Health Inequalities

1. Key messages

In Belgium, like in all EU countries, important socio-economic inequalities are observed in the whole spectrum of health indicators, starting from the health determinants to the health/disease status and ultimately the mortality. This report shows an overview of the extent of these inequalities using educational level as a marker for socio-economic position. Several of the socio-economic inequalities have increased over time.
People with a higher socio-economic status (SES) live longer. The gap in life expectancy (at age 25) between the highest and lowest educational levels is 6.1 years for men and 4.6 years for women. People with a higher SES also live longer in good health. The gap in health expectancy without disability (Healthy Life Years) between the highest and lowest educational levels is 10.5 years for men and 13.4 years for women. The gaps in health expectancy have increased over time.
People with a lower SES have a higher premature mortality rate. Men and women in the lowest educational category are, respectively, 1.9 and 1.6 times more likely to die before 75 years than men and women in the highest educational category.
Finally, people with a higher SES generally report better health and healthier behaviors. People in the lowest educational level rate almost 3 times more often their health as less than good compared with the highest educated people. People in the lowest educational level also report suffering from chronic diseases 1.5 to 2 times more often than people in the highest educational level. They also report a much higher prevalence of smoking and obesity, and poorer nutritional habits, such as insufficient consumption of fruits and vegetables and higher consumption of sugar-sweetened beverages.

2. Background

Socio-economic (SE) health inequalities refer to systematic disparities in health between SE groups, most often in disfavor of social groups lower on the social scale. SE health inequalities have been consistently observed throughout industrialized societies for the whole scope of health topics, ranging from health determinants to mortality [1;2]. Tackling health inequalities is a priority for the WHO [3], the European Union [4], and for Belgium [5-7]. In order to assess progress towards reducing health inequalities, it is important to measure and monitor them [8].

In this chapter, we provide an overview of the indicators from this Health Status Report in which inequalities were actually observed. Data from the 1997-2013 Health Interview Surveys, the 2014 Food Consumption Survey, and the censuses 2001 and 2011 linked with the mortality were used.

Preliminary to the inequality measurement, a marker of the socio-economic position has to be chosen: here, the educational level (EL) has been used as socio-economic variable (SE); it has been grouped in 3 or 4 levels, depending on the health indicator under study and the availability of the data. For the indicators extracted from the Health Interview Survey (HIS), EL was grouped in 4 levels: primary education or less, lower secondary education, higher secondary education and higher education. For life and health expectancy indicators and mortality rates, EL was grouped in 3 levels, pooling together the two lowest educational levels.

The measurement of the inequality requires 2 steps: firstly, the breakdown of the health outcomes by each level of the SES. Such breakdowns have been displayed on each specific indicator page. The second step is to measure the extent of the inequality which is the purpose of this chapter; here, we do more than simply displaying the breakdown of the health indicators by EL: we calculate the magnitude of the inequalities, using several inequality measures (see the 'Read More' section below for details). Depending on the nature of the indicator, the following inequality measures were computed:

  • For life and health expectancies:
    • The absolute difference in years between the lowest and the highest EL groups.
    • The absolute composite index of inequality (CIIabs): expresses the absolute differences in years among the whole population; it is also the number of years that could be gained, at population level, if there was no inequality and everybody had the life/health expectancy of the highest educated group
  • For mortality rates and the indicators extracted from the Health Interview Survey:
    • The absolute difference in age-adjusted rates between the lowest and the highest EL groups.
    • The relative difference in age-adjusted rates (or Rate Ratio) between the lowest and the highest EL groups.
    • The Population Attributable Fraction (PAF) corresponds to the percentage of gain in health expected in the whole population if all groups experienced the health of the more advantaged social group. 

By convention, for the calculation of relative inequalities across different indicators, those are expressed in term of adverse events [9] (for instance, self-rated health is expressed as the proportion of persons rating their health as less than good).

In this chapter, the emphasis is placed on health indicators with large inequalities. Therefore, beside the life and health expectancy disparities, for which the absolute differences were large but not the relative ones, we selected, for indicators expressed as a percentage, the ones having a relative risk (RR) of at least 1.4; this corresponds to a relative excess of bad health or unhealthy lifestyle of at least 40% in the lowest EL as compared to the highest EL. The indicators for which the values are not displayed here have smaller or no relative inequalities.

The evolutions in inequalities are only displayed when a significant change was observed.

3. Inequalities in life and health expectancies, and quality of life

For life expectancy at 25 years, based on the most recent census (2011), gaps of 6.1 and 4.6 years are observed between the lowest and the highest educational levels, respectively in men and women [10]. The CIIabs is 3.4 years in men and 2.4 in women (potential gain for the whole population if there was no inequality). For health expectancy at 25 years, the gap is even higher, reaching 10.5 years in men and 13.4 years in women. The CIIabs is 6 years in men and 7.2 years in women. In the case of life and health expectancies, relative measures are quite small, since they relate to large denominators.

Looking at the evolution between 2001 and 2011, a moderate increase of the LE gaps (+ 17% and 22% respectively in men and women) and a large increase of the health expectancy gaps (+61% and 45% in men and women) are observed (figure 1). When using the CII to account for the differences in the composition of the populations over time (figure 2), a small increase is observed in life expectancy CII among men (+ 7%), but not among women. The situation is different for health expectancy, where the population-level inequality index clearly increased among both men (+73%) and women (+20%).

Very large inequalities are observed in self-rated health (in 2013) with a prevalence of people rating their health as less than good being almost 3 times as high in low versus high educated people, and absolute inequalities reaching 25 percentage-point. The PAF reaches 31% (expressing the potential room for improvement at population level). No important inequality trend was observed for this indicator. The EQ-5D score of quality of life indicates also large SE differences, with a 17.6 gap on a scale of 100, between the lowest and the highest ELs.

Socio-economic inequalities in life expectancy and health expectancies and self-perceived, Belgium, 2011 or 2013.
Sources: (a) own calculations based on census 2011 linked with a 5-years' mortality follow-up, and EL grouped in 3 classes; (b) own calculations based on Health Interview Survey, with EL grouped in 4 classes, Sciensano, 2013

  Sex Lowest EL Highest EL Absolute difference
Relative difference
CIIabs PAF
Life expectancy at 25(a) Men 51.7 57.8 -6.1 - 3.4 -
Women 57.3 61.9 -4.6 - 2.4 -
Disability-free life expectancy at 25(a) Men 37.0 47.5 -10.5 - 6.0 -
Women 35.5 49.0 -13.4 - 7.2 -
Poor self-rated health (% people ≥ 15 yr)(b) Both 38.7% 13.5% 25.2% 2.87 - 31.1%
 
Evolution of the gaps between low and high educational levels for life expectancy and disability-free life expectancy, by sex, Belgium, 2001 and 2011
Source: Own calculations form censuses 2001 and 2011 linked with mortality
Evolution of population-level measure of educational absolute inequality in LE and DFLE, Belgium 2001 and 2011
Source: Own calculations based on censuses 2001 and 2011 linked with mortality, and HIS 2001 to 2013

4. Inequalities in premature mortality

Men in the lowest educational category are 1.9 times more likely to die before the age of 75 years than men in the highest educational category (after correction for age). Women in the lowest educational category are 1.6 times more likely to die prematurely than women in the highest educational category. Absolute differences are also very large, with excesses of 279 and 114 per 100.000 deaths respectively in men and women. The population attributable fraction is high (38% and 29% respectively in men and women), meaning that the premature mortality of the whole population could decrease by 38% in men if all ELs had a premature mortality rate as low as the one of the highest EL. The contribution to the population-level inequality in premature mortality was highest for Lung cancer, Ischemic heart diseases, suicide and Chronic Obstructive pulmonary diseases (COPD) in men, and for Ischemic Heart Disease, Lung Cancer, Cerebrovascular diseases and COPD in women [11]. This indicates the causes of deaths for which the inequality reduction would most benefit to the whole population, by reducing the global premature mortality level. This can suggest policy makers to prioritize tackling inequalities in mortality due to those specific causes.

Educational inequalities in premature mortality (<75 years), Belgium, 2001
Source: Own calculations based on census 2001 linked with a 5-years' mortality follow-up

Sex Lowest SES Highest SES Absolute difference Relative difference PAF
Premature mortality, all-cause (age-adj.rate per 100.000) Men 595.3 316.4 278.9 1.88 38.1%
Women 304.8 191.3 113.5 1.59 29.2%
 
  • Men
  • Women

Ranking of the causes of deaths by their contribution to the inequalities in premature mortality, measured as Population attributable fraction, Belgium, 2001
Source: Own calculations based on census 2001 linked with a 5-years' mortality follow-up

Ranking of the causes of deaths by their contribution to the inequalities in premature mortality, measured as Population attributable fraction, Belgium, 2001
Source: Own calculations based on census 2001 linked with a 5-years' mortality follow-up

5. Inequalities in morbidity

Relative inequalities are quite large for all considered indicators on non-communicable diseases, with low-versus-high risk ratios generally between 1.5 and 2. The highest risk ratios are observed for the self-reported prevalence of multi-morbidity (1.9) and diabetes (1.8). The largest population-attributable fractions are observed for diabetes and thyroid disorder. Important absolute inequalities are observed for the proportion of people who reported suffering from at least one chronic diseases (as a whole), for multi-morbidity and arthrosis (between 9 and 13 percentage-points).

For the three mental health disorders studied here (depression, anxiety and sleeping problems), relative inequalities are even larger, with risk ratios situated around 3 for depression and anxiety disorders. The inequality in the use of tranquillizers or sleeping pills is consequently also quite large. For all three disorders, the absolute inequalities are also large, reaching 15 percentage-points.

When looking at the evolution over time, absolute inequalities have increased for all three mental health disorders, while the relative inequalities have increased for anxiety only. This means that for depression and sleeping problems, the absolute prevalence has increased more among low than among highly educated people, but the proportional increase was the same in each EL. For anxiety, the increase in low EL exceeded the increase in high EL people in such an important way that both absolute and relative inequalities increased.

Educational inequalities in selected self-reported non-communicable disease prevalences, people aged 15 years and over, Belgium, 2013
Source: Own calculations based on Health Interview Survey, Sciensano, 2013

Lowest SES Highest SES Absolute difference Relative difference PAF
Non-communicable diseases     
Chronic disease 36.3% 22.9% 13.4% 1.59 13.6%
Multi-morbidity* 21.9% 11.6% 10.3% 1.89 15.8%
Diabetes 7.7% 4.2% 3.5% 1.84 22.9%
Osteoarthrosis 23.0% 14.1% 9.0% 1.64 17.3%
High blood pressure 21.3% 14.9% 6.5% 1.44 11.3%
Thyroid disorder 7.8% 4.6% 3.2% 1.69 21.4%
Mental health     
Depressive disorders 26.6% 9.8% 16.8% 2.71 31.5%
Anxiety disorders 21.3% 6.7% 14.6% 3.18 29.5%
Sleeping problems 40.8% 25.7% 15.1% 1.59 12.0%
Consumption of tranquillizers or sleeping pills 20.0% 10.1% 10.0% 1.99 24.9%
Suicidal ideation, past 12 months 5.5% 3.9% 1.6% 1.41 17.0%
*Multi-morbidity is defined as having at least 2 of the non-communicable diseases included in the Belgian Health Interview Survey.
  • Absolute
  • Relative

Absolute low-versus-high EL inequalities in mental health indicators, Belgium, 2001-2013
Source: Own calculations based on Health Interview Survey, Sciensano, 2001-2013

Relative low-versus-high EL inequalities in mental health indicators, Belgium, 2001-2013
Source: Own calculations based on Health Interview Survey, Sciensano, 2001-2013

6. Inequalities in health determinants

Inequalities in smoking behavior are large, with an absolute rate difference between lowest and highest EL of 15 percentage-points, and a relative difference reaching 2.6. Both absolute and relative differences have increased over time. People with the highest EL have considerably modified their smoking behavior, while the proportion of daily smokers remained stable over time among the lower ELs. Inequalities are even larger for the prevalence of obesity, with different evolutions in men and women. From 2008 to 2013, both absolute and relative inequalities have increased substantially among men. The increase in absolute inequality means that the increase of obesity prevalence in low EL men exceeded the increase in high EL men, and the increase on relative inequalities means that this excess was such important that even the proportional increase was higher in low than in high educated men. At the same time, for women, only absolute inequalities have steadily increased but relative inequalities remained stable, meaning that while obesity prevalence increased more in the lowest than in the highest EL, the proportional increase was the same in each group.

Poor nutritional habits are frequent in all ELs, but they are much worse in the lowest EL. For instance, the consumption of fruits and vegetables is much lower among the people in the lowest EL, who on average consume per day 90 g less fruits and vegetables than people in the highest EL. Only 14% of the population meets the recommendation to eat 400 g of fruits and vegetables per day; this proportion is as low as 6% among the lowest educated people versus 22% in the highest level, leading to an absolute rate difference of 16 percentage-point, and a rate ratio of 0.27, meaning that people of the lowest EL reach the recommended quantity 4 times less frequently than people of the highest EL.

Sugar-sweetened beverage consumption is 2.5 times higher in the lowest EL category compared with the highest one.

The important inequalities observed for these health determinants lead to very large population-attributable fractions: this means that an important improvement would result, at population level, if everybody could reach the value observed in the most advantaged group.

No systematic educational differences were observed for physical activity and alcohol consumption.

Socio-economic inequalities in some lifestyles and health determinants, Belgium, 2013/14
Sources: Health Interview Survey, Sciensano, 2013 and Food Consumption Survey, Sciensano, 2014

Lowest SES Highest SES Absolute difference Relative difference PAF
Daily smoking (%people ≥ 15 yr) 25.0% 9.5% 15.5% 2.63 45.1%
Obesity (%people ≥ 18 yr, BMI ≥ 30) 23.5% 7.7% 15.8% 3.05 37.4%
Fruits and vegetables consumption (g/day) 210 300 -90 - -
Fruits and vegetables consumption (% eating 400 g a day) 6.0% 22.4% -16.4% 0.27 61.2%
Sugar-sweetened beverage consumption (g/day) 227.3 89.5 137.8 - -
 
  • Absolute
  • Relative

Absolute low-versus-high EL inequalities in health determinants, Belgium, 1997-2013
Source: Own calculations based on Health Interview Survey, Sciensano, 1997-2013

Relative low-versus-high EL inequalities in health determinants, Belgium, 1997-2013
Source: Own calculations based on Health Interview Survey, Sciensano, 1997-2013

7. Read more

View the metadata for this indicator

Definitions

Socio-economic position
A number of indicators can be used to assess SES because the position of an individual within the social hierarchy may be determined by many dimensions such as occupation, income or education. The educational level (EL) has been used here as socio-economic variable (SE). The EL is measured using a standardized international scale (ISCED), then recoded into 3 or 4 levels, depending on the health indicator under study and the availability of the data.
International Standard Classification of Education (ISCED)
ISCED is the reference international classification for organizing education programs and related qualifications by levels and fields. It contains categories from 0 to 6:
  • 0: Early childhood education ('less than primary')
  • 1: Primary education
  • 2: Lower secondary education
  • 3: Upper secondary education
  • 4: Post-secondary non-tertiary education
  • 5: Short-cycle tertiary education, Bachelor’s, Master’s
  • 6: Doctoral or equivalent level
Inequality indices
Inequality indices are measurement of the magnitude of inequalities. A number of inequality indices have been described in the literature [12;13]. It is currently acknowledged, that the optimal approach to evaluate and track inequalities is through relying on a set of inequality measures rather than on a single measure because the measures of inequalities differ substantially in at least two important aspects [14]:
  • The relative vs. absolute nature of the comparison: to assess health inequalities, relative measures of inequalities (e.g. rate ratios) have been more commonly used in the literature, but inequalities can also be presented in absolute terms (e.g. rate difference) to account for the prevalence of the outcome in each group. Both measures are important, some authors according even more importance to the absolute measures in term of public health and in term of people’s point of view.
  • The scope: are simple pairwise comparison of health outcome between two SE groups, like the “absolute rate difference” or the “rate ratios”; more complex indices take also into account the distribution of the SES in the population, like the population attributable fraction (PAF), the composite inequality index (CII).
Composite Index of Inequality, absolute version (CIIabs)
The CII_abs expresses the absolute differences in years among the whole population; it is also the number of years that could be gained, at population level, if there was no inequality and everybody had the life/health expectancy of the highest educated group; is obtained by summing up the differences in life(health) expectancy of each EL group as compared to the group with the highest EL, weighted by the size of each group
Population Attributable Fraction (PAF)
The PAF corresponds to the relative gain in health (or health determinant) level that would be expected for the whole population if all groups experienced the value of the more advantaged social group. It is computed as the difference between the overall value in the population and the value in the more advantaged group, divided by the overall value in the population.
Premature mortality
Premature mortality is defined here as mortality occurring before the age of 75.
Age-standardized rates
Age-standardization is a method of adjustment of the rates (for instance mortality or prevalence rates), to eliminate the effects of differences in the age structure of different population. The most usual standardization (the direct standardization) computes a weighted average of all the age-specific rates, with the weights being the proportion of each age group in a reference population.

References

  1. Mackenbach J. Health inequalities: Europe in profile. Expert Report commissionned by the EU. Department of Health Publications; 2006.
  2. Feinstein JS. The relationship between socioeconomic status and health : A review of the literature. The Milkbank Quarterly 1993;71:279-94.
  3. WHO Commission on Social Determinants on Health. Closing the gap in a generation: health equity through action on the social determinants of health. Geneva: WHO; 2008.
  4. Executive Agency for Health and Consumer. Second Programme of Community Action in the Field of Health 2008-2013. European Commission; 2007.
  5. Vlaamse overheid. Vlaams Actieplan Geestelijke Gezondheid, Strategisch plan 2017-2019. 2017.
  6. Gouvernement wallon. Plan prévention et promotion de la santé en Wallonie. Partie 1: définition des priorités en santé. Namur: 2017.
  7. Arrêté royal du 18 juillet 2013 portant fixation de la vision stratégique fédérale à long terme de développement durable: http://www.etaamb.be/fr/arrete-royal-du-18-juillet-2013_n2013011468.html. Moniteur Belge 2013 Oct 8.
  8. Braveman PA. Monitoring equity in health and healthcare: a conceptual framework. J Health Popul Nutr 2003 Sep;21(3):181-92.
  9. Keppel K, Pamuk E, Lynch J, Carter-Pokras O, Kim I, Mays V, et al. Methodological issues in measuring health disparities. 141 ed. Washington DC: National Center for Health Statistics; 2005.
  10. Renard F, Devleesschauwer B, Van Oyen H, Gadeyne S, Deboosere P (2019) Evolution of educational inequalities in life and health expectancies at 25 years in Belgium between 2001 and 2011: a census-based study. Arch Public Health 77:6. doi: 10.1186/s13690-019-0330-8
  11. Renard F, Gadeyne S, Devleesschauwer B, Tafforeau J, Deboosere P. Trends in educational inequalities in premature mortality in Belgium between the 1990s and the 2000s: the contribution of specific causes of death. J Epidemiol Community Health 2017;71(4):371-80. doi: 10.1136/jech-2016-208370
  12. Mackenbach JP, Kunst AE. Measuring the magnitude of socio-economic inequalities in health: an overview of available measures illustrated with two examples from Europe. Soc Sci Med 1997 Mar;44(6):757-71.
  13. Harper S, Lynch J. Methods for Measuring Cancer Disparities:Using Data Relevant to Healthy People 2010 Cancer-Related Objectives. Bethesda, MD: National Cancer Institute, 2005: NIH Publication No. 05-5777.; 2005.
  14. Mackenbach JP, Kunst AE. Measuring the magnitude of socio-economic inequalities in health: An overview of available measures illustrated with two examples from Europe. Soc Sci Med 1997;44(6):757-71.