Habit or hiatus - Institute and Faculty of Actuaries · Source: "Overtreated: Why too much medicine...

© 2010 The Actuarial Profession www.actuaries.org.uk

Health & care conference 2011Sarah Bennett & Joanne Buckle

Global PMI utilisation patterns

Habit or hiatus19th May 2011

Global PMI utilisation patterns: Habit or hiatus

1© 2010 The Actuarial Profession www.actuaries.org.uk

0% 10% 20% 30% 40% 50% 60%

Malaysia

India Commercial

India micro

Hong Kong

Brazil

Mexico

Saudi

Lebanon

USA

Netherlands

UK Commercial

UK HES data

UK Seniors

Asia

Latin

A

mer

ica

Mid

dle

Eas

tU

SE

urop

e

Percentage of hospital admissions per population per year

Hospital admissions Day case admissions

Agenda

Background & contextSupply side issuesUtilisation comparisonsConclusionsQuestions/DiscussionAppendix – Country summaries


Background & Context

• Economic regression model of healthcare utilisation at a macro level– Income– Education

– Which are in themselves linked

– Supply/infrastructure– Demographics/health status

• Trend in utilisation tends to depend on – Inflation– GDP growth (more real GDP growth = higher propensity to spend

on healthcare)– Technology advances



• PMI utilisation patterns vary widely across (and within) countries

• Various factors/hypotheses– Structure of system/interaction with public system– Macroeconomic environment– Cultural issues– Supply & reimbursement of medical services– Benefits available under PMI– Political policy– Health status/demographics of underlying population– Voluntary versus compulsory – anti-selection issues


Background & ContextSpend on health care and the impact on quality


0

5

10

15

20

25

30

35

40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19Health care spend

Qua

lity

of h

ealth

car

e

After a certain point, increased health spend does not improve quality of health care

Up to a certain point, quality of health care improves rapidly as health care spend increases

As health care spend continues to spiral upwards, increased health spend actually worsens the quality of health care delivered

Background & ContextSpend on health care and the impact on quality


0

5

10

15

20

25

30

35

40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19Health care spend

Qua

lity

of h

ealth

car

e

China

US

India

Background & ContextInefficiencies in the US: very high users

• In the US, one fifth to one third of health care dollars are spent on care that does nothing to improve health

• The number of deaths due to unnecessary care is 30,000 per year in the US (+ ~100,000 per year due to medical error)

• Why so much unnecessary care?1. Doctors lack evidence to know which treatments and drugs are most effective

2. Doctors lack training to interpret the quality of evidence that’s available

3. They overtreat patients out of a desire to help, even if they don’t know it’s right

4. Malpractice fears drive defensive medicine

5. Medical custom varies remarkably from region to region in the US

6. Most doctors are reimbursed for how much health care they deliver, rather than quality

7. Member induced demand, for example, Generation Rx


Source: "Overtreated: Why too much medicine is making us sicker and poorer" – Sharon Brownlee

• Baby boomers • Generation X• Generation Y• Generation Rx


Background & ContextInefficiencies in the US: Prescription drugs

Source: "Overtreated: Why too much medicine is making us sicker and poorer" – Sharon Brownlee


• Countries we will look at:– Europe: UK, Germany, Netherlands– USA– Asia: India, Malaysia, Hong Kong– Middle East: Saudi Arabia, Lebanon, Israel– Latin America: Brazil, Mexico


Supply side statistics summaryNumber of hospital beds per 10,000 population

147

89

61

4642

3430 27 25 22 19

11 9

0

20

40

60

80

100

120

140

160Ja

pan

Ger

man

y

Isra

el

Net

herla

nds

UK

USA

Leba

non

Bra

zil

Chi

na

Sau

di A

rabi

a

Mal

aysi

a

Mex

ico

Indi

a

10© 2010 The Actuarial Profession www.actuaries.org.uk Source: World Health Organization, Most Recent Data available

Supply side statistics summaryNumber of physicians per 10,000 population

62

3734 32

28 28

21 21 2017 16 15

7 6

0

10

20

30

40

50

60

70Ita

ly

Isra

el

Ger

man

y

Net

herla

nds

USA

Leba

non

UK

Bra

zil

Japa

n

Mex

ico

Chi

na

Sau

di A

rabi

a

Mal

aysi

a

Indi

a

11© 2010 The Actuarial Profession www.actuaries.org.uk Source: World Health Organization, Most Recent Data available

Supply side statistics summaryNumber of health workers per 10,000 population

Source: World Health Organization, Most Recent Data available

169

134125

10799

75

4840

28 26 26 25 2114

0

20

40

60

80

100

120

140

160

180N

ethe

rland

s

Ger

man

y

USA

Japa

n

Isra

el UK

Sau

di A

rabi

a

Leba

non

Mex

ico

Bra

zil

Chi

na

Mal

aysi

a

Italy

Indi

a

Finland scores highest with 254 health workers per 10,000 population. The Netherlands is ranked 4th after Finland, Ireland and Norway. Liberia and Uganda score the lowest with 1 health worker per 10,000 population.


Utilisation comparisonsData sources

• Milliman Health Cost Guidelines (HCGs)• Swiss Re: data specifically requested from regional offices• Data requested on utilisation and costs for private

insurance• Data received in a consistent format for 11 countries• For some countries data for more than one population was

received, for example, for the UK we have:=> population data (HES data)=> an insured population of seniors=> a "commercial" insured population (working age)


Utilisation comparisonsImportance of utilisation (example)


• Total cost = Utilisation x Cost per procedure inclusive• 5 fold increase in CABG utilisation: 4% increase in total costs

Average Length of Stay Utilisation Cost (USD)

Total Cost (USD)

Hysterectomy 4.7 83 4,300 356,900 C‐Section Delivery 4.1 370 2,600 962,000 Normal Delivery 2.9 599 1,600 958,400 Coronary Artery Bypass Graft 9.6 8 22,400 179,200 Angiogram w/o stents 3.9 476 4,700 2,237,200 Angiogram with stents 5.1 96 10,300 988,800 Angiogram with drug‐eluting stents 3.9 24 12,700 304,800 Total Hip Replacement 9.3 32 16,900 540,800 Spinal Fusion 6.9 18 13,300 239,400 Arthroscopy 2.2 145 3,500 507,500 Gastroscopy 2.2 1,143 1,400 1,600,200 MRI 8.6 305 6,800 2,074,000 CT 7.3 1,130 6,500 7,345,000 Tonsillectomy 1.9 195 1,300 253,500 Mastectomy 2.7 101 3,500 353,500

4,725 18,901,200

Dr John Wennberg (Dartmouth Medical School)Widely recognised for pioneering research on health care

outcomes and patient-directed carePioneering study in 1967 – 1972:


Utilisation comparisonsProcedures chosen

Dr John E. Wennberg(photo by Jon Gilbert Fox – Dartmouth News)

• Observed huge differences for some• Tonsillectomy: 7% vs 70%• Hysterectomy: 2 per 1,000 vs 6 • Mastectomies, back surgery etc• Wennberg's unpopular conclusion: high

rates of surgery were not being driven by the patients but rather by the doctors

• - Source: "Overtreated: Why too much medicine is making us sicker and poorer" – Sharon Brownlee

Utilisation comparisonsMastectomies per 1,000 Medicare females 65-99


Source: Wennberg's Dartmouth Atlas: http://www.dartmouthatlas.org/data/map.aspx?ind=95

Utilisation comparisonsMastectomies per 1,000 population


0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8

Malaysia

India Commercial

India micro

Hong Kong

Brazil

Mexico

Saudi

Lebanon

Israel

USA

Germany

UK Commercial

UK HES data

UK Seniors

Asia

Latin

A

mer

ica

Mid

dle

Eas

tU

SE

urop

e

no data

no data

Utilisation comparisonsHysterectomies per 1,000 population

0 1 2 3 4 5 6 7 8 9 10

Malaysia

India micro

Hong Kong

Brazil

Mexico

Saudi

Lebanon

Israel

USA

Germany

UK Commercial

UK HES data

UK Seniors

Asi

aLa

tin

Am

eric

aM

iddl

e E

ast

US

Eur

ope

no data


Overall India privately insured = 1.2 – 1.5

Utilisation comparisonsProportion of C-sections

24%

67%

40%

53%

41%

49%

30%

56%

19%

9%

23%

42%

36%

12%

23%19%

30%28%

22%

14%

21%

0%

10%

20%

30%

40%

50%

60%

70%In

dia

Hon

g K

ong

Chi

na

Bra

zil

Mex

ico

Sau

di

Leba

non

Isra

el

USA

Ger

man

y

UK

Net

herla

nds

Sou

th A

frica

Asia Latin America Middle East US Europe Africa

Proportion of C-section deliveries (our data)

Proportion of C-section deliveries (WHO 2010 data)19


Utilisation comparisonsCABGs per 1,000 Medicare enrolees aged 65-99



Utilisation comparisonsCABGs per 1,000 population

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

Malaysia

India micro

India Commercial

Hong Kong

Brazil

Mexico

Saudi

Lebanon

Israel

USA

Germany

UK Commercial

UK HES data

UK Seniors

Asi

aLa

tin

Am

eric

aM

iddl

e E

ast

US

Eur

ope


no data

Utilisation comparisonsHip Replacements per 1,000 Medicare enrolees



Utilisation comparisonsHip Replacements per 1,000 population

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

Malaysia

India Commercial

India micro

Hong Kong

Brazil

Mexico

Saudi

Lebanon

Israel

USA

Germany

UK Commercial

UK HES data

UK Seniors

Asia

Latin

A

mer

ica

Mid

dle

Eas

tU

SE

urop

e


no data

Utilisation comparisons:MRI Scans vs MRI Scanners


Source of units per million population data: Health at a Glance 2009: OECD Indicators - OECD © 2009

26

78 8

4

6

2

118

39

110

29

16

58

14

0

20

40

60

80

100

120

140

0

5

10

15

20

25

30U

S (C

omm

erci

al)

Net

herla

nds

(All)

UK

(Com

mer

cial

)

UK

(NH

S)

Leba

non

(Com

mer

cial

)

Isra

el

(Com

mer

cial

)

Mex

ico

Scan

s pe

r 1,0

00 p

opul

atio

n

MR

I uni

ts p

er m

illio

n po

pula

tion MRI scanners per million population

MRI scans per 1,000 population

Utilisation comparisons:CT Scans vs CT Scanners


Source of units per million population data: Health at a Glance 2009: OECD Indicators - OECD © 2009

34

8 8 8

15

8

2

264

6033

59 51

116

5 0

50

100

150

200

250

300

0

5

10

15

20

25

30

35

40U

S (C

omm

erci

al)

Net

herla

nds

(All)

UK

(Com

mer

cial

)

UK

(NH

S)

Leba

non

(Com

mer

cial

)

Isra

el

(Com

mer

cial

)

Mex

ico

Scan

s pe

r 1,0

00 p

opul

atio

n

CT

units

per

mill

ion

popu

latio

n

CT Scanners per million population

CT Scans per 1,000 population

Utilisation comparisonsSpinal Fusions per 1,000 population


0 1 2 3 4 5 6

Malaysia

India Commercial

India micro

Hong Kong

Brazil

Mexico

Saudi

Lebanon

Israel

USA

Germany

UK Commercial

UK HES data

UK Seniors

Asia

Latin

A

mer

ica

Mid

dle

Eas

tU

SE

urop

e

no data


Utilisation comparisonsArthroscopies per 1,000 population

0 5 10 15 20 25 30

Malaysia

India Commercial

India micro

Hong Kong

Brazil

Mexico

Saudi

Lebanon

Israel

USA

Germany

UK Commercial

UK HES data

UK Seniors

Asia

Latin

A

mer

ica

Mid

dle

Eas

tU

SE

urop

e

no data

no data

Utilisation comparisonsGastroscopies per 1,000 population


0 5 10 15 20 25 30 35 40 45 50

Malaysia

India Commercial

India micro

Hong Kong

Brazil

Mexico

Saudi

Lebanon

Israel

USA

Germany

UK Commercial

UK HES data

UK Seniors

Asia

Latin

A

mer

ica

Mid

dle

Eas

tU

SE

urop

e

no data

no data

Utilisation comparisonsTonsillectomies per 1,000 population


0 1 2 3 4 5 6 7 8

Malaysia

India Commercial

India micro

Hong Kong

Brazil

Mexico

Saudi

Lebanon

Israel

USA

Germany

UK Commercial

UK HES data

UK Seniors

Asia

Latin

A

mer

ica

Mid

dle

Eas

tU

SE

urop

e

no data

Utilisation comparisonsNumber of GP consultations per life per year


0

2

4

6

8

10

12

14

OECD Economic Surveys: Japan - OECD © 2009 - ISBN: 978-92-64-05455-4

Our data: Data not available for many countriesUSA 3Malaysia 4.7Netherlands 4.1

Global PMI utilisation patterns% of hospital admissions per population per year


0% 10% 20% 30% 40% 50% 60%

Malaysia

India Commercial

India micro

Hong Kong

Brazil

Mexico

Saudi

Lebanon

USA

Netherlands

UK Commercial

UK HES data

UK Seniors

Asia

Latin

A

mer

ica

Mid

dle

Eas

tU

SE

urop

e

Hospital admissions Day case admissions

Cost comparisonAverage cost per hospital admission (not PPP adjusted)


$1,418

$1,174

$595

$3,711

$1,625

$1,603

$3,318

$6,665

$4,500

$0 $1,000 $2,000 $3,000 $4,000 $5,000 $6,000 $7,000

Malaysia

India Commercial

India micro

Hong Kong

Mexico

Saudi

Lebanon

USA

UK

Asi

aLa

tin

Am

eric

aM

iddl

e E

ast

US

Eur

ope

Conclusions

• Extreme differentials exist between PMI utilisation patterns currently

• Increased quantity does not necessarily imply better quality of health care

• Care must be taken to price PMI taking into account the uniqueness of each health care market and the product design

• For High Net Worth and high end products we are seeing some convergence – towards highest levels

• For microinsurance, the differential will persist indefinitely

Will these differentials persist or will utilisation patterns converge?Is this habit or hiatus?


Questions or comments?

Expressions of individual views by members of The Actuarial Profession and its staff are encouraged.The views expressed in this presentation are those of the presenters.


AppendixCountry summaries


– Europe: UK, Germany, Netherlands

– USA

– Asia: India, Malaysia, Hong Kong

– Middle East: Saudi Arabia, Lebanon, Israel

– Latin America: Brazil, Mexico

UK

• PMI is supplemental, voluntary and often overlapping• People have choice of using NHS or PMI for most

procedures• Use of PMI often depends on state of NHS in particular

area and specialty• Some drugs/procedures covered by PMI that not available

on NHS• Supply of hospital beds not constrained in private sector –

little control of utilisation• Mindset of “have paid premium, will use services”• Medium level of obesity & chronic disease in population,

but not high among people who have private coverage36


Germany

• 85% population have compulsory social insurance• 15% opt out and have alternative PMI• A high proportion also have supplemental insurance• PMI is heavily regulated in terms of premiums and UW• High cultural expectations of good coverage – traditionally

has included spa treatments!• High supply of medical infrastructure – docs, nurses, beds


Netherlands

• All private – every citizen must buy an approved level of private insurance. Top up insurance is voluntary

• Fairly supply-constrained in certain areas, with waiting lists• Reimbursement system does not encourage over-use –

tend to be episode based rather than Fee For Service ('FFS')

• Cultural expectations and ageing lead to high utilisation for certain services


USA

• PMI is first dollar coverage for most working population• Even government programmes often overseen by private

insurers• High supply and reimbursement encourages over-use,

despite efforts to control• Consumer & high spending power mindset encourages

excessive use• Cultural pre-occupation with consuming healthcare,

although politically not seen as universal right• Often generous coverage• High levels of obesity and chronic disease, but relatively

young population overall39


India

• PMI only covers 2-3% of population – relatively wealthy, privileged class

• Fast growing – premium growth of 30% + over last few years

• Also fast growing supply – PMI may be seen as a way of funding hospital building

• Tend to be inpatient-only coverage• Public health care very poor and struggling to deliver even

the most basic care to the population of 1.2 billion• Microinsurance fills the gap for some but cover is very

limited.


Malaysia

• Health care catered for by public and private providers• Lack of supply of health care resources is an issue• The high cost of private health care is also an issue• 65% of Medical and Health Insurance (“MHI”) premiums

relate to hospital and surgical cover (CI, LTC and hospital cash plans make up the balance)

• Increasing trend towards MHI, fuelled by increased tax relief in the mid 1990s

• Ongoing discussion about a National Health Financing Scheme for Malaysia


Source: Axco Report - Malaysia Life & Benefits

Hong Kong

• 2.7 million of the 7 million population in Hong Kong have some form of private medical insurance

• The government encourages take up of PMI by standardising products and requiring a savings facility for old-age premium costs

• Voluntary “top-up” policies are also available (mostly individual policies to supplement employer plans)

• Hong Kong is one of the most affluent societies in the world, standards of health care for the well-off are extremely high

• Lack of supply of private hospital beds due to demand from the mainland.

• Undisciplined market: high costs and high utilisation42

© 2010 The Actuarial Profession www.actuaries.org.ukSource: Axco Report - Hong Kong Life & Benefits

Saudi Arabia

• All expats must have approved minimum level of coverage• Historically some supply constraints, but large scale

hospitals building• Low utilisation for cultural reasons:

– Often lowly paid foreign workers don’t realise extent of insurance coverage

– Sick workers tend to go home, rather than be treated• By definition, covered people are fit enough to work


Lebanon


• Approx 1.5m people have PMI out of population of 4m• Much less ex-pat cover than other Middle Eastern

countries and more domestic cover• High utilisation trends as Lebanese awareness and

demand approaches that of Western countries like the US• Cultural trend towards increasing % of C-sections• Preference is to access specialist directly (gatekeeper

concept is not successful)

Israel

• Standards of health care are very high in Israel• Israel has a system of compulsory Social Insurance: every

resident is required to register as a member with one of the four Health Funds or “Kupat Holim”, deductions from income

• Supplementary benefits are available from each sick fund for an additional premium (known as Mashlim / SHABAN)

• In addition, top up cover is available from insurance companies, which covers, for example, medicines not covered by the sick funds, specified surgeries or organ transplants. This is essentially the PMI market in Israel.

• Even if the insured has cover through a Health Fund and supplementary benefits, they may not make use of them, choosing to place the full burden with the PMI insurer. 45


Brazil

• Public healthcare of very poor quality and 25% of population have private coverage – which is essentially duplicative

• No underwriting allowed for individual PMI, some for group• Cultural expectations of private healthcare is for treatment

without limit• Reimbursement usually encourages over-utilisation• Taking out of private healthcare encouraged by

government – private insurers obliged to cover new treatments by government– High utilisation trends due to new technologies/treatments


Mexico

• PMI is an alternative to comprehensive social insurance –essentially duplicates compulsory public insurance

• Approx 9m people have PMI out of population of 55m• Combination of major/minor medical coverage + dental,

optical etc• Little regulation as voluntary, with underwriting allowed• High net worth/middle class portfolio with high

expectations wanted freedom of choice of physicians and hospitals

• Few controls on utilisation – with open or preferred provider networks


Habit or hiatus - Institute and Faculty of Actuaries · Source: "Overtreated: Why too much medicine...

Documents

Transcript of Habit or hiatus - Institute and Faculty of Actuaries · Source: "Overtreated: Why too much medicine...