Scottish Health Survey - topic report: equality groups

1 Introduction

1.1 Background

Data on equality and its relationship to health is important because it helps us understand the characteristics of individuals which may have an impact on health and the need for health care services. To date, although we have some evidence to suggest how a person's characteristics can be associated with their health and care, this is limited in Scotland due to small size of some equality groups which makes it difficult to get representative information from surveys.

The analysis in this report represents an important step forward in the availability of survey data on health behaviours and health characteristics for equality groups in Scotland. Scottish Health Survey data has been combined from four consecutive years (2008-2011) in order to allow more in-depth analysis of smaller populations which would not be possible for individual survey years.

The UK Government's Equality Act 2010 includes a new public sector Equality Duty which requires public authorities to be active in promoting equality, eliminating unlawful conduct and fostering good relations. The new Duty came into force on 6 April 2011 and provides a single, consistent framework covering age, disability status, ethnicity, gender/sex, marriage and civil partnership, pregnancy and maternity, religion/belief, sexual orientation, and transgender identity¹. A public sector duty to provide evidence of compliance with this legislation and to set and report on equality outcomes came into force in Scotland in May 2012². All public sector bodies will require good quality information about equality groups to set and measure progress on equality outcomes. The analysis in this report should provide a useful contribution to the evidence base for setting national equality outcomes, which can then be drawn upon to inform local priorities where appropriate.

1.2 Equality Groups in the Scottish Health Survey

The equality groups considered in this report are gender, age, ethnicity, religion, disability and sexual orientation. Whilst data on gender and age is readily available and easily disaggregated, despite combining survey data for four years the sample sizes for the other equality characteristics are still relatively small, limiting the amount of analysis that can be undertaken.

1.2.1 Gender

Gender inequalities in Scotland and the UK have been well-documented in relation to income and employment.³ However, gender equality is also about the differing roles that men and women occupy within Scottish society: 90% of single parents are female and women still represent the majority of informal carers.⁴

In the 2008-2011 Scottish Health Surveys, 48% of respondents were men and 52% were women. These figures match the 2011 mid-year population estimates produced by National Records of Scotland.⁵

1.2.2 Age

Younger and older age groups are more likely to be disadvantaged in some way. Scotland has one of the highest teenage pregnancy rates in Europe and, amongst the working population in Scotland, many 18-24 year olds say that they are unable to cope financially. Smoking, drinking and poor diet start young for many Scottish men and women as identity-affirming habits. Although the young body is resilient, the effects of these behaviours are felt by the 45-55 age group.⁶

Older people in Scotland are less likely than the younger age group to display unhealthy behaviours such as smoking and excessive drinking; however their physical activity levels are lower than those of their younger counterparts. Poor oral health and limiting long-term conditions are more common in the older population.⁷

In the 2008-2011 Scottish Health Surveys, 14% of respondents were aged 16-24, with similar proportions in the next four age groups (ranging between 15% and 18% between the ages of 25-34 and 54-64. There were slightly less people aged 65-74 (11%) and fewer aged 75 and over (9%). These figures broadly matched the 2011 mid-year population estimates produced by National Records of Scotland.⁵ They differed very slightly due to the survey data covering four years and the population estimates covering one year.

1.2.3 Ethnic Group

There is some evidence to suggest that ethnic group is associated with health and healthcare outcomes. For example, people from minority ethnic groups generally have lower mortality than the general population in Scotland and there is a higher prevalence of heart disease and diabetes among those in the South Asian population.⁶

The question used to determine respondents' ethnic group was changed after the 2008 survey to match the harmonised ethnicity question which was being developed for the 2011 Census.⁸

In 2008, survey participants were asked which of the following groups they considered that they belonged to (only one could be chosen):

• White: Scottish
• White: Other British
• White: Irish
• White: Any other white background
• Mixed: Any mixed background
• Asian, Asian Scottish or Asian British: Indian
• Asian, Asian Scottish or Asian British: Pakistani
• Asian, Asian Scottish or Asian British: Bangladeshi
• Asian, Asian Scottish or Asian British: Chinese
• Asian, Asian Scottish or Asian British: Any other Asian background
• Black, Black Scottish or Black British: Caribbean
• Black, Black Scottish or Black British: African
• Black, Black Scottish or Black British: Any other black background
• Any other ethnic group

From 2009 - 2011, the question was changed slightly to ask respondents which of the following best describes their ethnic group or background:

• A - White: Scottish
• A - White: English
• A - White: Welsh
• A - White: Northern Irish
• A - White: British
• A - White: Irish
• A - White: Gypsy/Traveller
• A - White: Polish
• A - White: Any other white ethnic group
• B - Mixed: Any mixed or multiple ethnic groups
• C - Asian: Pakistani, Pakistani Scottish or Pakistani British
• C - Asian: Indian, Indian Scottish or Indian British
• C - Asian: Bangladeshi, Bangladeshi Scottish or Bangladeshi British
• C - Asian: Chinese, Chinese Scottish or Chinese British
• C - Asian: Other Asian ethnic group
• D - Black: African, African Scottish or African British
• D - Black: Caribbean, Caribbean Scottish or Caribbean British
• D - Black: Black, Black Scottish or Black British
• D - Black: Other Black ethnic group
• E - Other ethnic group: Arab
• E - Other ethnic group: other

In order to combine the data for 2008-2011, some of the categories had to be amalgamated where there were mismatches across the two questions. This related specifically to the White and Other ethnic group categories. In addition, some categories had to be combined where there were too few respondents to allow meaningful analysis. Black African, Black Caribbean and Black Other had to be combined into one category. In addition, the Asian Bangladeshi category was combined with Asian Other. Decisions on which groups to combine were based on analysis of the data and groups were combined where they showed similar results in terms of health behaviours / outcomes.

The ethnic group categories which have been used in this report are:

• White, British
• White, Irish
• White, Other
• Mixed
• Asian, Indian
• Asian, Pakistani
• Asian, Chinese
• Asian, Other
• African, Caribbean or Black
• Other

In the 2008-2011 Scottish Health Surveys, the combined non-white population accounted for only 3% of the total sample. The majority of respondents were white (97%). The largest non-white ethnic groups were Pakistani (0.7%), Indian (0.5%), Asian other and mixed (both 0.4%) and Chinese (0.3%). There were very few Bangladeshi and Black other respondents (0.03%). When categories were combined, African, Caribbean or Black respondents represented 0.5% of the population.

These figures compare to a non-white population of 2% in the 2001 census. At that time, Pakistani was also the largest non-white ethnic group (0.6%) followed by Chinese and Indian (both 0.3%).⁹

1.2.4 Religion

Previous research has found that self-reported poor health and limiting long-term illness appear to be consistently higher among people whose religion is Muslim, Sikh or Roman Catholic.⁶

In 2008, survey participants were asked whether they regarded themselves as belonging to any particular religion and were given the following list to choose from:

• No religion
• Christian - no denomination
• Roman Catholic
• Church of England/ Anglican/ Episcopal/ Church in Wales
• Presbyterian - Church of Scotland
• Presbyterian - Welsh Calvinistic Methodists
• Free Presbyterian
• Methodist - including Wesleyan
• Baptist
• United Reformed Church/ Congregational
• Brethren
• Other Protestant
• Other Christian
• Jewish
• Hindu
• Islam/Muslim
• Sikh
• Buddhist
• Other non-Christian
• Refused

From 2009 - 2011, the question was simplified. Respondents were asked what religion, religious denomination or body they belonged to and could answer one of the following:

• None
• Church of Scotland
• Roman Catholic
• Other Christian
• Muslim
• Buddhist
• Sikh
• Jewish
• Hindu
• Pagan
• Another religion
• Refused

For the analysis, some categories had to be combined where there were too few respondents to allow meaningful analysis. Sikh, Jewish and Pagan were included in the other religion category.

In the 2008-2011 Scottish Health Surveys, 41% of respondents had no religious faith, 32% were Church of Scotland, 15% Roman Catholic and 9% belonged to other Christian faiths. The largest non-Christian religious group was Muslim (1%).

These figures compare to 28% with no religious faith and 42% Church of Scotland in the 2001 census. The proportion of Roman Catholics (16%) and other Christian faiths (7%) was similar. Muslim was also the largest non-Christian group, representing 1% of the population.¹⁰ The differences between the survey and the 2001 census in the proportions of people who were Church of Scotland and those who had no religious faith are considerable. This may be an issue of timing given the lag between the 2001 census and the 2008-2011 surveys. It will be interesting to compare the 2011 census results when they are published.

1.2.5 Disability

Disabled people report some less positive experiences of inpatient care in Scotland in 2010, although it is not clear whether the variations in experience reported by these groups reflect real inter-group differences in the quality of services received, or inter-group differences in subjective factors such as expectations, perceptions or the way questions are answered, or some combination of these factors.¹¹

Disabled people have a diverse range of capacities and needs. There are an estimated 761,000 disabled people living in Scotland⁴ and as many as one in five Scottish men and women report living with some kind of disability.⁶ However, the wide variety of disabilities and long-term conditions means that the estimated numbers of this population depends on how disability is defined.

From 2008, the survey asked respondents whether they have any long-term physical or mental condition or illness lasting (or expected to last) more than 12 months. Those who reported having such a condition were then asked whether it limited their daily activities in some way. This allowed conditions to be classified as being either 'limiting' or 'non-limiting'. Limiting long-term condition has been used here as an indicator of disability. Note, however, that disability is self-reported, so may be subject to variation in respondents' perception and interpretations.

In the 2008-2011 Scottish Health Surveys, 42% of respondents reported a longstanding condition/illness. Of these, 27% had a limiting long-term condition and 15% had a non-limiting condition. 58% of respondents had no long-term condition.

Data from the 2001 census suggest that 1 in 5 people had a long term limiting illness.¹² However, this figure includes children, and previous health survey reports¹³ have shown that the prevalence of limiting long-term conditions among children are lower than in adults so we could expect that the census data for adults only would be slightly higher than 1 in 5.

1.2.6 Sexual Orientation

The Dimensions of Diversity report suggests that lesbian, gay and bisexual people in Scotland face a range of health issues arising from homophobic prejudice and discrimination - including verbal abuse, physical assault and fear of crime - with poor levels of mental health and higher levels of smoking and substance use.⁶

The survey has asked respondents about their sexual orientation as part of the self-completion questionnaire since 2008. Respondents are asked which of the following best describes their sexual orientation (if forming any of the following relationships: girlfriend / boyfriend / wife/ husband/ partner - with which sex(es) would that be?):

• Bisexual (both sexes)
• Gay or lesbian (same sex)
• Heterosexual (opposite sex)
• Other
• Prefer not to answer

In 2010 the question was adapted to remove the 'Prefer not to answer' category. Respondents were free (as in all questions in the survey) to refuse to answer the question even when that category was not explicitly displayed.

From 2012 the survey has adopted the following harmonised question on sexual orientation:

Which of the following options best describes how you think of yourself?

• Heterosexual or Straight
• Gay or Lesbian
• Bisexual
• Other

In the 2008-2011 Scottish Health Surveys, the vast majority (91%) of respondents were heterosexual. 0.9% reported being bisexual, 0.8% were gay or lesbian and 0.9% reported another sexual orientation. 7% of respondents either refused to answer the question or ticked 'prefer not to answer'. When those who refused to answer are removed from the analysis, the proportions become: 97% heterosexual, 1.0% bisexual, 0.9% gay or lesbian, and 1.0% other.

The Government Equalities Office report on UK inequalities says not only that there is insufficient data on this group, but that the information available is likely to be misleading due to participants under-reporting.¹⁴ It is understood that survey respondents typically under-report non-heterosexual orientations due to the sensitive nature of the topic. The Equality and Human Rights Commission report¹⁵ estimates the proportion of lesbian, gay and bisexual people to be more like 6%, although actually the true proportions are unknown.

There is evidence to suggest that lesbian and gay experience is very different, and it has been suggested that grouping gay and lesbian people into a single category for analysis may prove to be unsatisfactory. Due to the sample size for this group in the 2008-2011 surveys, it was not possible to split this group by gender for the purposes of this report. Further analysis could be done in future, either by combining more years of data together to get a larger sample, or by using the pooled data from the combined population surveys from 2012 onwards, as sexual orientation is one of the harmonised core questions.

1.3 Structure of the report

In the following report, each chapter focuses on a specific health condition or behaviour, such as smoking, obesity or coronary heart disease. Data from the six equality groups outlined above are presented and discussed.

As the survey is cross-sectional, it should be noted that, in all of the analyses, association between two variables does not imply causation.

1.4 Age Standardisation

To ensure that the comparisons are not confounded by the different age profiles of the sub-groups, the data have been age-standardised for all equality groups apart from age. Each of the five equality groups had different age profiles, for example, women are generally older than men and disabled people tend to be older than people without a disability.

Age standardisation is used in order to enable groups to be compared after adjusting for the effects of any differences in their age distributions. When different sub-groups are compared in respect of a variable on which age has an important influence, any differences in age distributions between these sub-groups are likely to affect the observed differences in the proportions of interest.

For example, as physical activity is strongly associated with age whereby activity levels fall as age increases, an analysis of physical activity by disability may show that disabled people were less likely to be physically active than people without a disability. It wouldn't be clear how much of this pattern was related to the older age profile of disabled people and how much was related to their disability. Age standardising removes the effect of the differences in age profiles to allow a clearer picture of the association between physical activity and disability.

1.5 Confidence Intervals

Due to the small sample sizes for some of the groups, confidence intervals have been provided for all estimates in this report. 95% confidence intervals have been used, meaning that if the survey were repeated 100 times, we would expect the estimate to lie within the range of the confidence interval 95 times out of 100.

Confidence intervals are depicted on the charts by the thin black vertical lines with bars on either end. If confidence intervals for two groups overlap the results are not considered to be significantly different¹⁶.