Your shopping cart is empty.

Inequalities in the distribution of health and health care and its access, including inequalities relating to social class, gender, culture and ethnicity, and their causes

Equality, Equity and Policy: Inequalities in the Distribution of Health and Health Care and its Access

The distribution of health is determined by a wide variety of individual, community, and national factors (See Figure 1). There is a growing body of evidence documenting inequalities in both the distribution of health (i.e. health outcomes) and access to health care both internationally and in the UK. Access to health care is a supply side issue indicating the level of service which the health care system offers the individual.

Figure 1: Determinants of health

4c10

Inequalities in the distribution of health

Researchers have documented inequalities in the distribution of health by social class, gender, and ethnicity. Inequalities in health have been measured using many different outcomes including infant deaths, mortality rates,
morbidity, disability, and life expectancy.

Social class (including income, wealth and education)

Research on socio-economic inequalities in health in the UK has a long history. In the early part of the 20th century the British government introduced questions on occupation in the decennial census. This allowed researchers to examine health outcomes by social class. The five class scheme Registrar General’s Social Class (RGSC) was created in 1911 and a variation of this scheme was still used until recently. In 2001, the National Statistics Socio-Economic Classification (NS-SEC) replaced the RGSC. For a description of the new scheme see http: /www.statistics.gov.uk

Table 1: Classifications of Social Classes.

RGSC

NC-SEC

I  Professional occupations

1 Senior professionals/senior managers

II Managerial and technical occupations

2 Associate professionals/junior managers

III Skilled occupations

3 Other administrative and clerical workers

  manual (M) and non-manual (N)

4 Own account non-professional

IV Partly-skilled occupations

5 Supervisors, technicians and related workers

V Unskilled occupations

6 Intermediate workers

 

7 Other workers

 

8 Never worked/other inactive

The 1970-1972 Decennial Supplement of occupational Mortality (OCPS) showed that men in social class V (unskilled) were 2.5 times as likely to die before the age of 65 than those in social class I (professional). Children in social class V families were twice as likely to die as those in social class I.

Table 2 shows more recent data on the relationship between social class and death,

Bartley and Blane (2008).

Table 2: Social class and health, 1991-1993 and 1993-1995

Social Class

Still-birth rate

Infant mortality rate

Mortality rate

(1-15 years)

Standardised mortality ratio (men 20-64 years)

I

4

4

18

66

II

4

5

16

72

IIIN

5

5

16

100

IIIM

5

6

26

117

IV

6

7

22

116

V

8

8

42

189

 

N=non-manual; M=manual

Still birth rate = number of deaths per 1000 live and death births, 1993-5

Infant mortality rate = number of deaths in the first year of life per 1000 live births, 1993-5

Mortality rate (1-15 years) = number of deaths per 100,000 population aged 1-15 years, 1991-3

Standardised mortality ratio (men 20-64 years) = The ratio of the observed mortality rate in a social class to its expected rate from the total population, multiplied by 100, 1991-3

Source: Bartley and Blane, 2008

Not only do social class inequalities in health still exist, they appear to be increasing over time (DHHS, 1980). Social class inequalities in the UK are present at every age and for all the major diseases.

There are four major models used to explain social class inequalities in health (Bartley and Blane, 2008; Bartley, 2004).

  1. Behavioural model: There are social class differences in health damaging or health promoting behaviours such as dietary choices, consumption of drugs, alcohol and tobacco, active leisure time pursuits, and use of immunisation, contraception and antenatal services. However, long-term studies (like the Whitehall study described below) have found that differences in health behaviour explain only one-third of social class differences in mortality. Furthermore, evaluations of interventions that seek to change health behaviours have rarely found clear cut improvements in health that would be predicted by the behavioural model.

  2. Materialist model: Poverty exposes people to health hazards. Disadvantaged people are more likely to live in areas where they are exposed to harm such as air-pollution and damp housing.  The Black Report (see below) found materialist explanations to be the most important in explaining social class differences in health. There is some specific evidence for materialist explanations. For example, many studies have associated higher rates of childhood respiratory disease with damp housing. The full impact of living standards, however, can only be understood over the course of the life term. While most experts in public health agree that materialist explanations play a role in explaining health inequalities, many find a simple materialist model to be insufficient. In the UK, relatively disadvantaged people receive various kinds of state help (rent, school meals etc) which, some argue, makes diet or poor housing unlikely to account for all inequalities health outcomes. Furthermore, in the UK and internationally, inequalities in health tend to follow a steady gradient, rather than there being poor outcomes for the most disadvantaged and equally good outcomes for the rest of society.

     

  3. Psycho-social model: Social inequality may affect how people feel which in turn can affect body chemistry. For example, stressful social circumstances produce emotional responses which bring about biological changes that increase risk of heart disease. Psycho-social risk factors include social support, control and autonomy at work, the balance between home and work, and the balance between efforts and rewards. There has been a plethora of research exploring associations between psycho-social factors and health. Evidence shows that people who have good relationships with family and friends, and who participate in the community, have longer life expectancies than those who are relatively isolated. Evidence of an association between stress at work and health is less clear, but most well designed studies show a higher risk of heart disease among individuals who work in jobs where demands are high and control is low. Furthermore, a number of studies have shown that an imbalance between effort and reward at work tends to be linked to high blood pressure, fibrinogen and a more adverse blood fat profile.

     

  4. Life-course model: Health reflects the patterns of social, psycho-social and biological advantages and disadvantages experiences by an individual over time. The chances of good or poor health are influenced by what happened to a child in-utero and in early childhood. Disadvantages are likely to accumulate through childhood and adulthood. For example, individuals who experienced poor home conditions in childhood are more likely to experience occupational disadvantage. The life-course model was developed relatively recently and studies investigating life-course explanations require detailed longitudinal data. Regardless, several studies have shown that health disadvantage accumulates over time.

Landmark studies in social class inequalities in health in the UK include:

The Black Report

             

The Black Report, published in 1980 confirmed social class health inequalities in overall mortality (and for most causes of death) and showed that health inequalities were widening. The report set out four possible mechanisms to explain widening socio-economic health inequalities:

Artefact: Population information came from the decennial census while death and cause of death information came from death certificates. An individual may have been described in different ways in the two data sources leading to numerator-denominator bias. The report also noted widening inequalities may be explained by the shrinking of social class V. With fewer people who were completely unskilled, the average health of social class V moved further from social class I. Furthermore, the report noted that the meaning of social class may have changed over time as some jobs disappear and others emerge.

Social selection: Health determines social position. Somewhat similar to Darwin’s ‘natural selection’, i.e. healthy people are more likely to get promoted while unhealthy people are more likely to lose their jobs.

Behaviour: individuals in the lower social classes indulge in comparatively more health damaging behaviour (see behavioural model above).

Material circumstances: poverty causes poor health (see materialist model above).

Whitehall Study of British Civil Servants

The ongoing Whitehall Study of British Civil Servants http://www.ucl.ac.uk/whitehallII/ is a cohort study following British civil servants over a long period of time. It collects detailed information on risk factors such as weight, cholesterol, smoking, and blood pressure. The study found inequalities in health and mortality between employment grades and found that risk factors could only explain one-third of the observed variation in health by employment grade.

The Acheson Report

The Acheson Report published in 1988 found that ,ortality had decreased in the last 50 years but that inequalities in health remained, and in some instances health inequalities had widened. The report recommended

  1. evaluating all policies likely to affect health in terms of their impact on inequalities

  2. giving high priority to the health of families with children

  3. the government should take steps to reduce income inequalities and improve living conditions in poor households.

Gender

Much research has shown that in industrialised countries women live longer than men (table 3) but appear to experience more ill health. While men have higher mortality from the most common single causes of death (ischemic heart disease and lung cancer), more women than men suffer from somatic complaints such as tiredness, headache, muscular aches and pains. However, some researchers have raised questions about the validity of studies that show higher illness rates in women, as many different health outcome variables have been used and not all show gender differences. There is more consistency in studies that examine minor psychological illness, anxiety, sickness absence from work, functional limitation, and depression (Bartley, 2004).

Table 3: Selected developed countries by order of female to male difference (in years) of life expectancy at birth in 1980 and 1996

1996

1980

Country

Female-Male difference

Ranking

Female-Male difference

Ranking

United Kingdom

4.9

1

6

4

Sweden

5

2

6

4

Denmark

5.2

3

6.1

6

Greece

5.3

4

4.6

1

Ireland

5.3

5

5.5

3

Netherlands

5.6

6

6.6

8

United States

6

7

7.4

12

Austria

6.3

8

7.1

11

Italy

6.4

9

6.8

10

Japan

6.5

10

5.3

2

Germany

6.5

11

6.8

9

Spain

7.2

12

6.1

7

Finland

7.5

13

9.1

14

France

7.8

14

8.2

13

Mean

6.2

6.6

 

 

Standard Deviation

0.987

1.156

 

 

Range

3

4.5

 

Source: Gjonca et al, 1999.

In the UK, mortality is greater in males than in females at all ages. In youth and early adulthood, males are more likely to die from motor vehicle accidents, other injury (such as fire and flames, accidental drowning and submersion), and suicide, contributing to higher mortality rates among young men and boys. Across the whole of adult life, mortality rates are higher for men than women for all the major causes of death including cancers and cardiovascular disease. However, women have much higher rates of disability than men, especially at older ages. Women have more morbidity from poor mental health, particularly those related to anxiety and depressive disorder (Acheson, 1998). 

WHO (2008) suggests that gender differences in health are a result of both (1) biology and (2) social factors (distinct roles and behaviours of a men and women a given culture, dictated by that cultures gender norms and values).

Social factors used to explain higher mortality rates in men (Scambler, 2008):

  • Employment: More occupations typically followed by men involve direct risk to life (such as dangerous machinery, weather, environmental hazards, and exposure to toxic chemicals).

     

  • Risk taking behaviour: Men are more socialized to participate in dangerous sports like motorbike racing, rock climbing etc. Men are at higher risk of road traffic injury and tend to drive more and faster when under the
    influence of alcohol compared to women.

     

  • Smoking: In the past, men had much higher smoking rates than women. However, the gender gap between men and women in smoking has narrowed in recent years and young girls (<15) are now more likely than boys to smoke.

     

  • Alcohol: Men drink significantly more than women in all age groups and are more likely than women to exceed their recommended daily alcohol intake.

Ethnicity and Culture

There is a growing body of evidence documenting ethnic inequalities in health outcomes in the UK, and internationally, despite difficulties with the conceptualisation and measurement of ethnicity as an epidemiological variable (see Box 1).

Box 1: Difficulties with the conceptualisation and measurement of ethnicity in health research.

Ethnicity is a fluid concept and takes on different meanings in different contexts. For example, a person may be considered (or consider his/herself) Pakistani when filling out the UK Census. The same person may be considered Asian on the US census or South Asian on other UK surveys. The definition of ethnicity is influenced by both historical value systems and the current social and political context (Bradby, 2003). Definitions of ethnicity change, but are likely to involve dimensions of race, skin colour, language, religion, nationality, country of origin, and ‘culture’. Each of these dimensions may have implications for health.  A major limitation of the concept of ethnicity in practice is that research specific definitions are often not clearly stated. Bhopal (1997) claims that ethnicity is “a euphemism for race”. Indeed, in a four year review of the literature, Comstock and colleagues (2004) found that researchers “frequently failed to differentiate between the concepts of race and ethnicity”.

There are a number of concerns about the reliability and validity of measurements of ethnicity. Researcher-assigned ethnic identities may not match respondent self-defined identities, threatening validity. Even when ethnicity is self-identified, the same person may use different ethnic identities in different situations at different times, compromising reliability. Fixed response categories such as those found in the UK Census and many other quantitative surveys have particular validity concerns. Bradby (2003) notes that the lack of theoretical coherence in defining fixed-response categories is a major problem in ethnicity related research. This has led some observers to describe data collection in the UK as ‘ad-hoc’ (Sheldon, 1992). While fixed response categories facilitate comparisons over time, and potentially across surveys, mutually exclusive groups cannot reflect mixed ethnic identities. Furthermore, fixed response categories such as ‘black’, ‘white’, or ‘Asian’ may mask considerable within-group differences and emphasise between-group differences. Ellison (2005) notes that the validity and reliability of ethnicity data depend on measurement techniques as well as the population. Broad categories, objective techniques and group homogeneity can improve validity and reliability of ethnicity measurement. Furthermore, qualitative research into ethnic identification and monitoring of open-ended ‘other-specify’ survey responses may help to define more accurate fixed-response categories (Aspinall, 1997).

These limitations of measurement, and the changing multidimensional nature of ethnicity, mean that quantitative researchers may never have a totally unbiased ethnicity variable. However, taking account of the methodological limitations and social context, these variables can be useful as a proxy for the complex concept of ethnicity (Ellison, 2005).

Ethnicity is not recorded on UK death certificates, and mortality data uses country of birth as a proxy, thus failing to identify ethnic minorities born in the UK.

There are some repeatedly documented findings on ethnic inequalities in mortality (Kelly & Nazroo, 2008):

  • Men and women born in the Caribbean have high rates of mortality from stroke. Men born in the Caribbean have low rates of mortality overall and low rates of mortality from coronary heart disease.

     

  • Individuals born in West/South Africa have high overall mortality rates, high mortality rates from stroke, but low mortality rates from coronary heart disease.

     

  • Individuals born in South Asia have high mortality rates form coronary heart disease and stroke.

     

  • Non-white migrant groups tend to have lower mortality rates from respiratory disease and lung cancer but higher mortality rates for conditions relating to diabetes.

Table 4: Standardised mortality ratios by country of origin, England and Wales, 1989-1992.

Cause of death

All

Coronary heart disease

Lung cancer

Breast cancer

Men

Women

Men

Women

Men

Women

Women

All

100

100

100

100

100

100

100

Scotland

132

136

120

130

149

169

114

Ireland

139

120

124

120

151

147

92

East Africa

110

103

131

105

42

17

84

West Africa

113

126

56

62

62

51

125

Caribbean

77

91

46

71

49

31

75

South Asia

106

100

146

151

45

33

59

Source: Wild and McKeigue (1997:705) in Bartly (2004)

Combining national origin data with data on social class (which is only available for men because social class is poorly recorded on women’s death certificates), Bartley (2004) reports that the relatively high mortality in men
born in Scotland, Ireland, and South Asia is only seen outside of social classes I and II.

Explanations for ethnic inequalities in health include:

  • They are a statistical artefact.

  • They are a consequence of the migration process.

  • They are due to genetic/biological differences between ethnic groups.

  • They are due to differences in culture and health behaviours.

  • They are a consequence of socioeconomic disadvantage.

  • Experiences of racism result in health differences.

Inequalities in health care and its access

Health care access is a supply side issue indicating the level of service which the health care system offers the individual. While the concept of equity in access to health care (horizontal equity) has been a central objective of the
NHS since it began, inequalities in health care access persist. The inverse care law, first described by Julian Tudor Hart in 1971, states: The availability of good medical care tends to vary inversely with the need for it in the population
served.

Equality of access requires that, for different communities (Wonderling et al, 2005):

  • Travel distance to facilities is equal.

  • Transport and communication services are equal.

  • Waiting times are equal.

  • Patients are equally informed about the availability and effectiveness of treatments.

  • Charges are equal (with equal ability to pay)

Many studies investigating access to health care use treatment received (i.e. utilisation) as a proxy for access. However, utilisation of health services may vary for many several reasons (such as perceptions of benefits or availability, availability of alternative therapies or services) and is an imperfect measure of access. Nonetheless, it is commonly used as such.

Goddard and Smith (2001) outline reasons for variations in access to health care:

Availability: Some health care services may not be available to some population groups, or clinicians may have different propensities to offer treatment to patients from different population groups, even where they have
identical needs.

Quality: The quality of services offered to patients may vary between population groups.

Costs: The health care services may impose costs (financial or otherwise) which vary between population groups.

Information: The health care organisations may fail to ensure that all population groups are equally aware of the services available.

 

References

  • Acheson D (1998). Independent inquiry into inequalities in health report. London: The Stationary Office.

  • Aspinall PJ (1997). “The conceptual basis of ethnic group terminology and classifications” Social Science and Medicine,45(5) 

  • Bartley M, Blane D (2008). Inequality and social class in Scambler G (ed) Sociology as applied to medicine. Elsevier Limited.

  • Bartley M (2004). Health inequality: an introduction to theories, concepts, and methods. Cambridge: Polity Press.

  • Bhopal R. (1997). “Is research into ethnicity and health racist, unsound, or important science?” BMJ, 314.

  • Bradby H. (2003) “Describing ethnicity in health research.” Ethnicity and Health, 8(1).

  • Comstock RD, Castillo EM, Lindsay SP (2004). “Four-year review of the use of race and ethnicity in epidemiologic and public health research” American Journal of Epidemiology. Vol. 159, No. 6.

  • Dalgren G (1995). European Health Policy Conference. Opportunities for the Future Vol 1-Intersectorial Action for Health, Copenhagen: WHO Regional Office for Europe.

  • Department of Health and Human Services (DHHS) (1980). Inequalities in health: report of a research working group. (The Black Report). HMSO, London.

  • Ellison, GTH (2005). “Population profiling and public health risk: when and how should we use race/ethnicity? Critical Public Health, 15(1).

  • Goddard M, Smith P (2001). “Equity of access to health care services: theory and evidence from the UK”. Social Science and Medicine 53:1149-62.

  • Gjonça A, Tomassini C, Vaupel J (1999). Male–female Differences in Mortality

  • in the Developed World. MPIDR Working Paper WP 1999-009.

  • Kelly M, Nazroo J (2008). Ethnicity and Health in Scambler G (ed) Sociology as applied to medicine. Elsevier Limited.

  • Scambler A (2008). Women and Health in Scambler G (ed) Sociology as applied to medicine. Elsevier Limited.

  • Sheldon TA, Parker H. (1992) “Race and ethnicity in health research.” Journal of Public Health Medicine,14(2).

  • WHO (2008) Why gender and health? http://www.who.int/gender

  • Wild S, McKeigue P (1997). “Cross sectional analysis of mortality by country of birth in England and Wales”. BMJ, 314:705.

  • Wonderling D, Gruen R, Black N (2005) Introduction to Health Economics. Understanding Public Health Series. Open University Press: London School of Hygiene and Tropical Medicine.

© Rebecca Steinbach 2009