johan mackenbach department of public health, erasmus mc rotterdam, netherlands

THE PERSISTENCE OF HEALTH INEQUALITIES IN MODERN WELFARE STATES

Johan MackenbachDepartment of Public Health, Erasmus MCRotterdam, Netherlands

INEQUALITIES IN LIFE EXPECTANCY ALONG ROTTERDAM’S METROLINES

Jonker e.a., 2013

Neighbourhood

Life expectancy (men, 2007)

Health expectancy (men, 2007)

Nesselande 78,8 71,9

Ommoord 77,2 69,0

Oosterflank 76,7 68,7

Prinsenland 77,8 69,1

Kralingen-Oost 78,4 70,1

Kralingen-West 75,9 67,7

Stadsdriehoek 76,5 67,7

Cool 76,0 67,0

Delfshaven 74,3 65,3

Spangen 74,9 65,5

INEQUALITIES IN LIFE EXPECTANCY IN THE NETHERLANDS

30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79-0.1

0.0

0.1

0.2

0.3

0.4

0.5

0.6

Contributions (in years) of causes of death and age-groups to inequalities in life expectancy between educational groups, men, 1998-2007

other

extern

respir

othercancer

capros

calung

colorect

othercvd

cervas

ischae

Kulhanova et al., submitted

The great paradox of public health:

despite prosperity, more equal income distribution, welfare state, equal access to health care, …

health inequalities persist, and even are widening

TWO RESEARCH STRATEGIES

Zooming in: individuals, and how they differ in socioeconomic position, specific risk factors, and health outcomes

e.g. Whitehall, GLOBE, birth cohort studies, …

Zooming out: societies, and how they differ in social structure, risk factor distribution, and health inequalities

e.g. international comparative studies

WHAT DO VARIATIONS BETWEEN COUNTRIES TELL US?

Socioeconomic inequalities in mortality are universal and substantial – not smaller in Northern but in Southern Europe

Socioeconomic inequalities in mortality are widening – not responsive to policies aiming to reduce inequalities

Lifestyle risk factors play important role – with variability illustrating robustness of socioeconomic position as “fundamental cause”

VARIATIONS IN MAGNITUDE OF INEQUALITIES IN MORTALITY (M)

Mackenbach et al., submitted

Finland

Sweden

Norway

Denmark

England & W

Netherlands

BelgiumFrance

Switzerla

nd

Austria

Barcelona

Basque C

MadridTurin

Tuscany

Hungary

Czech

Rep

Poland

Estonia

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

Relative Risks of all-cause mortalitylow vs. high education, men, 2000s

Rela

tive

Risk

(age

-adj

uste

d, 3

0-79

yea

rs)

VARIATIONS IN MAGNITUDE OF INEQUALITIES IN MORTALITY (W)


Finland

Sweden

Norway

Denmark

England & W

Netherlands

BelgiumFrance

Switzerla

nd

Austria

Barcelona

Basque C

MadridTurin

Tuscany

Hungary

Czech

Rep

Poland

Estonia

0.0

0.5

1.0

1.5

2.0

2.5

0.0

0.5

1.0

1.5

2.0

2.5

Relative Risks of all-cause mortalitylow vs. high education, women, 2000s

Rela

tive

Risk

, age

-adj

uste

d, 3

0-79

yea

rs

WHY DO NORDIC COUNTRIES HAVE LARGER INEQUALITIES THAN SPAIN?

TotalHomicide

COPDAlcohol abuse

DiabetesLung cancerPneumonia

Isch. heart dis.Hypertensive dis.

Road accidentsSuicide

Oth. heart dis.Falls

Stomach cancerCerebrovasc. dis.

Colorectal cancerProstate cancer

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5

Relative Risks by cause of deathSpain vs. Northern Europe, men, 2000s

3 Spanish regions4 Nordic countries

Relative Risk (age-adjusted, 0-79 years)

Unpublished data from EURO-GBD-SE project

INEQUALITIES IN SMOKING (M)

30-44 45-59 60-69 70-790

0.1

0.2

0.3

0.4

0.5

0.6

0.7

Current smokingSpain vs. Nordic countries, men, 2000s

Spain, LowSpain, HighNordic, LowNordic, High

Prev

alen

ce o

f sm

okin

g


INEQUALITIES IN SMOKING (W)

30-44 45-59 60-69 70-790

0.1

0.2

0.3

0.4

0.5

0.6

Current smokingSpain vs. Nordic countries, women, 2000s

Spain, LowSpain, HighNordic, LowNordic, High

Prev

alen

ce o

f sm

okin

g


Finl

and

Swed

en

Norway

Denm

ark

Engl

and

& W

Belgi

um

Fran

ce

Switz

erla

nd

Barce

lona

Basqu

e C

Mad

ridTu

rin

Hunga

ry

Lith

uani

a

Esto

nia

0

0.5

1

1.5

2

2.5

3

3.5

Rate Ratios of all-cause mortalitylow vs. high education, 1990s and 2000s, men

1990s2000s

Rate

Ratio (

age-a

dju

sted, 30-7

4 y

ears

)WIDENING RELATIVE INEQUALITIES (M)


Finl

and

Swed

en

Norway

Denm

ark

Engl

and

& W

Belgi

um

Fran

ce

Switz

erla

nd

Barce

lona

Basqu

e C

Mad

ridTu

rin

Hunga

ry

Lith

uani

a

Esto

nia

0

200

400

600

800

1000

1200

1400

1600

1800

Rate differences of all-cause mortalitylow vs. high education, 1990s and 2000s, men

1990s2000s

Rate

diff

ere

nce

(age-a

dju

sted, 30-7

4 y

ears

)VARIABLE TREND ABSOLUTE INEQUALITIES (M)

Mackenbach et al. submitted

13 YEARS OF LABOUR GOVERNMENT

UNIQUE POLICY EXPERIMENT

IF THIS DID NOT WORK, WHAT WILL?

DID ENGLAND HAVE MORE NARROWING OF ABSOLUTE INEQUALITIES?

1981-85 1986-90 1991-95 1996-00 2001-05 2006-100

100

200

300

400

500

600

700

800

900

Absolute inequalities in all-cause mortality1981-2010, men

FinlandNorwayEngland & WItaly, Turin

Diff

eren

ce b

etw

een

low

and

hig

h ed

ucat

ed

(dea

ths

per

1000

00)

Unpublished data from DEMETRIQ project



1981-85 1986-90 1991-95 1996-00 2001-05 2006-100

50

100

150

200

250

300

350

400

450

Absolute inequalities in Cardiovascular disease mortality1981-2010, men


Diff

eren

ce b

etw

een

low

and

hig

h ed

ucat

ed

(dea

ths

per

1000

00)



1981-85 1986-90 1991-95 1996-00 2001-05 2006-100

50

100

150

200

250

Absolute inequalities in smoking-related mortality1981-2010, men


Diff

eren

ce b

etw

een

low

and

hig

h ed

ucat

ed

(dea

ths

per

1000

00)

DIFFERENCE-IN-DIFFERENCE ANALYSIS OF INEQUALITIES IN 5 OUTCOMES

Hu et al., in preparation

ISCHEMIC HEART DISEASE (M)

Finlan

d

Swed

en

Norway

Denmark

Engla

nd & W

ales

Netherl

ands

Belgium

France

Switz

erlan

d

Austria

Barcelo

na

Basque C

ountry

Madrid

Turin

Tusca

ny

Hungary

Czech Rep

ublic

Poland

Estonia

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

0

0.5

1

1.5

2

2.5

3

3.5

Ischemic heart diseaseRelative Risks, low vs. high, 2000s, men

Rela

tive

Risk

(age

-adj

uste

d, 3

0-79

yea

rs)


ROLE OF SMOKING AND OVERWEIGHT (M)

Finla

nd

Sweden

Norway

Denm

ark

Engla

nd/W

Scotla

nd

Nethe

rland

s

Belgi

um

Franc

e

Switzer

land

Austri

a

Barce

lona

Basqu

e Cou

ntry

Mad

ridTur

in

Tusca

ny

Czech

Rep

.

Hunga

ry

Polan

d

Lith

uani

a

Estoni

a0

5

10

15

20

25

30

35

Percent contribution to inequalities in mortalityby education, 2000s, men

SmokingOverweight

Perc

enta

ge re

ducti

on o

f exc

ess r

isk

Eikemo et al., submitted

ROLE OF SMOKING AND OVERWEIGHT (W)

Finla

nd

Sweden

Norway

Denm

ark

Engla

nd/W

Scotla

nd

Nethe

rland

s

Belgi

um

Franc

e

Switzer

land

Austri

a

Barce

lona

Basqu

e Cou

ntry

Mad

ridTur

in

Tusca

ny

Czech

Rep

.

Hunga

ry

Polan

d

Lith

uani

a

Estoni

a0

5

10

15

20

25

30

35

Percent contribution to inequalities in mortalityby education, 2000s, women

SmokingOverweight

Perc

enta

ge re

ducti

on o

f exc

ess r

isk

Eikemo et al., submitted

PREVENTABLE VERSUS NON-PREVENTABLE CAUSES (M)

Hodgkin's disease and leukemia (M)Cancer of prostate (M)

Cancer of colorectum (N)Cancer of pancreas (N)

Cancer of kidney and bladder (N)Cancer of liver (N)

Hypertensive disease (B/M)Accidental falls (I)

Cerebrovascular disease (B/M)Ischemic heart disease (B/M)

Cancer of stomach (N)Road traffic accidents (I)

Suicide (I)Other liver and gall bladder diseases (N)

Diabetes mellitus (B)Cancer of buccal cavity, pharynx, and oesophagus (B)

Appendicitis, hernia and peptic ulcer (M)Pneumonia / influenza (M)

Cancer of trachea, bronchus and lung (B)Chronic Obstructive Pulmonary Disease (B)

Alcohol abuse (B)Cancer of larynx B)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5

Median Mortality Relative Risk19 European populations, 2000s, men

RR (low vs. high education)


PREVENTABLE VERSUS NON-PREVENTABLE CAUSES (M)

Hodgkin's disease and leukemia (M)

Cancer of prostate (M)

Cancer of colorectum (N)

Cancer of pancreas (N)

Cancer of kidney and bladder (N)

Cancer of liver (N)

Hypertensive disease (B/M)

Accidental falls (I)

Cerebrovascular disease (B/M)

Ischemic heart disease (B/M)

Cancer of stomach (N)

Road traffic accidents (I)

Suicide (I)

Other liver and gall bladder diseases (N)

Diabetes mellitus (B)

Cancer of buccal cavity, pharynx, and oesophagus (B)

Appendicitis, hernia and peptic ulcer (M)

Pneumonia / influenza (M)

Cancer of trachea, bronchus and lung (B)

Chronic Obstructive Pulmonary Disease (B)

Alcohol abuse (B)

Cancer of larynx B)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5

Median Mortality Relative Risk19 European populations, 2000s, men

RR (low vs. high education)


PREVENTABLE VERSUS NON-PREVENTABLE CAUSES

Men Women

Obs. Median RR Obs. Median RR

All causesof death 385 1,90 343 1,67

All preventablecauses 276 2,15 244 1,90

Amenable to behaviour change 161 2,35 146 2,30

Amenable to medical intervention 118 1,82 127 1,90

Amenable to injury prevention 51 1,94 38 1,40

All non-preventable causes 109 1,53 99 1,43

Median Mortality Relative Risk19 European populations, 2000s, men and women


ROBUSTNESS OF HEALTH

INEQUALITIES =

ROBUSTNESS OF SOCIAL

INEQUALITY

WHY HEALTH INEQUALITIES PERSIST DESPITE THE WELFARE STATE

Inequalities in access to material and immaterial resources have not been eliminated by the welfare state

Social mobility has become more sensitive to personal characteristics that are associated with health (mental health, cognitive ability, …)

Welfare state is not effective against determinants of disease that are linked to consumption behavior

Mackenbach 2012

REDISTRIBUTIVE EFFECTS OF THE WELFARE STATE, NETHERLANDS

0

200

400

600

800

1000

1200

BasicLower sec.

Higher sec.Lower voc. Higher

voc. University

Before govt. redistr.

After govt. redistr.

Ter Rele 2007

Lifetime wages before government intervention, vs. lifetime welfare after tax, cash transfers and non-cash benefits, in thousands of Euros, Netherlands, ca. 2002

RISE OF INTERGENERATIONAL SOCIAL MOBILITY

0

5

10

15

20

25

30

35

40

1914-391940-49

1950-591960-69

University

Higher voc.

Higher sec.

Lower sec.

Basic

RMO 2011

TOBACCO CONTROL AND INEQUALITIES IN QUIT RATIOS

Schaap et al. 2008

SOBERING CONCLUSIONS

Magnitude of socioeconomic inequalities in mortality varies substantially between countries, suggesting great potential for reduction – but smaller inequalities do not reflect more effective policies

Recently, absolute mortality inequalities have started to decline in some countries – but do not reflect success of national programs to reduce health inequalities, and relative inequalities continue to rise

Health inequalities are remarkably robust across time and place – we need better ideas for addressing inequalities in consumption behavior and/or their fundamental causes

THANK YOU!

Mackenbach JP. The persistence of health inequalities in modern welfare states: The explanation of a paradox. Soc Sci Med 2012;75:761-769.

Eikemo TA et al. How can inequalities in health be reduced? A study of 6 risk factors in 21 European populations. Submitted for publication

Kulhanova I et al. Why does Spain have smaller inequalities in mortality? Eur J Publ Health (in press)

Mackenbach JP et al. Widening inequalities in mortality in mortality: a study of 3.2 million deaths in 13 European countries. Submitted for publication

Mackenbach JP et al. Variations in the relation between education and cause-specific mortality in 19 European populations: a test of the “fundamental causes” theory of social inequalities in health. Submitted for publication

johan mackenbach department of public health, erasmus mc rotterdam, netherlands

Documents

magnitude of inequalities

larger inequalities

mortality mmackenbach

mortality wmackenbach

smoking munpublished

smoking wunpublished

health inequalitiese

health outcomese