Return to search

Association between HIV/AIDS related adult deaths and migration of household members in rural Rufiji District, Tanzania

MSc, Population-Based Field Epidemiology, Faculty of Health Sciences,University of the Witwatersrand / Introduction: The spread and prevalence of the HIV epidemic has resulted in extensive demographic,
social and economic impacts among families in the communities affected in Sub Saharan Africa
which increase with the severity and duration of the epidemic. The dramatic increase in adult
mortality attributable to HIV/AIDS in households in these communities may increase the number of
households that do not survive as a functional and cohesive social group in the years to come. The
migration of household members and possible dissolution of these households are the challenges
stemming from the epidemic. We therefore require rigorous empirical research on the socioeconomic
effects of HIV/AIDS in order to develop appropriate strategies to mitigate these impacts and
ultimately improve living standards in these communities. This report describes the extent at which
these impacts are felt by a rural community using data from the Rufiji HDSS in rural Tanzania.
Design: The study will use a longitudinal study design to identify antecedent events and dynamics
and trans-temporal aspects in establishing the effects HIV/AIDS, and particularly how adult deaths
from the disease determine migration of individual household members, controlling for other
individual level and household factors.
Objectives: The main objectives of the study include the description of the adult mortality patterns in
the area with an emphasis on the HIV/AIDS related adult deaths, the description of the socioeconomic
and demographic characteristics of households experiencing these adult deaths; the
characterisation of the members migrating from the households as a result of these adult deaths or
otherwise. We also estimate the proportion of household members migrating following the deaths of
adult members and further compare these rates of migrations from households experiencing adult
HIV/AIDS, Non-HIV/AIDS deaths and where there is no experience of death.
Methods: Migrating individuals from 4,019 households that experienced at least one adult death were
compared with migrating individuals from other households experiencing Non-HIV/AIDS deaths and
those from households without deaths. A total of 32, 787 households were included in the study. An
adult death was defined as a death of a household member aged 18 years and above. Those aged 60+
years were considered elderly deaths. A total of 4,603 adult deaths were recorded over the period 1st
January 2000 to 31st December 2007. The mortality trends were shown by the rates calculated by
Kaplan-Meier survival estimates expressed per 1000 PYO. Migration rates were computed while the
association between adult mortality and out-migration of household members was assessed using Cox
proportional Hazard model controlling for other individual level and household level factors.
Results: Adult deaths increase by about 9% the chance of a child, male or female, to migrate within or
without the DSA while HIV/AIDS adult deaths increase by a further 19 percentage point the risk of
5
the child to migrate out of the DSA. The results also show that HIV/AIDS adult deaths enhance the
risk of adult female internal migration by 6% (adj. HR 1.06; 95% CI 0.91-1.23, p-value 0.01) but is
not significantly associated with adult male migration. Non-HIV/AIDS adult deaths also enhance the
risk for female internal migration by 5% albeit hardly significantly (adj. HR 1.05; 95% CI 1.0-1.10, pvalue
0.05) but decreases the chance of male internal migration by 13% (adj. HR 0.87; 95% CI 0.81-
0.93, p-value 0.01).
Additionally, HIV/AIDS adult death is strongly associated with out-migration of adults, whatever the
gender. They predispose female out-migration to 19% (adj. HR 1.19; 95% CI 1.09-1.30, p-value
<0.001) and male migration to 30% increased risk (adj. HR 1.30; 95% CI 1.16-1.45, p-value <0.001).
This gender difference is however non-significant (the confidence intervals overlap). Non-HIV/AIDS
adult death has the inverse effect on out-migration, and the gender difference is significant: 18%
increased risk for males (adj. HR 1.18 95% CI 1.14-1.22, p-value <0.001) and 29% for females (adj.
HR 1.29; 95% CI 1.26-1.33, p-value <0.001).
Conclusion: Adult deaths have a positive impact on out-migration, with some variation by gender.
The effect of HIV/AIDS death on out-migration is not very different from other deaths‟ effect.

Identiferoai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:wits/oai:wiredspace.wits.ac.za:10539/9124
Date09 March 2011
CreatorsMurunga, Frederick Wekesah
Source SetsSouth African National ETD Portal
LanguageEnglish
Detected LanguageEnglish
TypeThesis
Formatapplication/pdf

Page generated in 0.0021 seconds