The impact of HIV/AIDS on household dynamics and household ...

hardtofindcurtainUrban and Civil

Nov 16, 2013 (3 years and 8 months ago)

89 views

The impact of HIV/AIDS on household
dynamics and household welfare

in rural northern Malawi

19
th

July, 2010


Sian Floyd, Angela Baschieri, Aulive Msoma, Albert Dube,
Elenaus Mwaiyeghele,

Andreas Jahn, Keith Branson,
Anna Molesworth, Judith Glynn, Neil French

Outline

1.
Background


2.
Research aim and questions


3.
Methods


4.
Results


5.
Interpretation / Conclusions



1) Background


Evidence from prospectively collected longitudinal data
on the impact of adult HIV/AIDS deaths on household
welfare in Southern and East Africa still relatively limited



Overall, evidence suggests:


a)
Largest impact if deceased
was

head of household
or
spouse of head of household
(and possibly smaller impact
on big households)


b)
Effect on household material welfare
(e.g. land, assets,
housing quality)
smaller than anticipated
early in HIV
epidemic

2) Research aim and questions

Aim



To add to the existing evidence, using data from a
demographic surveillance site (DSS) in rural Malawi


Questions


What is the effect of a young adult (age 15
-
59 years old)
HIV/AIDS death in the household on each of:


1.
Rate at which young adults join and leave a household

2.
Household dissolution

3.
Household migration

4.
Household material welfare

Setting

Karonga

district, rural northern
Malawi



Main livelihood = subsistence farming




HIV prevalence in adults 10
-
14%

(2000
-
2008)




DSS
established in south of district,
baseline
census
2002
-
2004




population
~30,000





Sufficient identifying information so can trace
all individuals, plus whole households, over
time




65% of young adult (15
-
59 years old) deaths
due to HIV/AIDS
(2002
-
2006)

3) Methods


From time household enrolled in baseline census, or subsequent
in
-
migration of whole household:





births and deaths


monthly




individual joining and leaving of households
-

once a year




verbal autopsies for
all

deaths
(so can distinguish AIDS vs non
-
AIDS)



Household = unit of analysis



Definition: group of people living together who recognise same head



Socio
-
economic data:

at baseline 2002
-
4; for households that
moved into the DSS area after baseline; follow
-
up survey 2007/8



Housing quality
(construction of roof, walls, floor)


Household sources of income


Household assets (
list of 10 items identified as of value in this setting)


Crops grown



Land owned and cultivated


Food and nutrition security (2007/8 survey only)

3) Methods (continued)



Between 2002
-
2007,
93% of all HIV/AIDS deaths were among 15
-
59 year
olds
, and
85% of HIV/AIDS deaths among 15
-
59 year olds were in 3
household types, with households categorised according to the age and
sex of their members



Thus

all analysis focused on effect of an HIV/AIDS death of a 15
-
59 year old
in these 3 household types
:


1)
Young adults (15
-
59 years old) only
(YA)

2)
Children (<15 years old) + young female adults + young male adults
(CYA)

3)
Children + young female adults + young male adults + older adults
(CYO)



Analysis

restricted to households first seen in DSS prior to 31 August 2005



ensures 2 years of “follow
-
up” for experience of HIV/AIDS deaths, from the time
a household was first seen in the DSS; and


a further 1
-
3 years until the follow
-
up socio
-
economic survey in 2007/8



All analysis compares households in which a young adult (15
-
59 years old)
died of HIV/AIDS during the first 24 months for which the household was
followed up in the DSS, with households in which no young adult died of
HIV/AIDS during this time period

4) Results

Households that experience an HIV/AIDS death also
have higher rate of joining and leaving of young adults


Joining (%)

Leaving (%)

Typology

(when first seen)

1

2+

1

2+

1.
Young adults only

No AIDS death (n=1240)

16

4

14

3

AIDS death (n=21)

24

19

19

5

2.

C + Y female + Y male

No AIDS death (n=3978)

15

5

20

8

AIDS death (n=102)

25

14

29

14

3.

C + Y female + Y male + O

No AIDS death (n=591)

27

15

37

29

AIDS death (n=31)

26

26

35

39

C = Child (<15 years old), Y = Young adult (15
-
59 years), O = older adult (60+ years)

Households that experience an HIV/AIDS death have
higher rate of household dissolution, but not migration


Dissolved
(%)

Migrated

(%)

Typology

(when first seen)

4 years later

4 years later

1.
Young adults only

No AIDS death

14

25

AIDS death (n=21)

63

31

2.

C + Y female + Y male

No AIDS death

7

15

AIDS death (n=102)

31

15

3.

C + Y female + Y male + O

No AIDS death

4

3

AIDS death (n=31)

16

0

YA with AIDS death:
90%
was death of household head or spouse

CYA
with AIDS death:
75%
was death of household head or spouse

CYO with AIDS death:
10%
was death of household head or spouse

Small negative impact of HIV/AIDS death on number of
household possessions*

*
For
young adult only (YA)
households with HIV/AIDS
death




very few still
existed by 2007/8,
so not shown

Median

number

of
household possessions in

2007/8

Count of househol
d possessions


at time first seen

No HIV/AIDS
death

HIV/AIDS
death

1. C + Y female + Y male

≤2

3

2

3
-
4

4

3



5

4

渽㈹㘹

渽㔱

㈮⁃‫ v⁦敭 汥l⬠v 汥l⬠佬H敲 慤畬瑳

≤2

㈮2

2

3
-
4

3

4



5

㐮4

渽㔱5

渽㈲

Apart from household possessions, no or small
effect of young adult HIV/AIDS death on
household material welfare






Measure

of material welfare

Evidence

of negative effect?

Housing quality

No

Food and nutrition security

No

Amount of land owned

No

Amount

of land cultivated

Suggestion of small fall

in
a
mount

of land cultivated

Household possessions, number and
value

Yes, but fall

in number/value was
relatively
small

Number of household income sources

Suggestion of increase, i.e.
diversification, but small effect

Interpretation / Conclusions



A common way in which households respond to an HIV/AIDS death is
with
“replacement” of the lost member and/or departure of surviving
adults



Household dissolution is a common response to an HIV/AIDS death



The apparently relatively small effect of an HIV/AIDS death in the
household on material welfare is partly because households are adapting
in these ways



Our
findings are consistent with those from rural Uganda and Tanzania



Important to be aware
, when designing programmes to help to mitigate
social and economic consequences of young adult HIV/AIDS deaths, that
households are “adjusting”/ responding in these ways



For the future:
impact of young adult HIV/AIDS death on the “next
generation”
-

children’s education and anthropometry
;
and the
effect
of
ART provision












-

Acknowledgements


Funded by Wellcome Trust



All the members of the Continuous Registration System
(CRS) team, who conducted the field work