document: Private payers of health care in Brazil: characteristics, costs and coverage.
Next Document: Knowledge, attitudes and practices during a community-level ivermectin distribution campaign in Guat...