healthscotland.com

Health Education Population Survey (HEPS) 1996-2003

9. Mental health

Mental health is an integral part of health as defined by the World Health Organization. The most common problems presented to GPs are linked to mental health and it is one of the top priorities for the NHS in Scotland (SODoH‚ 1999). As life expectancy increases and treatment or prevention of the main causes of mortality improves‚ issues such as mental health and well-being become increasingly salient on the health agenda. In addition‚ material and social disadvantage can affect health through indirect psychosocial pathways‚ thus contributing to the added burden of morbidity and mortality attributable to behavioural factors (Marmot and Wilkinson‚ 1999).

For health education‚ mental health has been a broad underlying theme of much public communications activity‚ especially that which concerns young people's decision making. Stress in the workplace has been a more specific focus of preventative actions. In 2001‚ a new national programme was launched in Scotland to promote mental health and well-being which will attempt to reduce the rising levels of suicide among young people‚ particularly young men‚ and which features a public education campaign (See Me) to reduce the stigma associated with mental illness. A separate national survey of public attitudes to mental health‚ well-being and mental health problems was carried out in 2002 (and will be repeated in 2004) to provide population-level information on a broad spectrum of mental health issues (Scottish Executive‚ 2002d).

9.1 Stress

In terms of the relationship between mental and physical health‚ it is interesting to note that respondents in the Health Education Population Survey have consistently mentioned stress management as one of the main factors in reducing hypertension (HEBS‚ 2000). Such perceptions echo findings from the longitudinal Health and Lifestyle Survey which suggest that exposure to stressful and disruptive life events is associated with physical and mental ill health (Cox et al‚ 1993). Evidence from a wide range of studies suggests that social isolation is associated with negative health outcomes‚ both mental and physical (Stansfeld‚ 1999)‚ thus highlighting the importance of taking into account wider issues of social health when considering variations in health status.

From 1999 onwards‚ specific questions on stress were included in the HEPS questionnaire. Around one in four respondents said they had experienced large amounts of stress (Table 9.1 overleaf).

Table 9.1: Percentage with large amount of stress‚ by sex/age

Women were more likely than men to report a large amount of stress. In terms of age‚ stress levels were lowest for those aged 65-74. It seems probable that the reported stress is related to the combined responsibilities of study‚ work and raising a family which are likely to be greater from 25-64 than for older people. Those aged 16-24 were also generally less likely to report stress than older respondents‚ but this did fluctuate year on year. There were no consistent differences by social grade.

Table 9.2: Percentage with harmful stress‚ by sex/age/social grade

If respondents reported experiencing stress‚ they were also asked if they considered this harmful. Over a third reported harmful stress (Table 9.2). Women were more likely to report harmful stress than men‚ and those aged 65-74 were less likely to do so than those aged 25-64. Those in social grade E were more likely to report harmful stress than those in the other social grades.

9.2 Mental health status (GHQ12)

As in the Scottish Health Survey‚ the mental health indicator used in HEPS is the twelve-item version of the General Health Questionnaire (GHQ12) which consists of a list of symptoms of mental distress and is scored on a scale from zero to twelve‚ with higher scores reflecting the reporting of more symptoms. This questionnaire has been widely used in general population surveys in order to assess levels of potential psychological morbidity (Bowling‚ 1991). The generally recommended threshold score for detecting potential psychiatric morbidity is two (Goldberg and Williams‚ 1988). For the purposes of this survey‚ a score greater than two is taken as an indicator of potential mental health distress. Consideration is also given to those with a score of four or more as an indicator of potential psychological disorder. The GHQ12 was part of the self-completion section of the questionnaire which approximately 5% of respondents refused to complete. The data in this section are based on responses from those who completed the section.

Table 9.3 Percentage with GHQ12 score greater than two‚ by sex/age/social grade/DEPCAT

Around a quarter of respondents had a score greater than two‚ and this showed no significant change over time (Figure 9.1). Women were more likely than men to report more than two symptoms.

This supports findings of other studies‚ such as the 1993 Survey of Psychiatric Morbidity carried out by the Office of Population Censuses and Surveys‚ that women show higher levels of psychiatric morbidity than men (Meltzer et al‚ 1995). However‚ there is also evidence that while men tend to report fewer symptoms of mental distress‚ they are more likely to exhibit behaviours linked to poor mental health such as heavy drinking and drug misuse (Johnson and Buscewicz‚ 1996). It is therefore possible that sex differences in self-reported mental health are due more to men under-reporting (or not recognising) symptoms of mental distress‚ than to being less prone to them.

There was also an age gradient‚ with fewer older respondents scoring over two‚ particularly those aged 55 or over. It is difficult to establish any patterns of change over the eight years. The proportion of those aged 16-24 with a score over two increased significantly between 1996 and 1998. However‚ data from the 1995 and 1998 Scottish Health Surveys show very little change for this group (Shaw et al‚ 2000) and‚ indeed‚ the increase was not consistently maintained in the 2001-2003 HEPS data.

There was a clearer pattern by social grade. Those in social grade E were consistently more likely to have a GHQ12 score over two. Again‚ it is difficult to establish any patterns of change across time. With the exception of 2001 there was a gradient by DEPCAT‚ with those in the more deprived areas most likely to have a GHQ12 score over two. The patterns described above are illustrated in Figure 9.1 using the 2003 data.

Fig 9.1: Percentage with GHQ12 score greater than two‚ by age/social grade/DEPCAT 2003

One in five had a GHQ12 score of four or more which is a possible indicator of psychological disorder (Table 9.4). As for scores over two‚ women were more likely than men and those in social grade E were more likely than those in other social grades to display four or more symptoms. Older respondents tended to be less likely to score four or more than those aged 25-54. These patterns and trends in GHQ12 scores of four or more are consistent with results from the Scottish Health Survey‚ which showed no significant changes between 1995 and 1998 and females suffering poorer mental health than males. In 1998‚ those living in deprived areas were around 1.4 times more likely to suffer from poor mental health than those in affluent areas (Shaw et al‚ 2000).

Table 9.4: Percentage with GHQ12 score of four or more‚ by sex/age/social grade

Similar patterns were observed for mean GHQ12 scores (Table 9.5). Women‚ those in social grade E and the under-55s had the highest mean scores.

Table 9.5: Mean GHQ12 scores‚ by sex/age/social grade

Main points

• Women were more likely than men to report symptoms of potential mental distress.
• There was a marked social gradient in symptoms of mental distress and disorder‚ with those living in more deprived areas and in social grade E reporting more symptoms.
• Older respondents (aged 55+) were less likely to report symptoms.
• There were no discernible changes over time.