Do Low Income Households Have Financial Protection from Illness
Costs in South Africa’s Mixed Health Care Financing System?
Health Economics Unit
Vimbayi Mutyambizi
Health Economics Unit
University of Cape Town
South Africa
Prepared for:
International Conference on Health Care Financing in Developing
Countries
CERDI 1 December 2005
OBJECTIVES
Health Economics Unit
• To describe the socio-economic environment in South Africa
• To provide a brief description of how the health system is
structured and is financed.
– Private sector
– Public sector
• To present the results of an analysis of household expenditure
on health care using data from the IES, focusing on
• Total expenditure and OOP expenditure
• Households with medical cover and households without cover
• To present results of an analysis of the determinants of health
service utilisation using the GHS 2003.
Health Economics Unit
Socio-economic Context
COUNTRY DATA
Health Economics Unit
• Population
– 46.5 million (SSA 2004)
• Less Indebted LMIC
• GDP
– R1.277B (169B USD 2003)
• Unemployment
– Official 27%
• Distribution of Income
– Gini coefficient 0.635 (2001) (UNDP 2003)
Monthly Employment Income in South Africa
Health Economics Unit
Monthly income
Formal
Informal
Domestic
Total
None
0.5%
11.4%
0.0%
2.5%
ZAR 1 – ZAR 500
6.6%
39.9%
47.3%
16.4%
ZAR 501 – ZAR 1,000
16.0%
23.5%
40.6%
19.6%
ZAR 1,001 – ZAR 2,500
30.3%
16.9%
12.1%
26.2%
ZAR 2,501 – ZAR 8,000
34.7%
6.7%
0.0%
26.3%
ZAR 8,001+
12.0%
1.5%
0.0%
9.0%
under ZAR 1,000
23.1%
74.8%
87.9%
38.6%
under ZAR 2,500
53.3%
91.8%
100.0%
64.7%
Source: IES 2000.
31 Dec 2000: 1 ZAR = 0.13184 USD
Health Economics Unit
Health System Structure and Financing
HEALTH CARE FINANCING AND STRUCTURE
Health Economics Unit
• Health care expenditure
– 8.7% GDP
• Distribution of Health Expenditure
– Private Sector
– 62% of expenditure
• 16% of the population
– Public Sector
– 38% of expenditure
• 84% of the population
FINANCING FLOWS (2000)
Health Economics Unit
Medical scheme
Contributions (voluntary)
Budget (general taxes)
Medical Schemes
Tax subsidy
(community rated,
open enrolment)
(R4-6 billion)
Prescribed Minimum Benefits
Funded utilisation (FFS)
Public Hospital Basic
Private Hospital fee-for-service
Public Primary Care
Private Primary Care
Source: McLeod 2003
Estimated Health Care Expenditure in South Africa,
2003/04
Health Economics Unit
National
Department
of Health
R2.7 bn
Local
Departments
of Health
R2.5 bn
Other
public
sector, incl.
admin.
R9.2 bn
Local &
provincial
PHC
R5.8 bn
Chronic &
other public
hospitals
R2.1 bn
Provincial
Departments
of Health
R35.6 bn
Household
s
(Out-ofpocket)
R15 bn
Medical
Schemes
R50.2 bn
Firms’
direct
payments
<R1bn
Other private
sector, incl.
admin.
R16.9 bn
Tertiary
hospitals
R7.7 bn
Private
hospitals
R14.4 bn
Regional
hospitals
R8.5 bn
District
hospitals
R8.6 bn
Major flows (considerably >R1
bn)
Medicine
s
R16.7 bn
Specialists
R8.7 bn
GPs
R5.2 bn
Dentist
s
R3 bn
Minor flows (<R1 bn & usually
<R500mill.)
Source: McIntyre et. al 2005
PRIVATE SECTOR COVERAGE AND BENEFITS
Health Economics Unit
• Members – 2.2 million (1999)
• Beneficiaries – 6.9 million (1999)
• Average contribution per beneficiary per month
– R600 (2003)
(Source registrar’s report 2003-2004)
• Benefits
– Prescribed minimum benefits
(exclude PHC)
EXPENDITURE PER BENEFICIARY (1996-1998 )
Health Economics Unit
Gross contributions in ZAR
Total beneficiaries
Total OOP per annum in ZAR(including OTC
1996
1997
1998
17,769,279,879
21,698,195,111
24,284,756,082
6,862,377
6,902,697
6,887,735
3,846,525,401
3,466,419,282
5,348,138,296
2,589
3,143
3,526
561
502
776
3,150
3,646
4,302
purchases)
Contributions/ beneficiary per annum in ZAR
OOP/beneficiary per annum in ZAR
Total expenditure per beneficiary per annum in ZAR
Source: Cornell et al. (2001).
Medical Scheme Membership
According to Expenditure Quintiles
Health Economics Unit
Quintiles of
households’
total annual
expenditure
1 (lowest)
Medical Scheme
Non-member
Member
98.61
1.39
2
98.12
1.88
3
95.47
4.53
4
86.29
13.71
5 (highest)
41.46
58.54
Total
83.99
16.01
Source: IES 2000
Health Economics Unit
Analysis of Household Health Care Expenditure
IES 2000
HEALTH EXP. AS A SHARE OF TOTAL EXP. & INCOME
(total sample)
Health Economics Unit
Relative health share
6%
5%
4%
Income
3%
Exp.
2%
1%
0%
1
2
3
Quintile
4
5
OOP EXP. AS A SHARE OF TOTAL EXP. & INCOME
(IF INCURRED)
Health Economics Unit
4.0%
Relative OOP Exp.
3.5%
3.0%
2.5%
Income
2.0%
Expenditure
1.5%
1.0%
0.5%
0.0%
1
2
3
4
5
Quintiles
OOP EXP. AS A SHARE OF TOTAL INCOME
(all: members and non-members)
Health Economics Unit
Relative OOP Exp.
3.0%
2.5%
2.0%
Members
1.5%
Non-members
1.0%
0.5%
0.0%
1
2
3
Quintiles
4
5
Health Economics Unit
ANALYSIS OF DETERMINANTS OF UTILISATION
GHS 2003
(WORLD BANK)
DETERMINANTS OF HEALTH SEEKING BEHAVIOUR
(Multinomial Logit Model) (1)
RRR 1
Health Economics Unit
Coefficient
Significance
No form al care vs. public
Medical schem e m em bership (no m em ber) 2
2.3409
0.851
0.000
G ender (m ale)
1.1720
0.159
0.019
Ethnic group (African)
Coloured
0.6251
-0.470
0.000
Indian
0.5232
-0.648
0.047
W hite
1.5783
0.456
0.015
Education of household head (no form al education)
Prim ary
0.9293
-0.073
0.392
Secondary
0.9196
-0.084
0.367
Post-secondary technical or diplom a
1.0821
0.079
0.687
University
1.9670
0.677
0.009
Second lowest
0.8663
-0.143
0.140
Medium
1.0181
0.018
0.859
W ealth quintile (low est quintile)
Second highest
1.0966
0.092
0.400
Highest
1.6563
0.505
0.000
Age category (0 - 14 years)
15 - 24 years
2.0589
0.722
0.000
25 - 34 years
1.3812
0.323
0.004
35 - 44 years
1.3250
0.281
0.011
45 - 54 years
1.0717
0.069
0.555
55 years and above
1.1148
0.109
0.282
-1.432
0.000
Constant
1. Relative risk ratio
2. Reference category variables in brackets
Source: Thiede& Mutyambizi (forthcoming)
DETERMINANTS OF HEALTH SEEKING BEHAVIOUR
(Multinomial Logit Model) (2)
RRR
Health Economics Unit
C o e ffic ie n t
S ig n ific a n c e
P r iv a te v s . p u b lic
M e d ic a l s c h e m e m e m b e rs h ip (n o m e m b e r)
G e n d e r (m a le )
1 0 .1 7 0 0
2 .3 1 9
0 .0 0 0
1 .0 9 3 0
0 .0 8 9
0 .1 4 3
E th n ic g ro u p (A fric a n )
C o lo u re d
0 .6 4 5 0
-0 .4 3 9
0 .0 0 0
In d ia n
0 .3 8 4 1
-0 .9 5 7
0 .0 0 0
W h ite
1 .2 6 5 3
0 .2 3 5
0 .1 3 7
P rim a ry
0 .9 4 3 7
-0 .0 5 8
0 .4 9 1
S e c o n d a ry
1 .2 3 5 9
0 .2 1 2
0 .0 1 4
E d u c a tio n o f h o u s e h o ld h e a d (n o fo rm a l e d u c a tio n )
P o s t-s e c o n d a ry te c h n ic a l o r d ip lo m a
2 .0 0 7 8
0 .6 9 7
0 .0 0 0
U n iv e rs ity
2 .8 4 6 3
1 .0 4 6
0 .0 0 0
W e a lth q u in tile (lo w e s t q u in tile )
S e c o n d lo w e s t
1 .1 5 1 4
0 .1 4 1
0 .1 5 6
M e d iu m
1 .6 2 0 2
0 .4 8 3
0 .0 0 0
S e c o n d h ig h e s t
1 .8 5 6 7
0 .6 1 9
0 .0 0 0
H ig h e s t
4 .0 8 9 4
1 .4 0 8
0 .0 0 0
A g e c a te g o ry (0 - 1 4 y e a rs )
1 5 - 2 4 y e a rs
1 .0 1 7 6
0 .0 1 7
0 .8 7 0
2 5 - 3 4 y e a rs
1 .1 4 8 0
0 .1 3 8
0 .1 7 9
3 5 - 4 4 y e a rs
1 .2 4 0 4
0 .2 1 5
0 .0 2 6
4 5 - 5 4 y e a rs
1 .0 3 7 0
0 .0 3 6
0 .7 1 0
5 5 y e a rs a n d a b o v e
1 .0 5 1 3
C o n s ta n t
0 .0 5 0
0 .5 8 3
-1 .5 2 7
0 .0 0 0
W a ld c h i2 (3 6 ) = 1 6 2 0 .2 8
P ro b > c h i2 = 0 .0 0 0 0
P s e u d o R 2 = 0 .1 4 7
Source: Thiede& Mutyambizi (forthcoming)
CONCLUSION
Health Economics Unit
•
Treatment seeking patterns are complex and include both the public/private sector
•
Causalities are not straightforward
– Membership of a medical schemes in different quintiles are different
implications
•
Membership helps lower the risk of variability of expenditure on medical
expenses but does not necessarily mean that they are lower in absolute terms
•
Other issues that need to be taken into account are :quality of care, health needs
and health status factors
•
Membership of schemes in different income groups does not mean needs are
covered
•
Move towards expanding coverage to lower income brackets – these intricacies
need to be understood in great detail in order to ensure that membership does
ensure financial protection from health care costs.
Health Economics Unit
THANK-YOU
REFERENCES
Health Economics Unit
•
Cornell J., J. Goudge, D. McIntyre, and S. Mbatsha. 2001. South African National Health
Accounts: The Private Sector. Pretoria: National Health Accounts Research Team and
National Department of Health.
•
McIntyre D., McLeod H., Thiede M. (2005) Comments on the National Treasury Discussion
Document on the Proposed Tax Reforms Relating to Medical Scheme Contributions
•
McLeod H. 2003. An Introduction to Medical Schemes and Social Health Insurance in South
Africa. Cape Town Centre for Actuarial Research, University of Cape Town.
•
Thiede M., Mutyambizi V. (2005) Role of Private Health Insurance in the South African
Health System: An Impact Evaluation Study: Draft chapter prepared for the book “Private
Health Insurance in Development” (forthcoming)