INTERNATIONAL COMPARISON OF SOUTH AFRICAN PRIVATE HOSPITAL
PRICE LEVELS
Francesca Colombo, Head of Health Division, OECD
Luca Lorenzoni, Economist, OECD
Tomas Roubal, Health Economist, WHO South Africa
Sarah L Barber, WHO Representative, South Africa
World Health
Organization
1
Why are South African private hospital prices a public policy concern?
South Africa spends a higher share of its total health expenditures on private voluntary health insurance (41.8%) than any country globally – >6x OECD average (6.3%) – even though it only serves 16% of the population. This is equivalent to 3.7% of South Africa’s GDP.
Source: OECD health data 2015, data for 2013 or latest year available Figure 1. Private health insurance as a share of total current health expenditures (%), 2013
2
41
,8
34
,9
20
,9
20
1
5,2
1
3,9
1
3,2
1
0
9,3
8
,9
7,2
6
,5
6,3
5
,9
5,3
5
4
,9
4,6
4
,4
4,2
4
,1
4
3,5
3
,5
3,2
3
2
,7
2,4
2
,1
2,1
1
,9
1,8
1
,3
1,3
1
0
,8
0,6
0
,2
0,2
0
5
10
15
20
25
30
35
40
45
Sou
th A
fric
aU
nit
ed S
tate
sB
razi
lC
hile
Slo
ven
iaFr
ance
Can
ada
Co
lom
bia
Ger
man
yA
ust
ralia
Swit
zerl
and
Ko
rea
OEC
D a
vera
geN
eth
erla
nd
sP
ort
uga
lA
ust
ria
New
Zea
lan
dLu
xem
bo
urg
Spai
nM
exic
oB
elg
ium
Po
lan
dU
nit
ed K
ingd
om
Ch
ina
Ind
iaG
reec
eH
un
gary
Jap
anFi
nla
nd
Ru
ssia
Den
mar
kIn
do
nes
iaC
ost
a R
ica
Latv
iaIt
aly
Lith
uan
iaSw
eden
Cze
ch R
epu
blic
Esto
nia
Private voluntary health insurance (PVHI) in OECD countries tends to fill in the gap above public cover- unlike in South Africa where
medical schemes mainly finance an alternative to the public sector.
Source: OECD, Health at a Glance 2015; WHO, Global Health Expenditure Database.
Countries with private voluntary health insurance that offer duplicative cover spend much less on PVHI (3.4-13.4% of total health expenditure).
Country PVHI as % of Total health
spending % pop covered by PVHI Type of coverage
South Africa 41.8 16 Duplicate, supplementary
USA 34.9 52.5 Primary
Chile 20.0 18.3 Primary
Ireland 13.4 44.6 Duplicate
France 13.9 95 Complementary Canada 13.2 67 Supplementary
Israel 10.7 82.9 Supplementary
Germany 9.3 11.0/22.0 Primary/complementary
Australia 8.9 54.9/47 Supplementary/duplicate
New Zealand 4.9 29.7 Duplicate
Portugal 5.3 21.1 Duplicate
Spain 4.4 12.5 Duplicate UK 3.4 10.6 Duplicate
Italy 1.0 15 Complementary
3
OECD countries have measures in place to cap, set or benchmark prices that RSA lacks
• OECD: public sector tends to have some form of price setting for specialist and hospital services, and this provides benchmarks for the private sector. South Africa lacks these price setting measures.
• For OECD countries, prices in the private sector are set using these benchmarks. This has been used as a means to contract/purchase private services to expand access.
• Without these measures in South Africa, negotiations between a handful of medical scheme administrators and private hospitals and specialists determine how a large section of the country’s funds for health are spent (3.7% GDP).
4
Objectives of the study
• Compare private hospital price levels in South Africa with OECD countries.
• Examine correlations between a country’s income and hospital prices.
• Estimate affordability of hospital services in comparison with general goods and services.
• Using South African data, assess the factors influencing private hospital prices: volume and components of the price (hospitals, specialists, pathology, radiology)
World Health
Organization
5
Methods (1)
• Cross-country price comparison requires standard units, definitions, and ways of measurement that can be applied uniformly across all countries to ensure comparison of “Like with Like.”
• Approach applied and validated in OECD countries: OECD Eurostat Purchasing Power Parities (PPPs) project. The PPP project identified 28 case types as a sample of hospital services that are the most common services appropriate for international comparison.
– 7 medical services
– 21 surgical services
World Health
Organization
6
Methods (2)
• These are services that are typically covered under the public basket, paid for through government contracting – regardless of whether they are delivered through public or privately-owned facilities.
• Prices reported represent the total price paid for each case type, including capital
• We also collected information on the number of admissions and average length of stay per case type
• South Africa collected information on the following price components: hospitals, specialists, pathology, radiology.
World Health
Organization
7
Methods (3)
South Africa
• Data from several large medical schemes from 2011-2013 – representing 59.4 % of members and total of 625,940 cases
OECD countries
• Used existing data already collected using same methodology for comparison (Purchasing Power Parities project) for representative sample
• 20 countries: Austria, Czech Republic, Estonia, Finland, France, Germany, Hungary, Iceland, Ireland, Italy, Luxembourg, Netherlands, Norway, Poland, Portugal, Slovenia, Spain, Sweden, Switzerland and United Kingdom
• Lower income subset of 7 countries with GDP per capita level closer to South Africa: Czech Republic, Estonia, Hungary, Poland, Portugal, Spain and Slovenia
World Health
Organization
8
Methods (4)
South Africa – data from medical schemes
• Insight Actuaries South Africa and other experts from medical scheme administrators worked with OECD to adapt the methodology
• Methodology was tested on the data from Government Employees Medical Scheme (GEMS)
• Several medical schemes shared their data including GEMS, Medscheme Holdings (Pty) Ltd, Bonitas Medical Fund
• Preliminary results were presented to and discussed with data providers and some adjustments were made to improve on the accuracy of the findings
• These organizations generously gave their time and expertise to this study, and openly shared their data. We thank them!
World Health
Organization
9
Methods (5)
Prices expressed in Purchasing Power Parities (PPPs)
• Convert different currencies to a common currency and uniform price level
• Equalize purchasing power across countries
• Enables cross-country comparison of prices
World Health
Organization
10
Methods (6)
Results are reported as comparative price levels
• Average for comparison group is calculated as the geometric mean of the comparative price levels across all countries included in the comparison group, and is then set equal to 100.
• Each country’s comparative price level is then expressed in relation to the mean of 100. Results should be interpreted looking at the relative positions of countries rather than looking at absolute levels.
Exchange rates fluctuations are captured in both the CPI and hospital prices.
World Health
Organization
11
Presentation of
Results
South African prices by case type
• Prices in Rand by case type, 2011-13
• Increases in prices over time
Cross country comparison
• Correlations between hospital price levels and income
• Comparison of hospital price levels across countries
• Affordability of hospital prices relative to other goods and services
What is driving South African prices?
• Length of stay and admissions
• Components of the price: hospitals, specialist, pathology
World Health
Organization
12
Average price (in Rand) and average annual rate of change by case type: 7 medical cases studied
M01 Acute Myocardial Infarction 56,726 59,038 64,308 6.5
M02 Angina Pectoris 25,729 27,562 29,445 7.0
M03 Cholelitiasis 11,954 13,788 14,639 10.7
M04 Heart Failure 25,325 27,168 28,959 6.9
M05 Malignant Neoplasm of Bronchus and Lung 24,147 25,651 26,412 4.6
M06 Normal Deliveries 16,251 17,056 17,954 5.1
M07 Pneumonia 17,002 18,359 19,731 7.7
S01 Appendectomy 23,448 25,725 27,836 9.0
S02 Caesarean section 27,802 29,915 31,912 7.1
S03 Cholecystectomy 34,390 36,753 39,071 6.6
S04 Colorectal resection 95,850 99,863 103,543 3.9
S05 Coronary artery bypass graft 220,344 237,497 253,901 7.3
S06 Discectomy 80,847 83,851 87,977 4.3
S07 Endarterectomy 69,901 73,993 82,917 8.9
S08 Hip replacement: total and partial 111,210 116,976 120,758 4.2
S09 Hysterectomy: abdominal or vaginal 29,108 31,616 34,432 8.8
S10 Knee replacement 105,536 110,801 114,044 4.0
S11 Mastectomy 31,427 33,944 37,437 9.1
S12 Open prostatectomy 68,556 73,683 76,645 5.7
S13 Percutaneous transluminal coronary angioplasty 88,827 93,466 101,689 7.0
S14 Peripheral vascular bypass 95,008 94,889 106,260 5.8
S15 Repair of inguinal hernia 20,428 21,704 23,857 8.1
S16 Thyroidectomy 29,204 31,907 34,623 8.9
S17 Transurethral resection of prostate 29,176 31,500 33,196 6.7
S18 Arthroscopic excision of meniscus of knee 21,002 22,974 24,748 8.6
S19 Lens and cataract procedures 19,367 20,464 21,633 5.7
S20 Litigation and stripping of varicose veins - lower limb 21,060 22,122 24,275 7.4
S21 Tonsillectomy and/or adenoidectomy 10,044 10,853 11,442 6.7
S18 Arthroscopic excision of meniscus of knee 17,279 18,152 19,497 6.2
S19 Lens and cataract procedures 18,520 19,299 20,115 4.2
S20 Litigation and stripping of varicose veins - lower limb 12,277 14,518 16,144 14.7
S21 Tonsillectomy and/or adenoidectomy 7,957 8,341 8,822 5.3
Annual rate of
change (%)
Year
Case type 2011 2012 2013
13
Average price (in Rand) and average annual rate of change by case type: surgical cases studied
M01 Acute Myocardial Infarction 56,726 59,038 64,308 6.5
M02 Angina Pectoris 25,729 27,562 29,445 7.0
M03 Cholelitiasis 11,954 13,788 14,639 10.7
M04 Heart Failure 25,325 27,168 28,959 6.9
M05 Malignant Neoplasm of Bronchus and Lung 24,147 25,651 26,412 4.6
M06 Normal Deliveries 16,251 17,056 17,954 5.1
M07 Pneumonia 17,002 18,359 19,731 7.7
S01 Appendectomy 23,448 25,725 27,836 9.0
S02 Caesarean section 27,802 29,915 31,912 7.1
S03 Cholecystectomy 34,390 36,753 39,071 6.6
S04 Colorectal resection 95,850 99,863 103,543 3.9
S05 Coronary artery bypass graft 220,344 237,497 253,901 7.3
S06 Discectomy 80,847 83,851 87,977 4.3
S07 Endarterectomy 69,901 73,993 82,917 8.9
S08 Hip replacement: total and partial 111,210 116,976 120,758 4.2
S09 Hysterectomy: abdominal or vaginal 29,108 31,616 34,432 8.8
S10 Knee replacement 105,536 110,801 114,044 4.0
S11 Mastectomy 31,427 33,944 37,437 9.1
S12 Open prostatectomy 68,556 73,683 76,645 5.7
S13 Percutaneous transluminal coronary angioplasty 88,827 93,466 101,689 7.0
S14 Peripheral vascular bypass 95,008 94,889 106,260 5.8
S15 Repair of inguinal hernia 20,428 21,704 23,857 8.1
S16 Thyroidectomy 29,204 31,907 34,623 8.9
S17 Transurethral resection of prostate 29,176 31,500 33,196 6.7
S18 Arthroscopic excision of meniscus of knee 21,002 22,974 24,748 8.6
S19 Lens and cataract procedures 19,367 20,464 21,633 5.7
S20 Litigation and stripping of varicose veins - lower limb 21,060 22,122 24,275 7.4
S21 Tonsillectomy and/or adenoidectomy 10,044 10,853 11,442 6.7
S18 Arthroscopic excision of meniscus of knee 17,279 18,152 19,497 6.2
S19 Lens and cataract procedures 18,520 19,299 20,115 4.2
S20 Litigation and stripping of varicose veins - lower limb 12,277 14,518 16,144 14.7
S21 Tonsillectomy and/or adenoidectomy 7,957 8,341 8,822 5.3
Annual rate of
change (%)
Year
Case type 2011 2012 2013
M01 Acute Myocardial Infarction 56,726 59,038 64,308 6.5
M02 Angina Pectoris 25,729 27,562 29,445 7.0
M03 Cholelitiasis 11,954 13,788 14,639 10.7
M04 Heart Failure 25,325 27,168 28,959 6.9
M05 Malignant Neoplasm of Bronchus and Lung 24,147 25,651 26,412 4.6
M06 Normal Deliveries 16,251 17,056 17,954 5.1
M07 Pneumonia 17,002 18,359 19,731 7.7
S01 Appendectomy 23,448 25,725 27,836 9.0
S02 Caesarean section 27,802 29,915 31,912 7.1
S03 Cholecystectomy 34,390 36,753 39,071 6.6
S04 Colorectal resection 95,850 99,863 103,543 3.9
S05 Coronary artery bypass graft 220,344 237,497 253,901 7.3
S06 Discectomy 80,847 83,851 87,977 4.3
S07 Endarterectomy 69,901 73,993 82,917 8.9
S08 Hip replacement: total and partial 111,210 116,976 120,758 4.2
S09 Hysterectomy: abdominal or vaginal 29,108 31,616 34,432 8.8
S10 Knee replacement 105,536 110,801 114,044 4.0
S11 Mastectomy 31,427 33,944 37,437 9.1
S12 Open prostatectomy 68,556 73,683 76,645 5.7
S13 Percutaneous transluminal coronary angioplasty 88,827 93,466 101,689 7.0
S14 Peripheral vascular bypass 95,008 94,889 106,260 5.8
S15 Repair of inguinal hernia 20,428 21,704 23,857 8.1
S16 Thyroidectomy 29,204 31,907 34,623 8.9
S17 Transurethral resection of prostate 29,176 31,500 33,196 6.7
S18 Arthroscopic excision of meniscus of knee 21,002 22,974 24,748 8.6
S19 Lens and cataract procedures 19,367 20,464 21,633 5.7
S20 Litigation and stripping of varicose veins - lower limb 21,060 22,122 24,275 7.4
S21 Tonsillectomy and/or adenoidectomy 10,044 10,853 11,442 6.7
S18 Arthroscopic excision of meniscus of knee 17,279 18,152 19,497 6.2
S19 Lens and cataract procedures 18,520 19,299 20,115 4.2
S20 Litigation and stripping of varicose veins - lower limb 12,277 14,518 16,144 14.7
S21 Tonsillectomy and/or adenoidectomy 7,957 8,341 8,822 5.3
Annual rate of
change (%)
Year
Case type 2011 2012 2013
14
7,1
6,9
6,8
7
6
6,2
0 1 2 3 4 5 6 7 8
Medical
Surgical
Private hospitals, average
% change in price
2013/2012
2012/2011
CPI index
Private hospital prices in South Africa increased over time (6.2-6.8%), which is higher than the price
increase for other goods/services (CPI: 5.6% and 5.7%)
15
Strong correlation (r=0.82) between country’s income (GDP) and hospital prices – but South Africa is the outlier
Each country’s comparative price level is expressed relative to the mean of 100.
South Africa’s prices are on par with France, UK, Germany – countries with much higher income levels. 16
South Africa has the lowest GDP per capita and a price level for private hospitals comparable to the average
observed across OECD countries
17
Private hospitals are charging high-prices for their services in an environment where other goods and services are delivered at lower cost.
South African economy wide price level represents 53% of OECD average in 2013: but private hospital prices are
comparable or higher than OECD averages Consistent over three years
GDPpercapita(inUS$PPPs)
OECDaverage 40,169 40,700 41,224
SouthAfricaaverage 12,144 12,555 12,891
18
100 = OECD mean
For lower GDP subset: South African economy wide price level is 74% of the subset average; but private hospital
prices approximately double the subset average
ComparativePriceandGDPLevels
SouthAfricasample2011 2012 2013
Medicalservices 174 168 161
Surgicalservices 207 201 211
Total 195 189 192
GrossDomestic
Product85 86 74
GDPpercapita(inUS$PPPs)
Subsetof7lowerincomeOECD
countriesonaverage26,404 26,702 27,323
SouthAfricaaverage 12,144 12,555 12,891
19
100=OECD subset mean
South Africa private hospital prices rank as least affordable in comparison with other countries
20
…because there is the largest difference between hospital price levels and general price levels – i.e., prices of food, clothing and other common goods
Within South Africa, private hospital prices are likely to be expensive for 90% of South Africans
- even for people with higher incomes
Hospital comparative price levels and household consumption expenditure pc (US$PPP), including South Africa’s high income populations (expenditure deciles 7-10)
21
What is driving prices? • Volume
– Length of stay
– Number of Admissions
• Unit price components dedicated to: – Specialists
– Pathology
– Hospitals
– Radiology
• Other factors – Organization of care (supply)
– Patient preferences (demand)
World Health
Organization
22
Average length of stay in private hospitals in South Africa (3.3 days) is lower
than OECD average (4.7 days)
23
8-41% lower average length of stay across all medical and surgical cases studied
3,5
2,9
4,4
2,3
5,0
4,7
2,7
6,0
4,7
6,9
3,3
7,1
6,4
3,6
41% 39% 37%
31% 29% 27%
24%
0%5%10%15%20%25%30%35%40%45%
-
1
2
3
4
5
6
7
Ave
rage
len
gth
of
stay
in d
ays
South Africa OECD average difference in % 24
No changes in overall admission rates on average in the South African sample between 2011-13 … Admissions increased in line with the increase in medical scheme membership by 12% between 2011-13.
42,1 38,2
42,4
38,0
42,2
37,6
0
5
10
15
20
25
30
35
40
45
Rate per 1000 Bene-
ficiaries
2011
2012
2013
25
But the structure of admissions changed – with increases mostly in selected surgical cases (2011-13).
. Surgical inpatient cases increased by 16%
All but 2 medical case admissions increased by 25% Large increases in hip and knee replacements (31% and 53%)
26
5 7
96
3 2
86
2 5
74
2 8
98
6 1
29
3 5
31
3 3
96
4 4
55
5,7%
7,5% 31,9%
53,7%
0%
10%
20%
30%
40%
50%
0
1000
2000
3000
4000
5000
6000
Appendectomy Repair of inguinalhernia
Hip replacement Knee replacement
2011 2013 2011-2013 increase increase in membership
Relatively high admission rates for some services with standardized age/sex rates in comparison with
OECD countries
Hospital utilisation rates per 100,000, South African sample and OECD countries:
Hysterectomy
27
0%
10%
20%
30%
40%
50%
AnginaPectoris
AcuteMyocardialInfarction
Cholelitiasis
Heart Failure
Neoplasm ofbronchus
Pneumonia
Specialists
Radiology
Hospital
Hospital component accounts for the largest share of prices for both medical (40-52%) and surgical
cases (49-62%)
• Pathology: 21 – 35 % • Specialists: 13 – 19 % 28
10%
12%
14%
16%
18%
20%
Acute MyocardialInfarction
Angina Pectoris
Cholelitiasis
Heart Failure
Malignant Neoplasmof Bronch
Pneumonia
2011
2012
2013
But the share of the price devoted to specialists increased over time, and comprises 14%–29%
Changes in share of price for specialist fees, medical cases, 2011-2013
29
In summary… Strengths of the study
• Transparent and reproducible methodology, which was tested and validated on medical scheme data
• Hospital cases comparable, clearly defined and representative of hospital production
• Price estimate based on payment from purchaser – In OECD countries public purchaser’s price was used as
they pay the same to public and private hospitals – This enabled us to compare private hospitals in South
Africa with public/private hospitals in OECD countries
• All costs included (specialists, nursing, pharma, pathology, investment)
• When comparing prices, PPPs convert currency into common currency and eliminate differences in price levels between countries.
World Health
Organization
30
Limitations
• Comparison done with countries for which data are available from the OECD-EUROSTAT project for 2011-2013
• No data on cost components of hospitals and specialists for comparator countries
• Cannot determine what is behind large share of price devoted to hospitals
• No data on individual medical scheme prices
• No information on the market structure/power and its impact on prices
World Health
Organization
31
Conclusion 1:
Prices are high and increasing.
• South Africa’s private hospital prices are expensive relative to what could be predicted given the country’s wealth.
• Prices are increasing over time above the rate of inflation, or at a higher rate than other goods and services in the economy.
World Health
Organization
32
• South Africa data demonstrate the greatest difference between private hospital price level and general prices level among all countries in the sample.
• Thus South Africa’s private hospital price level is ranked as the least affordable among all countries analysed.
• Prices are also likely to be expensive for 90% of South Africans.
World Health
Organization
Conclusion 2: Prices are not affordable for most South Africans.
33
Low ALOS are likely related to the financing and organization of care
--weak/no controls over admissions given that specialists work independently in hospitals.
-- including cost control measures by medical scheme administrators, i.e., pre-authorized length of stay
Impact on quality of care and health outcomes are a concern and should be evaluated in the future.
World Health
Organization
Conclusion 3: Unusually low lengths of stay probably result from
cost-control efforts.
34
• Planned surgical procedures, i.e., hip and knee replacements increased by 31% and 53% over a very short time (2011-13).
• These changes cannot be fully explained by the 12% increase in medical scheme membership (2011-13)
• They also cannot be fully explained by aging population or other external factors.
• They are likely driven by supplier induced demand.
World Health
Organization Conclusion 4: Large increases in high volume surgical procedures
cannot be fully explained by changes in
membership.
35
• Hospitals are the main component of prices (40-62%) for both medical and surgical cases.
• The increase in the prices observed in this study however, were driven by the increase specialists fees.
• This has implications for the health system as a whole and access to services.
World Health
Organization
Conclusion 5: Hospital share is the main component of the price,
but specialists fees are driving the increase.
36
• Given the magnitude of private voluntary health insurance spending (3.7% GDP), market interactions between medical schemes and private health care providers spills over to the whole health system.
• Prices set in the private sector set labour market benchmarks that specialists face in choosing between working in public/private facility.
• High private prices restrict the ability of the government to use private services to achieve universal health coverage under National Health Insurance (NHI).
• Increases in prices are passed on as increases in premiums to members and employers – which can then lead to individuals bearing the burden (i.e., higher co-payments, reduced benefits)
World Health
Organization Conclusion 6: The high prices in private health sector spill over
to the health system and economy as a whole.
37
• High and increasing prices suggest that current methods of controlling private hospital prices are not effective in curbing price inflation.
• Other OECD countries have measures to prescribe, cap or benchmark prices that South Africa lacks.
This study suggests that efforts to control prices while ensuring accessibility and quality are needed, which could help individual South Africans and the country at large get more value from their
considerable spending on health care.
World Health
Organization Conclusion 7: The current ways in which the private sector
controls prices are not effective.
38
Thank you!
World Health
Organization
39
International Comparison of South African Private Hospitals Price Levels; OECD Working Paper No. 85. DELSA/HEA/WD/HWP(2015)85. Available at http://www.oecd.org/health/workingpapers Contact: WHO Representative Office in South Africa [email protected]