Scottish Health Survey - topic report: mental health and wellbeing

1. Introduction

1.1 Policy context

The improvement of mental wellbeing is a national indicator in the Scottish Government's National Performance Framework. In support of this, the 'Mental Health Strategy: 2012-2015' sets out the Scottish Government's priorities with regard to improving mental health services, promoting mental wellbeing, preventing mental illness and ensuring that individuals and communities can maintain and improve their own health. The Strategy describes 36 commitments it will adhere to in achieving these priorities.

A key element in the Strategy is enabling people to become more involved and active in their own health and wellbeing. The evidence base for people taking a leading role in managing their own illness over time and the wider benefits to them that this approach offers is well established. The Strategy focuses on things people and communities can do for themselves which are particularly valuable given the additional benefits that people derive from taking control of their own health and wellbeing.

Examples of this approach include the Living Life Guided Self Help Service operated by NHS 24, the Steps for Stress resources managed by NHS Health Scotland, and Ginsberg - a web-based tool launched by the Scottish Government to help people manage their wellbeing in relation to other aspects of their lives. Ginsberg allows people to see patterns that are developing, to draw links between what they are doing with their time and how they are feeling, and to see the changes they can make to improve their wellbeing.

SHeS is the data source for 28 of NHS Health Scotland's 54 national mental health indicators for adults^[a], intended to allow national monitoring of adult mental health and covering outcomes and risk factors for poor mental health.

1.2 Aims of the report

The primary aim of this report is to investigate the factors that are significantly associated with poor mental health and wellbeing among adults of 16 years of age and older in Scotland.

The report examines how mental wellbeing varies by socio-demographic, behavioural and health condition factors based on Scottish Health Survey (SHeS) data and using the measures described in section 1.4. Other measures of mental health and wellbeing covered by the survey, such as anxiety, depression and life satisfaction, are not analysed in this report.

1.3 Scottish Health Survey background

The Scottish Health Survey was established in 1995 to provide data on the health of the population living in private households. The survey was repeated in 1998 and 2003, and has been carried out continuously since 2008. As well as general health and mental wellbeing, topic areas include long term conditions, obesity, physical activity, alcohol consumption, smoking, diet, dental health and gambling.

Since 2008, SHeS has presented data on WEMWBS scores, while GHQ12 scores have been presented since 2003. With the exception of high-level trends over time, all analyses which feature in this report are based on SHeS data spanning the 2012 and 2013 surveys.

Figure 1A

Sample structure of 2012, 2013 and 2012/2013 combined datasets

SHeS has a core and modular structure whereby most questions are asked of all participants (the core sample) while others (modular questions) are asked only of a proportion of the total sample. Questions on mental wellbeing, as well as the factors explored in this report, were included as part of the core SHeS questionnaire in both 2012 and 2013.

1.4 Measurement of Mental Health and Wellbeing

Two measures of mental health in the Scottish Health Survey are examined in this report. Below is a description of each measure, how they were developed and have been used.

1.4.1 The Warwick-Edinburgh Mental Wellbeing Scale (WEMWBS)

The Warwick-Edinburgh Mental Well-being Scale (WEMWBS) was developed by researchers at the Universities of Warwick and Edinburgh, with funding provided by NHS Health Scotland, to enable the measurement of mental wellbeing of adults in the UK. Within the Scottish Health Survey, it has been used to monitor the Scottish Government National Indicator "improve mental wellbeing". It was adapted from a 40 item scale originally developed in New Zealand, the Affectometer 2^[2].

The WEMWBS scale comprises 14 positively worded statements with a five item scale ranging from '1 - None of the time' to '5 - All of the time'. The lowest score possible is therefore 14 and the highest is 70. The 14 items are designed to assess positive affect (optimism, cheerfulness, relaxation); and satisfying interpersonal relationships and positive functioning (energy, clear thinking, self-acceptance, personal development, mastery and autonomy).

WEMWBS is not designed to identify individuals with exceptionally high or low levels of positive mental health, so cut off points were not developed at the indicators inception^[3]. However, recent work by Bianco^[4] suggests that a cut-off of 43.5 performed well in screening depressive symptomatology and may be an accurate tool for the assessment of symptoms of depressive disorders. The Scottish Health Survey 2011 annual report applied a cut off of one standard deviation less than the mean. Scores below this threshold indicate respondents with below average mental well-being. For the same reasons, and to allow comparison, the same methodology has been used in this topic report - a cut-off of 42 has therefore been applied.

1.4.2 Component Questions of WEMWBS

The fourteen positively worded statements that make up the WEMWBS scale are listed below. All of the component questions have a five item scale ranging from '1 - None of the time' to '5 - All of the time', so the summed WEMWBS scale ranges from 14 to 70 with higher scores indicating better mental wellbeing. Respondents are asked to pick the score that best describes their experience over the last two weeks for each question.

• Been feeling optimistic about the future
• Been feeling useful
• Been feeling relaxed
• Been feeling interested in other people
• I've had energy to spare
• Been dealing with problems well
• Been thinking clearly
• Been feeling good about myself
• Been feeling close to other people
• Been feeling confident
• Been able to make up own mind about things
• Been feeling love
• Been interested in new things
• Been feeling cheerful

1.4.3 The General Health Questionnaire (GHQ12)

The General Health Questionnaire is a scale designed to detect possible psychiatric morbidity in the general population, which has been validated in the UK and worldwide^[5]. Within the Scottish Health Survey, it was administered to participants aged 13 and above.

The questionnaire contains 12 questions about the informant's general level of happiness, depression, anxiety and sleep disturbance over the past four weeks, arranged so that no reverse scoring is required. The scoring takes the form of a four point Likert scale with response options 'less than usual' (score 0), 'no more than usual' (1), rather more than usual' (2) or 'much more than usual' (3) (or in the opposite order to avoid the need for reverse scoring). To identify psychiatric morbidity these scores are converted into binary scores, with 0 or 1 set to zero and 2 or 3 set to 1.

The scores for the summed variable then range from zero to 12, with higher scores indicating greater likelihood of possible psychiatric morbidity. With this new variable, a cut-off score can be selected to signify the possible presence of psychiatric morbidity. Although there is considerable variety in literature on the choice of the most appropriate score to use as the cut-off,^{[6] [7] [8] [9]} a figure of four has been used for this report, in line with previous Scottish Government reports and examples from literature^{[10] [11] [12] [13]}

1.4.4 Component questions of GHQ12

There are twelve component questions in GHQ12, six positively and six negatively worded. Each question has a four item scale arranged so that no reverse scoring is required and respondents are asked to pick the score that best reflects their experiences over the last four weeks for each question.

• Able to concentrate
• Lost sleep over worry
• Felt playing useful part in things
• Felt capable of making decisions
• Felt constantly under strain
• Felt couldn't overcome difficulties
• Able to enjoy day-to-day activities
• Been able to face problems
• Been feeling unhappy and depressed
• Been losing confidence in self
• Been thinking of self as worthless
• Been feeling reasonably happy

1.4.5 Limitations in measures

WEMWBS has shown good reliability in terms of internal consistency and stability at a population level^{[3], [14]} and has proved to be a psychometrically strong population measure of mental wellbeing.^[15] Although the measure was not designed as a screening tool for the assessment of depressive symptoms via a cut-off score, a score of 43.5 has been shown to be an appropriate discriminating point. As in previous reports, the methodology applied here defines low WEMWBS scores as one standard deviation below the population mean or lower. This is equivalent to scores of 41 or lower, and has been used to identify respondents at risk of low mental wellbeing.

GHQ12 has been found to be 'uniformly good' in identifying anxiety and mood disorder cases among adults in clinical settings^[5] and its ability to estimate the prevalence of such disorders appears reasonable^{[16], [17]} However, some doubt has been raised as to the validity of GHQ12 as a screening tool for non-specific psychiatric morbidity. In particular, response bias in the negatively worded component questions is thought to lead to measurement error.^[18] The selection of the most appropriate cut-off point in GHQ12 is subject to some discussion, but, for consistency with other reports, a cut-off of four has been applied in this analysis.

Although WEMWBS and GHQ12 are designed to be analysed as aggregate measures, some selected findings for component questions are described in this report where they stand out and are statistically significant.

1.5 Literature review

A review of the literature has been carried out to identify factors associated with problem and positive mental health. Mental disorders increase the risk for physical health conditions and contribute to unintentional and intentional injury, while conversely many health conditions increase the risk for mental disorders, and co-morbidity complicates help-seeking, diagnosis and treatment, and influences prognosis^[19]. The NHS Health Scotland report 'Scotland's Mental Health: Adults 2012'^[20] provides a comprehensive list of indicators and contextual factors associated with mental health. Some of the factors for which data is available in the Scottish Health Survey are examined in more detail below.

Socio-demographic factors

The socio-demographic factors of age, gender, marital status, education levels and income have consistently been identified as having an association with levels of depression^[21] and psychiatric disorders^[23]. A study on minor psychiatric morbidity across the UK has found that women have a significantly higher GHQ12 score, and thus poorer mental health, than men^[11]. In Northern Ireland, having housing worries is a predictor of increased risk of anxiety and depression^[24] and, in Australia, research suggests a link between housing affordability, tenure type and mental health^{[25], [26]}. Being a single parent is associated with increased risk of poor mental health^{[27], [28]}. Another socio-demographic factor linked with increased levels of common mental illness is the actual^[29] and perceived^[30] level of neighbourhood crime.

Occupational class and household income are both positively associated with mental health^{[31] [32] [33]}. Mental disorders occur more frequently in deprived urban areas^[34] and in a disadvantaged area of Berlin psychological stress was associated with a younger age, being female and living alone^[35]. In Sweden, living alone and economic insecurity both showed strong association with anxiety and depression^[36]. Deprivation is associated with a higher prevalence of depression and anxiety among people with multi-morbidity in Scotland^[37]. The effect of unemployment on mental health has been extensively examined both in terms of the immediate impact^{[38], [39]} and the longer-term mental health scarring effects of multiple exposure to unemployment during the life course^[40]. Among those in work, several psychosocial factors within the work environment have been shown to impact on well-being and psychological distress^{[41], [42]}.

Within the rural population of the UK, the farming community exhibit higher GHQ12 scores, and thus poorer mental health, than the non-farming community^[43]. However, the general rural population have a lower risk of depression compared to their urban counterparts^[44].

Mental health disorders such as depression are higher among informal caregivers than the general adult population but the degree of any association varies with both the number of caregiver burdens and the behavioural or health problems among the care recipients^{[45], [46]}.

Lifestyle factors and health behaviours

People with mental health problems are much more likely to develop poor physical health when compared to the general population^[49]. Mental disorders such as depression have been inversely associated with physical exercise^{[50] [51] [52] [53]} and the causal link is considered bi-directional^{[54], [55]}.

Recent evidence suggests that unhealthy diets are risk factors for some mental disorders, particularly depression and dementia^{[56], [57]}. In particular, fruit and vegetable intake has been shown to have a strong inverse association with mood and anxiety disorders^[58]. Frequent family meals have been shown to have a positive impact on mental health, particularly among children and adolescents^{[59] [60] [61]}. Most of these studies have only considered the mental wellbeing of children, whereas this report examines associations between family meals and mental wellbeing in adults.

Alcohol use disorders are recognised and classified as mental disorders^[62]. Excessive alcohol consumption is associated with an increased prevalence of depression^[63],[64] and psychological distress^[65]. The literature suggests it is the highest levels of alcohol consumption that are associated with poor results across various measures, including mental wellbeing^[66].

A meta-analysis of research looking at the association between smoking and depression noted a two-fold increased risk of depression among smokers relative to those who have never smoked or formerly smoked^[67].

Health conditions

Some mental health problems such as depression, bipolar disorder and anxiety are associated with obesity^[68] although the direction of causality is uncertain. In addition, gender and age each alter the association between obesity and mental health problems with overweight men having better mental health^[69] and young women the reverse.^[70],[71] In addition, it is likely that the association between mental health and obesity (as measured by the Body Mass Index) may vary by type of mental health problem^[72].

Social involvement with community groups and resources has been shown to support mental well-being among people with long-term conditions^[73]. A retrospective study has shown that self-reported general health is a significant predictor of clinical outcomes including cancer, coronary heart disease and psychiatric hospitalisation among Scottish adults^[74]. However, the same study found self-reported mental health was a predictor only of psychiatric hospitalisations. Long-standing illnesses, disability and adverse life events are associated with increased anxiety and depression in the Northern Ireland population^[24].