Downloaded from http://jech.bmj.com/ on January 18, 2018 - Published by group.bmj.

com
158

RESEARCH REPORT

Has the relation between income inequality and life

expectancy disappeared? Evidence from Italy and top
industrialised countries
Roberto De Vogli, Ritesh Mistry, Roberto Gnesotto, Giovanni Andrea Cornia
...............................................................................................................................
J Epidemiol Community Health 2005;59:158–162. doi: 10.1136/jech.2004.020651

Objective: To investigate the relation between income inequality and life expectancy in Italy and across
wealthy nations.
Design and setting: Measure correlation between income inequality and life expectancy at birth within
Italy and across the top 21 wealthy countries. Pearson correlation coefficients were calculated to study
these relations. Multivariate linear regression was used to measure the association between income
inequality and life expectancy at birth adjusting for per capita income, education, and/or per capita gross
domestic product.
Data sources: Data on the Gini coefficient (income inequality), life expectancy at birth, per capita income,
and educational attainment for Italy came from the surveys on Italian household on income and wealth
See end of article for 1995–2000 and the National Institute of Statistics information system. Data for industrialised nations were
authors’ affiliations taken from the United Nations Development Program’s human development indicators database 2003.
.......................
Results: In Italy, income inequality (b = 20.433; p,0.001) and educational attainment (b = 0.306;
Correspondence to: p,0.001) were independently associated with life expectancy, but per capita income was not (b = 0.121;
Dr R De Vogli, Department p.0.05). In cross national analyses, income inequality had a strong negative correlation with life
of Epidemiology and
expectancy at birth (r = 20.864; p,0.001).
Public Health, University
College of London (UCL), Conclusions: In Italy, a country where health care and education are universally available, and with a
London WC1E 6BT, UK; strong social safety net, income inequality had an independent and more powerful effect on life
devogli@libero.it expectancy at birth than did per capita income and educational attainment. Italy had a moderately high
Accepted for publication degree of income inequality and an average life expectancy compared with other wealthy countries. The
28 April 2004 cross national analyses showed that the relation between income inequality and population health has not
....................... disappeared.

I
n the past three decades there has been great interest in have been proposed to explain away the effect of income
examining the relation between income inequality and inequality on health: per capita income and educational
health. Income inequality has been found to be associated, attainment. Gravelle argued that the relation between
in a cross national comparison, with a series of health income inequality and mortality is an artefact of the non-
indicators including infant mortality, and life expectancy at linear relation between income and mortality at the
birth and at age 5.1 Such findings have been replicated using individual level.19 However, it is precisely the non-linear
different indicators across both a wide range of countries2–9 relation between micro-level income and life expectancy at
and within countries.10–12 birth that justifies the inclusion of income inequality in the
More recently, the evidence that income inequality is one study of mortality at the macro-level, for example, when
of the major determinants of population health has been making cross national comparisons.1 Later, Wolfson and
questioned. Some authors claimed the relation between colleagues20 using multilevel analyses provided substantial
income inequality and health vanished to a large extent evidence for a non-artefactual explanation.21 22 Others still
when new studies with better data from different countries maintain that population health measured by life expectancy
were available.13 Lynch and colleagues suggested that the does not depend on how income is distributed, but individual
relation between income inequality and life expectancy may level income instead.16 17 23
have resulted from the small number of countries for which With regard to educational attainment, Muller using cross
comparable data were initially available.4 A recent study from sectional data on income inequality and age adjusted
Canada did not find a significant association between income mortality in the US concluded that the proportion of adults
inequality and mortality concluding that the relation with high school diploma in a population accounts for the
between income distribution and health is not universal, income inequality effect.24 These findings have been subse-
but instead dependent on social and political characteristics quently augmented with data from Brazil, which showed that
specific to place.14 However, a subsequent study from Canada the introduction of illiteracy into the analysis explained
showed that labour market inequality was related to working away the association between life expectancy and income
age mortality.15 Associations were not found between income inequality.25
inequality and different health outcomes such as all cause In summary, many questions remain unanswered about
mortality and self rated health in Denmark,16 Japan,17 and whether and why there is an association between income
New Zealand.18 inequality and health outcomes, which require further
As results on the relation between income distribution and evidence from a variety of social, economic, and political
health outcomes have been inconsistent, previous evidence contexts. If the relation between income inequality and
has been dismissed as spurious. At least two major factors health varies according to characteristics specific to place, it is

www.jech.com
 Downloaded from http://jech.bmj.com/ on January 18, 2018 - Published by group.bmj.com
Income inequality and life expectancy at birth 159

of great interest to investigate such association in different

A
nations. To date, we are not aware of studies that have 81.5
undertaken this research in Italy, a country characterised by Marches
large regional differences in per capita income and income 81

Life expectancy at birth (2001)

Abruzzo
inequality, for example, the relative affluence of the north as Umbria
Tuscany Molise
compared with the south of Italy.26 80.5 Trentino Alto Adige
Veneto
In this research, we analysed the relation between the Gini Emilia-Romagna Calabria
Apulia
coefficient and life expectancy at birth using aggregated data 80 Fruili Venezia
Laetium Giulia Basilicata
for all 20 Italian regions. We also examined whether or not Aosta Valley Lombardy
Sardinia
Liguria
this association is the result of variation in per capita income 79.5 Piedmont Sicily
and educational attainment. Finally, we place the Italian data
within an international context by studying the relation 79
between income inequality and life expectancy at birth across r = –0.785; p < 0.001
top 21 industrialised countries. 78.5 Campania

78
METHODS 0.2 0.22 0.24 0.26 0.28 0.3 0.32 0.34 0.36 0.38 0.4
Italy Gini coefficient (1995–2000)
Data on life expectancy at birth for the year 2001, the study B
dependent variable, were extracted from the Italian National 81.5
Institute of Statistics information system.27 The Gini coeffi- Marches
81

Life expectancy at birth (2001)

cient was calculated using pooled data from the surveys on Abruzzo
Italian household on income and wealth (SIHIW) conducted Molise Umbria Tuscany Trentino
80.5 Alto Adige
by the Bank of Italy in 1995 (n = 8135), 1998 (n = 7147), and Veneto
2000 (n = 8001). Pooling of the three surveys was performed Emilia
Calabria Apulia Romagna
80 Basilicata
to assure a sample of at least 1000 subjects for 11 regions and Laetium Lombardy
Piedmont Fruili Venezia
at least 500 subjects for the remaining nine regions.28 Face to Sardinia Liguria Giulia
79.5 Aosta
face interviews of household members were conducted. The Sicily Valley

sampling strategy consisted in two stages: firstly, a random

79
sample of 300 municipalities (stratified by region and county
size) was selected; secondly, a random sample of Italian 78.5 r = 0.538; p < 0.001
Campania
households was drawn within each municipality. Post-
stratification adjustment or raking was used to assure that 78
regional samples were similar to the general population in 15 000 19 000 23 000 27 000 31 000 35 000
terms of sex, age, occupation, educational attainment, and Per capita income in euros (2000)
municipality size. A more detailed description of the C
81.5
sampling methodology is discussed elsewhere.26
Covariates for the within Italy analysis included per capita Marches
81
Life expectancy at birth (2001)

income and educational attainment for each Italian region. Molise

Data on per capita income for each Italian region came from Abruzzo
Trentino Umbria
80.5 Tuscany
the SIHIW conducted in 2000.28 Using data from the National Alto Adige Veneto Emilia-Romagna

Institute of Statistics, we measured educational attainment Apulia Calabria

80 Basilicata Fruili Venezia
as the proportion of persons 19 years old in each Italian Aosta Valley Giulia
Lombardy Liguria Laetium
region that obtained a high school diploma as of 1997.29 Data Sardinia Piedmont
79.5
on population size, life expectancy at birth, educational Sicily

attainment, per capita income, and Gini coefficients across 79

Italian regions are shown in table A (see appendix, available
on line http://www.jech.com/supplemental). 78.5 r = 0.47; p < 0.001
Campania

Cross national 78
55 60 65 70 75 80 85
Data on life expectancy at birth for the year 2001, Gini
coefficient (data were available for different years—that is, Percentage of persons 19 years old with a
high school diploma (1997)
from 1990 to 1998), and per capita gross domestic product
(GDP) for Italy and the other 20 industrialised countries
Figure 1 (A) Income inequality and life expectancy at birth among
included in this study were taken from the United Nations Italian regions (n = 20). Data are weighted by population size and
Development Program’s human development indicators adjusted for gross domestic product per capita. r (crude) = 0.607;
published in 2003.30 Countries selected for the analysis were p,0.005. r (adjusted for per capita income) = 20.934; p,0.011.
the top 25 nations in terms of per capita GDP—Australia, r (weighted by population size) = 0.658; p,0.001. (B) Economic
Belgium, Canada, Denmark, Finland, France, Germany, development and life expectancy at birth among Italian regions (n = 20).
Greece, Italy, Japan, Luxembourg, Netherlands, New Data are weighted by population size. (C) Association between
educational attainment and life expectancy at birth among Italian
Zealand, Norway, Portugal, Singapore, Spain, Sweden, regions (n = 20). Data are weighted by population size.
Switzerland, UK, and USA—excluding Austria, Ireland,
Israel, and Slovenia. Austria was excluded because income
inequality data were not available. Ireland, Israel, and capita income.31 Data on population size, life expectancy at
Slovenia were excluded because of their social instability birth, per capita GDP, and Gini coefficients across top
and conflicts. Data on the Gini coefficient were calculated industrialised nations are shown in table B (see appendix).
using per capita income as a living standard indicator and
came from nationally representative household surveys using Statistical analyses
similar methodologies. The Gini coefficient was adjusted for For analyses within Italy, Pearson correlation coefficients
household size, providing a more consistent measure of per were calculated to measure the association between life

www.jech.com
 Downloaded from http://jech.bmj.com/ on January 18, 2018 - Published by group.bmj.com
160 De Vogli, Mistry, Gnesotto, et al

Table 1 Linear regression on life expectancy at birth across Italian regions (n = 20)
Standardised b coefficients

Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7

Gini index 20.658*** – – 20.785*** 20.559*** – 20.443***

Per capita – 0.538*** – 20.146* – 0.496*** 0.121
income
Education – – 0.476*** – 0.276*** 0.426*** 0.306***
Constant 82.652 78.287 76.315 83.616 80.187 75.259 79.126
2
r 0.433 0.290 0.226 0.438 0.500 0.469 0.502
Model statistics:
F (df, N) 363.9*** (1, 20) 194.3*** (1, 20) 139.4*** (1, 20) 185.4*** (2, 20) 237.5*** (2, 20) 210.5*** (2, 20) 159.8*** (3, 20)

*p,0.05; **p,0.01; ***p,0.001.

expectancy at birth and income inequality, per capita income,

82
and educational attainment. Multivariate linear regression
was used to study the relation between income inequality Japan
81

Life expectancy at birth (2001)

and life expectancy when adjusting for per capita income and
educational attainment. Problems of collinearity could have
80
arisen because of the high correlation between income Sweden

inequality and per capita income (r = 20.851; p,0.001). As Canada Spain

Australia
an indicator of collinearity, we used variance inflation factor 79 Norway
France Switzerland
Italy
(VIF) values greater than or equal to 10,32 however in none of Belgium Luxembourg Greece New Zealand

the multivariate regression models did the VIF exceed this 78 Netherlands Germany Singapore
Finland UK
threshold.
For analyses between countries, Pearson correlation 77 USA

coefficients adjusted for GDP per capita were calculated to Denmark

measure the association between life expectancy at birth and 76 Portugal

r = –0.864; p < 0.001
income inequality.
All analyses were weighted by population size. 75
20 25 30 35 40 45
Gini coefficient (1990–1998)
RESULTS
Figure 1A shows the relation between the Gini coefficient and Figure 2 Income inequality and life expectancy at birth among
life expectancy at birth across Italian regions. Income inequality industrialised countries (n = 21). Data are from the human development
had a strong negative correlation with life expectancy indicators 2003. The correlation presented in the figure is weighted by
population size and adjusted for per capita gross domestic product
(r = 20.785; p,0.001). Regions whose income inequality was
(GDP). r (crude) = 0.415; p,0.65. r (adjusted for per capita GDP =
higher, such as Campania and Sicily, had a significantly lower 20.433; p,0.065. r (weighted by population size) = 20.907;
life expectancy than regions where income inequality was p,0.001.
comparatively low, for example, Marches and Umbria.
There were large variations among Italian regions in terms
of per capita income (fig 1B), which ranged from J16 492 for inequality remained strongly associated with life expectancy
Basilicata to J33 774 for Emilia-Romagna.28 Per capita (b = 20.443; p,0.001; model 7). Moreover, the adjusted r2
income was positively correlated with life expectancy at birth increased only slightly.
(r = 0.538, p,0.001). Regions with low per capita income Figure 2 is a scatterplot of data on the Gini coefficient and
such as Sicily and Campania had significantly lower life life expectancy at birth across the top 21 industrialised
expectancy at birth compared with regions whose per capita countries, including Italy.30 Income inequality was strongly
income was high such as Emilia-Romagna, Trentino Alto negatively correlated with life expectancy at birth
Adige, and Tuscany. (r = 20.864; p,0.001). Among wealthy countries, Italy has
a moderately high degree of income inequality (Gini = 0.36)
Educational attainment was also positively correlated with
and an average life expectancy at birth (78.6 years).
life expectancy at birth (r = 0.476; p,0.001; fig 1C). Regions
with a higher proportion of persons 19 years old having a
high school diploma performed better in terms of life DISCUSSION
expectancy compared with regions where a lower proportion This study has a number of important findings. The first set
of young adults of the same age has a high school diploma. of findings pertains to the Italian context. In Italy, income
Table 1 shows the results from multivariate linear inequality was associated with life expectancy at birth even
regressions. The Gini coefficient by itself explained 43.3% of when per capita income and education (that is, high school
the variation in life expectancy at birth (model 1). When per completion rate of persons 19 years old) were held constant.
capita income was introduced into the regression (model 4), Per capita income did not explain the relation, but education
the strength of the association between the Gini coefficient did to a small extent. There was an independent effect of
and life expectancy increased. The fit of the regression model education, but not of per capita income, on life expectancy at
improved, as shown by the increase in the adjusted r2 from birth. The second set of findings situates the data from Italy
43.3% (model 1) to 50.0% (model 5), when educational in an international context. It is a country with a moderately
attainment was added to model with the Gini coefficient. high degree of income inequality and an average life expectancy
Per capita income and educational attainment together compared with other wealthy nations. Among these nations,
explained 46.9% of the variance on life expectancy (model income inequality was negatively associated with life expec-
6). When all three explanatory factors were included into the tancy at birth, which is consistent with earlier reports.1 3 6
multiple regression analysis, the b coefficient for per capita Amid recent data supporting the contrary,14 16–18 our
income was not significant while the coefficient for income findings support the relative income hypothesis, which states

www.jech.com
 Downloaded from http://jech.bmj.com/ on January 18, 2018 - Published by group.bmj.com
Income inequality and life expectancy at birth 161

Results, showing that education played a significant part in

Key points explaining the relation between income inequality and life
expectancy, are partially consistent with the material
N Among Italian regions, income inequality has an deprivation hypothesis. Income inequality may be partially
independent and more powerful effect on life expec- associated with poorer infrastructures and lack of resources
tancy than per capita income and education. Per among the most disadvantaged populations, especially in
capita income does not account for the effect of income southern Italy. However, we did not find an independent
inequality on life expectancy while education plays a effect of per capita income, yet an independent effect of
partial, but limited part in explaining the relation. income inequality predominates. In contrast with previous
studies,24 25 the effect of education was small and did not
N Italy has a moderately high degree of income inequal-
account for the relation between income inequality and
ity and an average life expectancy compared with health. This may suggest that income inequality may exert an
other wealthy countries. influence through psychosocial mechanisms such as social
N Data from the top 21 industrialised nations including stress, for example from the unequal distribution of power
Italy show that there is a strong negative relation among citizens, lack of political participation, and social
between income inequality and life expectancy at birth. cohesion. However, this study was not equipped to investi-
gate this. Rather than considering the two major pathways of
causation (material deprivation and psychosocial stress) as
that for regions with GNP per capita beyond US$ 8000– mutually exclusive, our study suggests that a combination of
10 000, health status is mainly affected by the spread of the two pathways are at work.34
income.3 Conversely, the absolute income hypothesis assert- The results from analyses between wealthy countries
ing that health is primarily influenced by economic develop- corroborate previous evidence showing a significant correla-
ment, even in a comparatively rich nation, is not supported tion between income inequality and life expectancy.1 6
by the data on Italy. Income inequality explained most of the Countries having lower levels of income inequality such as
variation in life expectancy, and the effect of per capita Japan and Sweden enjoy a better health compared with
income on life expectancy disappeared when controlling for countries with high levels of income inequality such as the
income inequality. Regions such as Marches, Umbria, and USA and Portugal. These results, however, are in contrast
Tuscany that are more equitable in terms of income with the evidence presented by Lynch and colleagues who
distribution and enjoyed a higher life expectancy. On the concluded that the higher income inequality was not
other hand, highly inequitable regions such as Sicily and associated with lower life expectancies.4 They suggested that
Campania performed quite poorly in terms of life expectancy. the relation had disappeared because of the higher number of
Two major pathways of causation have been proposed for countries in their study, in comparison with the analysis
the effect income inequality exerts on health status: the presented by Wilkinson in 1992.3
material deprivation pathway and the psychosocial path- Our findings indicate that with updated data and the
way.3 33 The material deprivation pathway states that income inclusion of five additional top industrialised countries in the
inequality is related to health through a combination of analysis the relation reappears. Inclusion of Greece, Japan,
negative exposures and lack of resources held by individuals, New Zealand, Portugal, and Singapore, which were excluded
along with systematic underinvestment across a wide range in the analysis conducted by Lynch and colleagues because of
of human, physical, health, and social infrastructures (for unavailability of data,4 may explain, at least partially, the
example, education, health services, transportation, environ- difference in results. However, a sensitivity test showed that,
mental controls, availability of good quality food, quality of when these countries were excluded from the analysis, there
housing, and occupational health regulations).33 The psycho- was a moderate correlation between income inequality and
social pathway argues that income inequality affects health life expectancy (data not shown).
through individual perceptions of place in the social Discrepancies may also be attributable to the delayed effects
hierarchy producing negative emotions such as stress, shame, of changes in income distribution on mortality rates.35 During
and distrust that are translated ‘‘inside’’ the body into poorer the late 1980s and 1990s some countries (for example, United
health via psycho-neuro-endocrine-immunological mechan- Kingdom and Finland) experienced changes in income
isms and stress induced behaviours such as smoking and distribution. The true effect on life expectancy of these changes
overeating. Simultaneously, perceptions of relative position may have been difficult to detect (especially in the oldest age
and negative emotions are also translated ‘‘outside’’ the groups) in 1990–93, the years when the life expectancy data
individual into antisocial behaviours such as homicides, used by Lynch and colleagues were collected.4
traffic accidents, and reduced civic participation, which result in This study has several limitations. As shown in other
less social capital and social cohesion within the community.3 research, the level of geographical aggregation influences the
This study shows that income inequality is strongly association between income inequality and health. At higher
associated with health in a country where health care and levels of aggregation it is easier to find independent effects of
education are publicly funded with practically free access to income inequality on health outcomes.22 We do not know
all citizens, and unemployed people receive social benefits. whether or not the use of the Gini coefficient at lower
geographical levels such as provinces or municipalities would
have produced different results. However, the level of
Policy implications aggregation needs not to be too small to allow income
distribution to exert an effect independent of individual
N In the Italian context, improving population health may income.36 Further studies from Italy using measures of
income distribution at lower levels of aggregation would be
not only require policies that promote economic
development, but also those that reduce income useful to clarify this point. Another limitation is that the
inequality and increase educational attainment. cross sectional data used precludes investigating how
changes in income inequality over time influence population
N To promote population health, governments of wealthy health. It is true that inequalities can take time to show their
nations are advised to minimise income inequality. effects, but northern Italy has been more equitable than
southern Italy at least since the creation of the unified

www.jech.com
 Downloaded from http://jech.bmj.com/ on January 18, 2018 - Published by group.bmj.com
162 De Vogli, Mistry, Gnesotto, et al

state—that is, as far as income inequalities are concerned the REFERENCES

situation has not substantially changed when we compare 1 Rodgers G. Income and inequality as determinants of mortality: an
different regions. This analysis may also incur ‘‘ecological international cross-sectional analysis. Population Studies 1979;33:343–51.
2 Waldman R. Income distribution and infant mortality. Quarterly Journal of
fallacies’’ when interpreting aggregated data to make Economics 1992;107:1283–302.
conclusions about individuals.37 Multilevel analyses using 3 Wilkinson R. Unhealthy societies: the affliction of inequalities. London:
individual level data on income and aggregated data on Routledge, 1996.
income inequality within Italy may contribute in the 4 Lynch JW, Davey Smith G, Hillemeier M, et al. Income inequality, the
psychosocial environment and health: comparisons of wealthy nations. Lancet
confirmation or rejection of study findings. However, as 2001;358:194–200.
income inequality is inherently a social concept, it may not be 5 LeGrand J. Inequalities in health. Some international comparisons. European
plausible, both from a theoretical and methodological view- Economic Review 1987;31:182–91.
6 Wilkinson R. Income distribution and life expectancy. BMJ 1992;304:165–8.
point, to evaluate its effects at the individual level. Finally,
7 Marmot M, Bobak M. International comparators and poverty and health in
our cross national results must be interpreted with caution Europe. BMJ 2000;321:1124–8.
because of the difference between the reference dates of the 8 Davey Smith G, Egger M. Understanding it all—health, meta-theories, and
Gini coefficients. While most of the Gini coefficients were mortality trends. BMJ 1996;313:1584–5.
9 Walberg P, McKee M, Shkolnikov V, et al. Economic change, crime, and
collected between 1994 and 1998, data for Switzerland, mortality crisis in Russia: regional analysis. BMJ 1998;317:312–18.
Japan, and Spain were from 1990 and 1993. Moreover, 10 Kaplan G, Pamuk ER, Lynch JW, et al. Inequality in income and mortality in
comparability problems remain in the use of data from the United States: analysis of mortality and potential pathways. BMJ
household surveys across countries, even though these 1996;312:999–1003.
11 Kennedy B, Kawachi I, Prothrow D. Income distribution and mortality: cross-
problems are diminishing as survey methodologies are sectional ecological study of the Robin Hood Index in the United States. BMJ
improving and becoming more standardised.31 1996;312:1004–7.
Despite the limitations, our research suggests that the 12 Ben Shlomo Y, White I, Marmot M. Does the variations in the socioeconomic
relation between income inequality and population health characteristic of an area affect mortality. BMJ 1996;312:1013–14.
13 Mackenbach JP. Income inequality and population health: evidence favouring
among wealthy nations has not disappeared as claimed by a negative correlation between income inequality and life expectancy has
some authors.4 13 Furthermore, using data from Italy, a disappeared. BMJ 2002;324:1–2.
country with a moderately high degree of income inequality 14 Ross NA, Wolfson M, Dunn JR, et al. Relation between income inequality and
mortality in Canada and in the United States: cross sectional assessment using
among the industrialised nations, we show that, in contrast census data and vital statistics. BMJ 2000;320:898–902.
with studies from other countries,23–25 income inequality had 15 Sanmartin C, Ross NA, Tremblay S, et al. Labour market income inequality
an independent and more powerful effect on life expectancy and mortality in North American metropolitan areas. J Epidemiol Community
at birth than per capita income and educational attainment. Health 2003;57:792–7.
16 Osler M, Prescott E, Gronbaek M, et al. Income inequality, individual income,
More research is needed to analyse the material and and mortality in Danish adults: analysis of pooled data from two cohort
psychosocial mechanisms between income inequality and studies. BMJ 2002;324:1–4.
health in wealthy nations including Italy. However, despite 17 Shibuya K. Individual income, income distribution, and self rated health in
which pathway of causation predominates, income inequa- Japan: cross sectional analysis of nationally representative sample. BMJ
2002;324:1–5.
lity remains a key determinant of population health. 18 Blakely T, Atkinson J, O’Dea D. No association of income inequality with
Consequently, public policy and public health initiatives to adult mortality within New Zealand: a multi-level study of 1.4 million 25–64
reduce or contain such inequalities are essential to reach year olds. J Epidemiol Community Health 2003;57:279–84.
19 Gravelle H. How much of the relation between population mortality and
public health objectives at the population level.
unequal distribution of income is a statistical artefact? BMJ 1998;316:382–5.
20 Wolfson M, Kaplan G, Lynch JW, et al. Relationship between income
ACKNOWLEDGEMENTS inequality and mortality: empirical demonstration. BMJ 1999;319:953–7.
We greatly appreciate the invaluable review and comments provided 21 Daly M, Duncan G, Kaplan GA, et al. Macro-to-micro linkages in the
by Richard Wilkinson and George Davey Smith in the preparation of inequality-mortality relationship. Milbank Mem Fund Q 1998;76:315–39.
this manuscript. 22 Soobader M, LeClere F. Aggregation and the measurement of income
inequality: effects on morbidity. Soc Sci Med 1999;48:733–44.
23 Fiscella K. Poverty or income inequality as predictor of mortality: longitudinal
CONTRIBUTORS cohort studies. BMJ 1997;314:1724.
RD performed most of the analyses and wrote most of the paper. RM 24 Muller A. Education, income inequality and mortality: a multiple regression
participated in the writing of the paper, reviewed its contents, and analysis. BMJ 2002;324:23–5.
conducted some statistical analyses. RG inspired the development of 25 Messias E. Income inequality, illiteracy rate, and life expectancy in Brazil.
research hypotheses, reviewed the paper, and participated in the Am J Public Health 2003;93:1294–6.
26 Cannari L, D’Alessio G. La distribuzione del reddito e della ricchezza nelle
interpretation of results. GAC participated in the design of the regioni Italiane. Rome: Banca d’Italia, 2003.
manuscript and inspired the framing of the paper. 27 ISTAT. Indicatori Demografici. Rome: Istituto Nazionale di Statistica, 2003.
http://demo.istat.it/ind_dem2002.htm (accessed 10 Jan 2003).
28 Banca d’Italia. Survey on Italian households wealth and income (SIHWI).
The appendix is available on line (http://www.jech. Rome: Banca d’Italia, 2003.
com/supplemental). 29 ISTAT. Indicatori Sociosanitari, 1997. Rome: Istituto Nazionale di Statistica,
1998.
30 UNDP. Human Development Report 2003. New York: Oxford University
Press, 2003.
..................... 31 World Bank. World Development Indicators, 2003 [CD ROM]. Washington,
DC: World Bank, 2003.
Authors’ affiliations 32 Belsley D. Conditioning diagnostics: collinearity and weak data in regression.
R De Vogli, The University of California Los Angeles (UCLA), School of New York: Wiley, 1991.
Public Health, USA 33 Lynch JW, Davey Smith G, Kaplan GA, et al. Income inequality and mortality:
R Mistry, The University of California Los Angeles (UCLA), School of importance to health of individual income, psychosocial environment, or
Public Health material conditions. BMJ 2000;320:1200–4.
R De Vogli, R Gnesotto, The Epidemiology Centre of the Veneto Region, 34 Wilkinson R. Two pathways, but how much do they diverge? BMJ
1999;319:956–7.
Italy
35 Lobmayer P, Wilkinson R. Income inequality and mortality in 14 developed
G A Cornia, The University of Florence, Faculty of Economics, Italy countries. Sociology of Health and Illness 2000;22:401–14.
Funding: the work of RD and GAC was supported in part by the John D 36 Kennedy B, Kawachi I, Glass R, et al. Income distribution, socioeconomic
and Catherine T MacArthur Foundation Research Network Project on status and self-rated health in the United States: multilevel analysis. BMJ
1998;317:917–21.
Socioeconomic Status and Health.
37 Gravelle H. Diminishing returns to aggregate level studies. BMJ
Competing interests: none declared. 1999;319:955–6.

www.jech.com
 Downloaded from http://jech.bmj.com/ on January 18, 2018 - Published by group.bmj.com

Has the relation between income inequality and

life expectancy disappeared? Evidence from
Italy and top industrialised countries
Roberto De Vogli, Ritesh Mistry, Roberto Gnesotto and Giovanni Andrea
Cornia

J Epidemiol Community Health2005 59: 158-162

doi: 10.1136/jech.2004.020651

Updated information and services can be found at:

http://jech.bmj.com/content/59/2/158

These include:

Supplementary Supplementary material can be found at:

Material http://jech.bmj.com/content/suppl/2005/01/19/59.2.158.DC1
http://jech.bmj.com/content/suppl/2005/01/26/59.2.158.DC2
References This article cites 29 articles, 19 of which you can access for free at:
http://jech.bmj.com/content/59/2/158#ref-list-1

Email alerting Receive free email alerts when new articles cite this article. Sign up in the
service box at the top right corner of the online article.

ErrataAn erratum has been published regarding this article. Please see next page
or:
/content/59/6/531.3.full.pdf
Topic Articles on similar topics can be found in the following collections
Collections Health service research (832)

Notes

To request permissions go to:

http://group.bmj.com/group/rights-licensing/permissions

To order reprints go to:

http://journals.bmj.com/cgi/reprintform

To subscribe to BMJ go to:

http://group.bmj.com/subscribe/
J Epidemiol Community Health 2005;59:531 531

PostScript ..............................................................................................

problems in industrialised as well as devel- the contents date from 2002), the challenge
BOOK REVIEWS oping countries. This shift calls for urgency of globalisation, and how to solve social
and action to create an applied global inequalities in health. Ethics is given the
surveillance system that would permit com- detail it deserves, in relation to human rights.
Global behavioral risk factor parison of chronic disease risk factor trends, Some complex and difficult issues are ade-
surveillance similarities, and experiences across popula- quately discussed: how to prioritise public
tions. Surveillance of chronic disease risk health problems, how to identify and carry
factors is essential in planning and evaluating out cost effective strategies, and the resolu-
Edited by D McQueen, P Puska. Kuwer
disease prevention and health programmes tion of public health problems not from a
Academics/Plenum, 2003, $65.00, pp 255.
and policies. disease oriented approach but from a more
ISBN 0-306-47777-7
comprehensive point of view.
Qaiser Mukhtar In summary, this textbook continues to be
This book is a godsend to those working in
the area of monitoring and understanding an excellent reference for public health
changes overtime in chronic disease risk Oxford textbook of public health, practitioners, teachers, and students.
factors. It is a collection of 18 chapters 4th ed Miguel Delgado-Rodrı́guez
authored by experts around the world and
provides a comprehensive insight into estab-
Edited by R Detels, J McEwen, R Beaglehole,
lishing and maintaining the surveillance of
H Tanaka. Oxford: Oxford University Press,
behavioural risk factors both in developed
2002, pp 1955.
and developing countries. CORRECTIONS
The examples of behavioural risk factor The paperback edition of the Oxford Textbook of
surveillance systems in a variety of inter- Public Health has in one volume the same
national settings provide not only a glimpse contents as the hardback edition of 2002. The doi: 10.1136/jech.2004.020651
into the diversity of issues but also suggest price is a clear advantage over the hardback
creative solutions to these challenges. edition, less than one third of the latter, Three editorial errors occurred in this paper
Furthermore, what makes this book a prac- making the book affordable for practitioners by Dr R De Vogli and others (2005;59:158–
tical public health resource are topics such as and students. 62). In the third line of the legend to figure 1
analysis, interpretation, comparison, and use Compared with the hardback edition, there it should read r (crude) = 20.607 [not
of behavioural risk factor surveillance data. is nothing new in the contents of this edition. 0.607]. In the fifth line of the legend to
This book will serve as a guide for those new The volumes of the hardback edition are figure 1 it should read r (weighted by
to chronic disease surveillance but there is sections in this one volume edition. The first population size) = 20.658 [not 0.658]. In
plenty here for the seasoned public health section presents the scope of public health the fifth line of the legend to figure 2 it
professionals looking to hone their skills. I (22 chapters in four parts), the second should read r (crude) = 20.415 [not 0.415].
was pleased to see the reporting of the introduces the methods of public health (38
essence of the discussions that occurred at chapters in four parts), and the third details doi: 10.1136/jech.2003.019547corr1
the four global meetings on risk factor the practice of public health (41 chapters in
surveillance attended by international com- four parts). An authors’ error occurred in this paper by
munity members beginning in 1999. This book is an up to date text on the Dr J Doyle and others (2005;59:193–7). Dr
The authors have successfully shown that important topics of public health. It deals Kwok Cho Tang from the World Health
the global burden of disease is changing with emerging and re-emerging infectious Organisation, Geneva was omitted from the
and chronic diseases are important health diseases (although SARS is not featured as author list.

bmjupdates+

bmjupdates+ is a unique and free alerting service, designed to keep you up to date with the
medical literature that is truly important to your practice.
bmjupdates+ will alert you to important new research and will provide you with the best new
evidence concerning important advances in health care, tailored to your medical interests and
time demands.

Where does the information come from?

bmjupdates+ applies an expert critical appraisal filter to over 100 top medical journals
A panel of over 2000 physicians find the few ’must read’ studies for each area of clinical
interest

Sign up to receive your tailored email alerts, searching access and more…

www.bmjupdates.com

www.jech.com