Psychological, social and biological determinants of ill health (pSoBid) in Glasgow: a cross-sectional, population-based study FINAL STUDY REPORT Updated February 2014 Contents Acknowledgements Abbreviations and glossary Summary 1. Introduction 8 2. Review of the evidence: The psychological, social and biological 10 determinants of health 3. Why study the impact of socioeconomic status on health in Glasgow? 35 4. Study aims and hypotheses 39 5. Methods 41 6. Results and published key findings 53 pSoBid study protocol 55 pSoBid investigations into the biological determinants of ill health 57 pSoBid investigations into the social determinants of ill health 67 pSoBid investigations into the psychological determinants of ill health 71 7. Study population profile: descriptive statistics 74 8. Study strengths and limitations 77 9. Public health implications 79 10. Conclusions and next steps 84 11. References 87 12. Study team 111 13. pSoBid published article references 112 14. Appendices 116 2 Acknowledgements Special thanks are due to the study team, past and present, for their enthusiasm and energy, knowledge and insight, time and dedication to the study. Thanks also to the large number of contributors and staff who shared their skills, knowledge and expertise and who supported and contributed to the preparation and publication of study manuscripts. Grateful thanks to Carol Tannahill for helpful, constructive and insightful comments on this report. Thanks are also due to the Robertson Centre for Biostatistics, the University of Glasgow for data management, statistical support and analysis; the administrative staff in the Glasgow Centre for Population Health; the Health Information and Technology section of NHS Greater Glasgow and Clyde for sample selection and analysis; Alasdair Buchanan for GPASS data and analysis. Finally, thanks to all members of GP Practices who participated in the study and to the participants themselves. This study was funded by the Glasgow Centre for Population Health which is a partnership between NHS Greater Glasgow and Clyde, Glasgow City Council and the University of Glasgow, supported by the Scottish Government. Jennifer McLean July 2013 3 Abbreviations and glossary AVLT Auditory Verbal Learning Test: assesses short-term auditory-verbal memory BMI Body mass index BP Blood pressure CHD Coronary heart disease cIMT Carotid intima-media thickness: a surrogate marker of atherosclerosis CRP C-reactive protein: levels of this protein in the blood rise in response to inflammation CRT Choice reaction time: assesses reaction time in a task where the actor has to make one of two or more choices FEC Forced vital capacity: the maximum amount of air able to be exhaled on a single breath FEV1 Forced expiratory volume in 1 second GCPH Glasgow Centre for Population Health GM Grey matter GP General Practitioner GPASS General Practice Administration System for Scotland: clinical record and practice administration software ICAM Intercellular adhesion molecule-1: associated with immune response, becomes up-regulated in response to bacteria IL-6 Interleukin-6: a cytokine with both pro-inflammatory and anti- inflammatory roles; an important mediator of fever and the acute phase response LD Least deprived MD Most deprived MRI Magnetic resonance imaging: used to visualise internal structures of the body pSoBid Psychological, Social and Biological Determinants of Ill-Health Study RGSC Registrar General’s Social Classification SES Socioeconomic status SIMD Scottish Index of Multiple Deprivation: ranks datazones from most deprived to least deprived TNF-α Tumour necrosis factor alpha VBM Voxel based morphometry: a neuroimaging analysis technique that investigates differences in brain anatomy vWF von Willebrand Factor: a glycoprotein present in blood plasma; plays a major role in blood coagulation WM White matter 4 Summary Socioeconomic inequalities in health are essentially universal: poorer health is more common among people in less advantaged circumstances. Social gradients in a range of biological and psychosocial variables exist which indicate that living in deprivation may increase the propensity to develop chronic disease, through as yet unknown mechanisms. This study, pSoBid (pronounced ‘so-bid’), sought to examine the psychological, social, behavioural and biological determinants of ill health within population groups in Glasgow that differed in socioeconomic status and in their susceptibility to develop chronic disease, especially coronary heart disease and Type 2 diabetes mellitus. The study also explored these aetiological links to try to identify potential explanatory pathways for the burden of physical and mental ill health in deprived communities. pSoBid was established and funded by the Glasgow Centre for Population Health (GCPH). The research fieldwork was carried out from December 2005 to May 2007. This report presents the background to the study, the full study methodology and key findings to date. It also presents the implications of these findings for future research and policy development and outlines the next steps and future direction for pSoBid. Study design and participants In a cross-sectional, population-based study, 666 participants took part. The sample population was selected firstly on the basis of how their area of residence ranked in the Scottish Index of Multiple Deprivation 2004. Repeated stratified random sampling was deployed to achieve approximately equal numbers of participants from the most deprived areas and the least deprived areas of Greater Glasgow, as well as approximately equal numbers of men 5 and women and of participants from each age group studied (35-44, 45-54 and 55-64 years). Methods Individuals were invited by letter to attend for assessment of their medical and early life history, lifestyle, classical and novel risk factors for chronic disease, cognitive function and psychological profile, and carotid intima-media thickness (cIMT) and plaque count as indices of atherosclerosis. A small sub- sample of male participants was also invited to attend later for MRI scanning. Results Study findings to date across a number of areas of interest are presented as a series of published journal article abstracts across the psychological, social and biological determinants of health as investigated by the study team. The study was successful in recruiting a sample of subjects collectively comprising a balanced sex and age profile from the most and least deprived communities of Glasgow. This study also illustrates the willingness of subjects to volunteer for a variety of investigations involving psychological, behavioural, sociological and medical questions and tests. The depth and range of the data collected and the analyses undertaken in this study yield important information concerning the relationships between health and socioeconomic status, inflammation, atherosclerosis, mental outlook, cognitive performance and personality, early life family circumstances, genetic disposition and brain morphology. The characteristics of the least and most deprived participant groups varied, in the large majority of cases, in the expected direction across a number of indices of adult socioeconomic status, early life conditions at age 11 years, health behaviours, mental wellbeing and cognitive ability and biomarkers of systemic inflammation and carotid atherosclerosis. 6 Conclusion The valuable and important source of information on the determinants of ill health across the socioeconomic gradient in Glasgow which pSoBid has collected places the study in a good position to provide further insight into the pathways between people’s social circumstances, mental wellbeing and biological markers of disease. However, as the sample was selected from the ends of the socioeconomic gradient, subjects are not representative of the population of interest as a whole. The cross-sectional design of the study also means that it is not possible to identify causal pathways or the relationships between variables and can therefore only report associations. Acknowledging the limitations of the study and the challenges of integrating a range of professional perspectives, the multidisciplinary approach employed in pSoBid has enabled a more holistic understanding of the diverse characteristics of individuals who reside in affluent and deprived communities and their influence on health and health inequalities. Future study analyses will continue to build an understanding of the relationships between the different types of measure, and of the pathways that link poverty, biology, behaviour and psychology and lead to health inequalities in Glasgow and beyond. 7 1. Introduction Heart disease, diabetes, some cancers, rheumatoid arthritis and mental illness are examples of the burden of ill health that is carried disproportionately by those living in deprived communities (Davey Smith, 1997; Mackenback et al., 2003; Marmot, 2005). Not only is the prevalence and incidence of disease higher in areas of deprivation but also the nature of the problem appears to be qualitatively different, and treatment less successful (Ionescu et al., 1998). This inequality in disease risk can partially be explained by the higher prevalence of classical risk factors in deprived areas, but this explanation fails to account for the totality of the variation (Shewry et al., 1992; Tunstall-Pedoe et al., 1997; Capewell et al., 1999) and there is the need to uncover other potential explanatory variables. There are social gradients in a range of biological and psychosocial variables which indicate that living in a deprived environment may increase the propensity to develop chronic disease, through as yet unknown mechanisms. A potential underlying cause of increased prevalence of disease is chronic inflammation (Steptoe et al., 2002; Owen et al., 2003). This has been observed to be more common in deprived than affluent populations, linked to coronary heart disease (Ross, 1999), increased risk of type 2 diabetes (Stern, 1995) and other disorders (Sattar et al., 2003), as well as cognitive dysfunction and altered psychological profile (Weaver et al., 2002; Yaffe et al., 2003; Schram et al., 2007). These aetiological links continue to need further exploration as potential explanations of the burden of physical and mental ill health in deprived communities. This study, pSoBid (pronounced ‘so-bid’), sought to examine the psychological, social, behavioural and biological determinants of ill health within population groups in Glasgow that differed in socioeconomic status and in their propensity to develop chronic disease, especially coronary heart disease and Type 2 diabetes mellitus. 8 The pSoBid study brought together expertise from social epidemiology (the study of how social interactions affect the health of populations), public health, biochemistry, psychology, neuroscience and genetics to build a better understanding of why living in poorer, more stressful circumstances results in higher levels of disease and ill-health. The study sought to relate the social conditions of the population of Glasgow to their psychological profile and their biological status. pSoBid was established and funded by the Glasgow Centre for Population Health (GCPH). The research fieldwork was carried out from December 2005 to May 2007. The Principal Investigator and study director was Professor Chris Packard, Director of Research and Development, NHS Greater Glasgow & Clyde, and Dr Yoga Velupillai, GCPH Programme Manager, was study project manager. The study drew together academic expertise from a range of different disciplines and units from across the University of Glasgow and NHS Greater Glasgow and Clyde. This report presents the background to the study, a review of relevant literature, the full study methodology and key findings to date as a series of abstracts from academic published papers. It also presents the public health implications of these findings for future population health research and policy development and outlines the next steps and future direction for pSoBid. 9 2. Review of the evidence The psychological, social and biological determinants of health Socioeconomic inequalities in health are essentially universal: poorer health is more common among people in disadvantaged circumstances. In all countries where data are available, mortality has been shown to be higher among those in less advantaged socioeconomic positions, regardless of whether socioeconomic position is indicated by education level, occupational social class, home ownership or income level (Adler and Ostrove, 1999; Mackenbach et al., 2003; Lahelma et al., 2004) and this is evident for both men and women. These variables are interrelated, but represent different dimensions of socioeconomic status (Kristenson et al., 2004). Compared to present occupational status, education relates more to social status in early life, whereas income describes the availability of material resources but also a level of status. For measures of education, occupation and income, on average, the more advantaged individuals are, the better their health. In a number of large scale studies a gradient appears across the social spectrum, rather than a threshold effect, indicating that it is the position within the social hierarchy that is important for health (Marmot and Wilkinson, 1999). Studies examining the associations of each socioeconomic indicator with mortality and morbidity have repeatedly shown consistent gradients. These gradients have been shown for all-cause mortality, but also for a wide range of diseases, especially coronary heart disease, diabetes, respiratory diseases, arthritis, poor birth outcomes, and for accidents and violent deaths (Marmot and Wilkinson, 1999). The inverse relationship between socioeconomic position/status and health is one of the most consistent epidemiological findings. The social distribution of physiological risk is partly a reflection of the social patterning of unhealthy behaviours. Unhealthy diet, lack of exercise, tobacco and drug use have now become strongly associated with social disadvantage. Notably, material constraints, prevalent social norms and limited opportunities to make healthy 10