Download Consumption Patterns and Levels Among Households With HIV PDF

TitleConsumption Patterns and Levels Among Households With HIV
TagsHiv/Aids Poverty Poverty & Homelessness Standard Deviation Poverty Threshold
File Size270.3 KB
Total Pages6
Table of Contents
                            Abstract
Introduction
Methods
	Study sites
	Selection criteria
	Sampling procedure and sample size
	Data collection tool
	Analytical strategy and statistical analysis
Results
Discussion
Acknowledgements
Disclosure statement
Notes
References
                        
Document Text Contents
Page 1

Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=caic20

Download by: [Orta Dogu Teknik Universitesi] Date: 03 December 2015, At: 00:17

AIDS Care
Psychological and Socio-medical Aspects of AIDS/HIV

ISSN: 0954-0121 (Print) 1360-0451 (Online) Journal homepage: http://www.tandfonline.com/loi/caic20

Consumption patterns and levels among
households with HIV positive members and
economic impoverishment due to medical
spending in Pune city, India

Varun Sharma, Divya Krishnaswamy & Sanjeevanee Mulay

To cite this article: Varun Sharma, Divya Krishnaswamy & Sanjeevanee Mulay (2015)
Consumption patterns and levels among households with HIV positive members and
economic impoverishment due to medical spending in Pune city, India, AIDS Care, 27:7,
916-920, DOI: 10.1080/09540121.2015.1015482

To link to this article: http://dx.doi.org/10.1080/09540121.2015.1015482

Published online: 04 Mar 2015.

Submit your article to this journal

Article views: 43

View related articles

View Crossmark data

http://www.tandfonline.com/action/journalInformation?journalCode=caic20
http://www.tandfonline.com/loi/caic20
http://www.tandfonline.com/action/showCitFormats?doi=10.1080/09540121.2015.1015482
http://dx.doi.org/10.1080/09540121.2015.1015482
http://www.tandfonline.com/action/authorSubmission?journalCode=caic20&page=instructions
http://www.tandfonline.com/action/authorSubmission?journalCode=caic20&page=instructions
http://www.tandfonline.com/doi/mlt/10.1080/09540121.2015.1015482
http://www.tandfonline.com/doi/mlt/10.1080/09540121.2015.1015482
http://crossmark.crossref.org/dialog/?doi=10.1080/09540121.2015.1015482&domain=pdf&date_stamp=2015-03-04
http://crossmark.crossref.org/dialog/?doi=10.1080/09540121.2015.1015482&domain=pdf&date_stamp=2015-03-04

Page 2

Consumption patterns and levels among households with HIV positive members and economic
impoverishment due to medical spending in Pune city, India

Varun Sharmaa*, Divya Krishnaswamyb and Sanjeevanee Mulayc

aSchool of Health Systems Studies, Tata Institute of Social Sciences, Mumbai, India; bCentre for Budget and Policy Studies,
Bangalore, India; cPopulation Research Centre, Gokhale Institute of Politics and Economics, Pune, India

(Received 26 August 2014; accepted 27 January 2015)

HIV infection poses a serious threat to the economy of a household. Out of pocket (OOP) health spending can be
prohibitive and can drag households below poverty level. Based on the data collected from a cross-sectional survey of
401 households with HIV+ members in Pune city, India, this paper examines the consumption levels and patterns among
households, and comments on the economic impoverishment resulting from OOP medical spending. Analysis reveals
that households with HIV members spend a major portion of their monthly consumption expenditure on food items.
Medical expenditure constitutes a large portion of their total consumption spending. Expenditure on children’s education
constitutes a minor proportion of total monthly spending. A high proportion of medical expenditure has a bearing on the
economic condition of households with HIV members. Poverty increases by 20% among the studied HIV households
when OOP health spending is adjusted. It increases 18% among male-headed households and 26% among female-
headed households. The results reiterate the need of greater support from the government in terms of accessibility and
affordability of health care to save households with HIV members from economic catastrophe.

Keywords: coping strategies; economic impoverishment; HIV; out of pocket health spending; India

Introduction

HIV-related illnesses result in increased medical expend-
iture and households mostly meet this by out of pocket
(hereafter referred to as OOP) spending (Riyarto et al.,
2010). Frequent illness associated with HIV infection
can undermine livelihoods on one hand and increase
medical expenditure on the other, contributing to impov-
erishment (Barnett, Whiteside, & Desmond, 2001;
Lopera, Einarson, & Ivan Bula, 2011; Whiteside,
2001). HIV/AIDS has the greatest effect on families
with only one employed person (Kebede & Retta, 2004).
Studies reveal that households with HIV+ members
usually have limited financial resources compared to
households without HIV+ members (Oni, Obi, Okorie,
Thabede, & Ordan, 2002). These financial resources are
disproportionately distributed amongst food, non-food
and medical expenditure. Various studies indicate a
change in consumption patterns of households with
HIV+ members and show how resources are diverted
towards health care (Bachmann & Booysen, 2003;
Bechu, 1998; Duraisamy et al., 2006; Oni et al., 2002;
Tibaijuka, 1997). The OOP health spending accen-
tuates financial burden and leads to impoverishment
(Duraisamy et al., 2006). Studies also showcase the
visible difference in how HIV/AIDS impact poor and
non-poor households (Mather et al., 2004; Yamano &
Jayne, 2004).

This paper broadly examines consumption levels and
patterns among households with HIV+ members and
analyses the poverty inducing effect of OOP medical
spending.

Methods

Study sites

Cross-sectional data were collected through interviews1

(conducted between January 2008 and September 2008)
with households with HIV+ members in Pune city of
India.

Selection criteria

Adult (over 18 years) people living with HIV (PLHIV)
were selected based on average household monthly
income (indicator of economic condition; less than/equal
to Rs 2000; Rs 2001–4000; Rs 4001–6000; more than/
equal to Rs 6000) and CD4 cell/µL count (indicator of
disease progression; less than 200 cell/µL; 200–499 cell/
µL; more than/equal to 500 cell/µL) followed by the
Centre for Disease Control.

Sampling procedure and sample size

The list of 1400 individuals registered under the Adher-
ence, Counseling and Treatment project (till December

*Corresponding author. Email: [email protected]

© 2015 Taylor & Francis

AIDS Care, 2015
Vol. 27, No. 7, 916–920, http://dx.doi.org/10.1080/09540121.2015.1015482

D
ow

nl
oa

de
d

by
[

O
rt

a
D

og
u

T
ek

ni
k

U
ni

ve
rs

it
es

i]
a

t
00

:1
7

03
D

ec
em

be
r

20
15



mailto:[email protected]
http://dx.doi.org/10.1080/09540121.2015.1015482

Page 3

2007) at the Positive Living Centre (PLC)2 was filtered
as per the selection criteria and 735 PLHIV shortlisted
adhering to age criteria. Final sample of 401 adult PLHIV
was drawn from that list. Study objectives and instruments
(interview schedules, checklists, consent forms) were
approved by the Network of Maharashtra PLHIV Net-
work of Maharashtra People Positive (NMP+) organisa-
tion. Written consent was obtained from respondents and
their privacy and confidentiality was given utmost
importance during data collection (respondents were
assigned a unique number; names/addresses were not
recorded on the interview schedule).3

Data collection tool

A structured questionnaire (translated into Marathi)4 was
used to collect information regarding respondents’ –
demographic/social status, consumption patterns/levels,
experience of living with HIV.

Analytical strategy and statistical analysis

Bivariate analysis, chi-square test and t test were used
to estimate the statistical significant difference. Poverty
induced effect of OOP medical spending was ex-
plored based on World Bank Institute’s operational
guidelines (O’Donnell, Doorslaer, Wagstaff, & Lindelow,
2008).

Poverty head count gross and net of OOP health
payments, poverty gap gross and net of health payments,
mean poverty gap, normalized poverty gap and gross and
net intensity of poverty were estimated. Poverty line
estimates released by the Planning Commission, India
for Urban Maharashtra were used to estimate poverty
head counts. For Urban Maharashtra, the poverty line
was estimated at Rs 665.90 (2004–2005) per capita per
month (Planning Commission, 2007). The Planning
Commission of India uses the Expert Group Method to
estimate poverty relying on large sample survey data on

Table 1. Background characteristics of households with HIV+ members.

Headship of household with HIV+ members

Characteristics Male Female p valuea

Household headship 253 (63.1) 148 (36.9)
Composition of household 0.001
Less than or equal to two members 24 (9.5) 32 (21.6)
Three to four members 135 (53.4) 81 (54.7)
More than or equal to five members 94 (37.2) 35 (23.6)

Mean household size (SD) 4.23 (±1.41) 3.63 (±1.37)
Household structure 0.311
Joint/extended 82 (32.4) 40 (27.0)
Nuclear 171 (67.6) 108 (73.0)

Religion 0.633
Hindu 218 (86.2) 134 (90.5)
Muslim 11 (4.3) 4 (2.7)
Christian 3 (1.2) 1 (0.7)
Buddhist/Neo-Buddhist 19 (7.5) 9 (6.1)
Any other 2 (0.8) 0 (0.0)

Caste 0.563
SC 83 (32.8) 46 (31.1)
ST 35 (13.8) 20 (13.5)
OBC 27 (10.7) 23 (15.5)
Others 108 (42.7) 59 (39.9)

Monthly income (INR) <0.001
<=2000 122 (48.2) 100 (67.6)
2001–4000 78 (30.8) 36 (24.3)
4001–6000 28 (11.1) 9 (6.1)
>6000 25 (9.9) 3 (2.0)

HIV status of head of the household <0.001
Positive 182 (71.9) 126 (85.1)
Negative 70 (27.7) 22 (14.9)
Do not know 1 (0.4) 0 (0.0)

Note: Figures in bracket are percentages of n.
ap value is estimated based on chi-square test.
SC, schedule caste; ST, schedule tribe; OBC, other backward classes; SD, standard deviation.

AIDS Care 917

D
ow

nl
oa

de
d

by
[

O
rt

a
D

og
u

T
ek

ni
k

U
ni

ve
rs

it
es

i]
a

t
00

:1
7

03
D

ec
em

be
r

20
15

Page 4

household consumer expenditure conducted by the
National Sample Survey Organization.

Results

Mean household size was 4.01 and nearly 70% were
nuclear. Almost 37% of households were female-headed
(Table 1).

Majority of female-headed households were earning
less than Rs 2000 (67.6%); 85% heads of female-headed
households were HIV+. On the other hand the same
figures for male-headed households stood at 48.2% and
71.9%, respectively.

Consumption patterns showed stratification of
expenditure incurred on food and non-food items, while
consumption levels showed quantification of expendi-
ture; these revealed the welfare aspect of a household.

Average total expenditure per capita per month among
male-headed households (Rs 906.12) was higher than
female-headed households (Rs 765.87; refer Table 2).

Per capita medical expenditure among male-headed
households was more than one and a half times that for
female-headed households. The average per capita total
food expenditure between the two groups of household
was statistically significant.

The share of total expenditure incurred on non-food
items depicts the economic prosperity of a household
(refer Table 3).

Households with HIV+ members spend a consider-
able percentage of their total household expenditure on
food items (52.11%). Within non-food-related expendi-
ture, medical expenditure (9.62%) constitutes a large

proportion. Expenditure on education of children
(2.79%) constitutes a tiny slice of the total expenditure
of a household.

A high proportion of medical expenditure is relevant
to the economic condition of households with HIV+
members. In the study, when OOP health spending is
adjusted, poverty increases by 20%. It increases 18%
among male-headed and 26% among female-headed
households. The poverty gap indicates the amount re-
quired to be transferred in order to bring the expenditure

Table 2. Average per capita monthly expenditure on major food and non-food items (in INR).

Items

Headship of household with HIV+ members

Male Female t statistics p valuea

Cereal 105.52 102.46 0.560 0.576
Pulses 28.03 24.22* 1.853 0.062
Other food items 286.54 242.93** 2.803 0.005
Total food 418.76 368.63** 2.681 0.008
Fuel 63.96 66.03 −0.360 0.719
Electricity 40.33 35.74 1.247 0.214
Rent 61.54 63.69 −0.168 0.867
Clothing and footwear 37.68 23.67** 4.287 <0.001
Durable goods 4.41 2.22 0.577 0.564
Education 21.12 30.82 −1.643 0.101
Medical expenditure 152.88 93.62* 1.783 0.075
Other non-food 105.45 81.46** 3.216 0.001
Total non-food 487.36 397.24** 2.037 0.042
Total 906.12 765.87** 2.497 0.013

ap value is estimated based on student t test to estimate the statistical difference of average per capita monthly expenditure on various food and
non-food items between male-headed and female-headed households with HIV+ members.
INR, Indian Rupees.
**95% CI; *90% CI.

Table 3. Share of expenditure on some major items (in
percentages).

Headship of household with
HIV+ members

Items Male Female Total

Cereal 13.66 15.21 14.23
Pulses 3.52 3.50 3.51
Other food items 34.71 34.18 34.51
Total food 51.78 52.68 52.11
Fuel 8.32 9.11 8.61
Electricity 4.97 4.71 4.87
Rent 5.26 6.78 5.82
Clothing and footwear 4.41 3.32 4.01
Durable goods 0.28 0.27 0.27
Education 2.33 3.59 2.79
Medical expenditure 10.24 8.54 9.62
Other non-food 12.41 11.00 11.89
Total non-food 48.22 47.32 47.89
Total 100.00 100.00 100.00

918 V. Sharma et al.

D
ow

nl
oa

de
d

by
[

O
rt

a
D

og
u

T
ek

ni
k

U
ni

ve
rs

it
es

i]
a

t
00

:1
7

03
D

ec
em

be
r

20
15

Page 5

of households up to the poverty line; from Table 3 we see
that an additional Rs 59 per capita per month is the
increase in poverty gap due to OOP health payments
(Table 4).

Female-headed households were more vulnerable to
economic impoverishment than male-headed households.
Table 3 shows Rs 67 and Rs 55 per capita monthly needs
to be transferred to female-headed and male-headed
households, respectively, to bring their expenditure up
to the poverty line.

Discussion

Theoretically and empirically studies have found that
poor people spend more of their total expenditure on
food-based items (Moneta & Chai, 2013). The study
found that households with HIV+ members spent a
major share of their total household expenditure on food
and medical care, indicating that they belong to the
lower economic strata. It highlights their economic
conditions and accentuates the impoverishment due to
OOP health spending. Share of OOP health payments
among households with HIV+ members is higher than
their counterparts (National Sample Survey Organisa-
tion, 2008; Pradhan, Sundar, & Singh, 2006) making
them susceptible to a Medical Poverty Trap. Education
of children constituting a meagre proportion of the total
consumption expenditure in households with HIV+
members is indicative of how HIV/AIDS affects chil-
dren’s education and hinders formation of future human
capital (Pradhan et al., 2006).

HIV infection affects the economic condition of a
household; hence policies need to complement existing
strategies for PLHIV. In India, the National Commission
on Macroeconomics and Health reported that OOP
expenditure on HIV+ members and individuals on anti-
retroviral treatment is significantly high over a six-month
reference period (Rs 6000 and nearly Rs 18,150, respect-
ively) and roughly 40–70% of such expenditures are
financed by borrowing (NCMH, 2005). Various studies
have reported that OOP expenditure on treatment and
services is high in India and other countries (Duraisamy

et al., 2006; Kumarasamy, Venkatesh, Mayer, & Freed-
berg, 2007; Riyarto et al., 2010).

Economic burden of HIV-related illness at the house-
hold level is embodied in OOP health spending resulting
in an increase in poverty among households with HIV+
members. Illness-related medical costs affect below pov-
erty line households disproportionately (Kamolratanakul
et al., 1999). Thus, the economic burden of illness
associated with HIV infection is a serious concern for
health policy-makers. Findings suggest that when OOP
expenditure is adjusted, poverty increases by 20% in the
studied households. Gupta (2009) found that when OOP
spending is adjusted in the Indian context, poverty
increases by 3.6% and 2.9% in rural and urban areas,
respectively.

However, the results of the study need interpreting
with certain caveats: Responses were based on a reference
period (last month before the survey) and may have recall
bias. Ideally, longitudinal data are required to examine
disruption in consumption path due to health shocks and
assess impact on the economic well-being of a household/
individual. Results are also not comparable in nature due
to the absence of a control group of similar socio-
economic nature.

Overall the study observed that medical expenditure
of a household with HIV+ members constitutes a fairly
large proportion of the household budget. In the absence
of social support schemes and health insurance for
PLHIV, households spend OOP; this is enough to drag
some households below the poverty line. Households with
HIV+ members need much greater support from the
government in terms of accessibility and affordability of
health care to avoid economic catastrophe. The policy
response needs to comprise a suitable balance of targeted
health insurance intervention and social security schemes
to protect households with HIV+ members from impover-
ishment due to HIV-related illness.

Acknowledgements

The authors would like to thank the NMP+ for allowing us
to conduct this study at their PLC and the Jawaharlal Nehru

Table 4. Estimation of poverty with and without health expenditure.

Headship of household with HIV+ members

S. No. Variables Male Female All

A Headcount ratio (%) 50.7 53.4 51.6
B Health expenditure adjusted head count ratio (%) 68.3 79.3 72.0
C Poverty gap (Rs) – mean 97.3 113.3 102.6
D Poverty gap health expenditure adjusted (Rs) – mean 152.4 180.0 161.6
E Percentage increase in poverty (%) 17.7 25.9 20.4
F Increase in poverty gap (Rs) 55.1 66.8 59.0

AIDS Care 919

D
ow

nl
oa

de
d

by
[

O
rt

a
D

og
u

T
ek

ni
k

U
ni

ve
rs

it
es

i]
a

t
00

:1
7

03
D

ec
em

be
r

20
15

Similer Documents