![]()
![]()
Psychosocial factors and their health effects are widely discussed, controversial topics. By using epidemiological methodology and linking psychological, social, biological and medical sciences, psychosocial epidemiology seeks to increase understanding of the effects of psychosocial factors on population health. This scientific discipline deals with a wide range of questions including, for example, how much positive and negative emotions affect our risk of developing physical disease; what role stressful life events play in health and disease; whether there is a disease-prone personality; and, importantly, whether we can improve our disease-free life expectancy and well-being by avoiding psychosocial risk factors.
The earliest attempts to link psychosocial factors to disease endpoints are almost contemporaneous with the emergence of modern epidemiology itself. Thus, shortly after the emergence of an association between cigarette smoking and lung cancer in British physicians (Doll & Hill, 1954), the suggestion was raised that type A behaviour (a behaviour pattern characterized by aggressiveness, ambitiousness, restlessness and a strong sense of time urgency) might precipitate cardiovascular disease (Friedman & Rosenman, 1959) and that life events were linked to chronic diseases (Holmes & Rahe, 1967). The concept of stress at work, defined as high demands combined with low job control, was launched in the 1970s (Karasek, 1979). Research on psychosocial factors in health continues to grow at a rapid pace: during the last 10 years, for example, the number of published studies on this topic has more than doubled (Figure 1.1). Pooling of data sets into mega studies during the 2010s has further boosted progress in psychosocial research (Kivimäki et al., 2012; Russ et al., 2012).
In this chapter, we outline the methodological principles behind psychosocial epidemiology, a branch of epidemiology that offers effective scientific structure for the study of the links between putative risk factors and disease conditions at a population level. Our aim is to provide the reader with the essentials to evaluate the diverse set of studies on psychosocial factors and health that are published today. In particular, we describe commonly used study designs in this field of research, key principles for judging the validity of epidemiological evidence, as well as challenges in translating evidence into action.
What is epidemiology?
Epidemiology is the study of the distribution, causes and control of disease in human populations. Although the randomized controlled trial is a study design in epidemiology, the discipline typically relies on observation (rather than experimentation) where the natural history of an exposure and outcome are measured rather than the exposure being directly manipulated. Although epidemiological (observational) studies increasingly use data sets that are sufficiently large and long-term for well-powered analyses, causal inference poses a particular challenge given the influence of confounding and biases. Currently, several methodological approaches and concepts, described in this chapter, are available to address this challenge.
The public health importance of any given psychosocial risk factor is measured by the strength of the association between the risk factor and the disease outcome, as well as an assessment of prevalence and malleability of the risk factor. In addition, public health impact is also dependent on the outcome of interest, with diseases that are common causes of mortality and morbidity ranking having priority. Such outcomes include, for example, coronary heart disease, stroke, diabetes, selected cancers and chronic obstructive respiratory disease, which contribute to a substantial burden of life-years lost across different regions of the world, but also common causes of disability and discomfort, such as depression, low back and neck pain and other injuries (Global Burden of Disease Collaboration, 2016a). With population ageing, the importance of dementia is increasing, while in low-income countries communicable diseases (e.g. malaria and HIV/AIDS) and maternal, neonatal and nutritional diseases still represent an important burden (Global Burden of Disease Collaboration, 2016a).
In Table 1.1, we list some important epidemiological studies from western societies, some of which have been crucial in understanding the aetiology of these common diseases, particularly cardiovascular disease. Findings from these successfully conducted cohort studies have led to favourable changes in health policies and clinical practice, and improvements in population health, a feature that makes epidemiology a science which can directly impact our lives. Epidemiological research, for example, played a pivotal role in identifying the key risk factors for cardiovascular disease (stroke and heart attack). Cardiovascular disease was very common in the 1940s, particularly in the USA, being responsible for one in every two deaths, including that of President Roosevelt from uncontrolled, ‘malignant’ hypertension in 1945. Three years later, the Framingham Heart Study, a population-based study, was established in the USA to identify risk factors for chronic disease with the aim of developing more effective strategies for prevention. In a small, affluent town in Massachusetts, investigators recruited 5,209 men and women aged 30–62 years who subsequently had biennial biomedical examinations and surveillance for cardiovascular disease events. Within two decades, several risk factors in this study were shown to predict coronary heart disease, including: elevated blood pressure (Kagan, Gordon, Kannel, & Dawber, 1959), high serum cholesterol levels (Kannel, Castelli, Gordon, & McNamara, 1971), cigarette smoking (Doyle, Dawber, Kannel, Heslin, & Kahn, 1962) and diabetes (Kannel & McGee, 1979). Similarly, the importance of physical activity as a protective factor for cardiovascular disease was originally identified by an epidemiological study. Taking an ingenious approach in what would now be described as a natural experiment, Jerry Morris and co-workers showed that the sedentary drivers on London’s double-decker buses suffered twice as many heart attacks as the more physically active conductors (Morris, Heady, & Raffle, 1956). Distrusting his own results, Morris replicated the finding by comparing heart attack rates in postmen relative to desk-bound telephonists (Morris, Heady, Raffle, Roberts, & Parks, 1953).
Table 1.1Examples of well-known cohort studies including assessment of psychosocial factors (chronological order)
| The Framingham Heart Study began in 1949 and targeted a representative sample of 5,209 men and women, aged 30–62, which was selected from approximately 10,000 persons of that age living in Framingham, a small town near Boston, MA. Of these, 5,127 were free of coronary heart disease and were re-examined biennally. Biomedical examinations included assessments of psychosocial factors, such as Type A behaviour pattern, and a broad array of attributes of social networks. |
| The National Child Development Study (also known as the 1958 Birth Cohort Study) follows the lives of 17,415 people born in England, Scotland and Wales in a single week of 1958. Data collections have been at ages 7, 11, 16, 23, 33, 42, 45 (biomedical examination), 50 and 55. Psychosocial factors include IQ, financial difficulties, partnerships, social participation, attitudes and social interactions. |
| The Dunedin Multidisciplinary Health and Development Study was established in 1972–73 and the assessments of the 1,037 participants from a general sample of New Zealanders started at the age of 3 years in 1975–76. Since then they have been assessed every two years until the age of 15, then at ages 18, 21, 26, 32 and 38. Psychosocial factors assessed include childhood maltreatment, IQ, personality and social relationships. |
| The Nurses’ Health Study was established in 1976 with a study population of 121,700 married registered nurses, aged 30–55 from California, Connecticut, Florida, Maryland, Massachusetts, Michigan, New Jersey, New York, Ohio, Pennsylvania and Texas. Cohort members received a follow-up questionnaire every two years, including questions on psychosocial factors, such as personal characteristics, job strain, interpersonal violence, social relationships and sexual orientation. |
| The Whitehall II Study (also known as the Stress and Health Study) was established in 1985 to explore the relationship between socioeconomic status, stress and cardiovascular disease. A cohort of 10,308 participants aged 35–55, of whom 3,413 were women and 6,895 men, was recruited from the British Civil Service. Since this first wave of data collection, self-completion questionnaires and clinical data have been collected from the cohort every two to six years, including information on a wide range of psychosocial factors, such as stresses at work and private life, social activities, social support, conflict and networks. |
| The Atherosclerosis Risk in Communities (ARIC) Study began in 1987 when each of six ARIC field centers randomly selected and recruited a cohort sample of approximately 4,000 individuals aged 45–64 from a defined population in their community. A total of 15,792 participants took part in baseline examination, and were re-examined in 1990, 1993 and 1996 and have then been followed up with telephone interviews and substudies. Psychosocial factors measured include personality, work stressors and social support. |
| The English Longitudinal Study of Ageing (ELSA), established in 2002, includes 12,099 members aged 50 years or older drawn from respondents to the population-based Health Survey for England (HSE). Study members have a face-to-face interview (a computer-assisted personal interview followed by a self-completion questionnaire) every two years of the study and a nurse assessment every four years. Psychosocial factors include post-retirement labour market activity, social networks and social supports and social, civic and cultural participation. |
An increased awareness of these predictors provided the necessary stimulus for public health initiatives to curb smoking and physical inactivity, as well as the mass screening and detection of hypertension and hypercholesterolaemia. The demonstration of hypertension and high serum cholesterol as risk factors for cardiovascular disease in turn paved the way for the development of highly effective pharmaceuticals to treat these conditions (Cholesterol Treatment Trialists’ Collaboration, 2012; Law, Morris, & Wald, 2009).
Psychosocial factors in an epidemiological context
Epidemiology can be divided into a number of distinct subfields, one criterion being the set of risk factors that are studied as determinants of health and disease, including biological, behavioural, genetic, social and psychosocial (Table 1.2). Clinical epidemiology, for example, is concerned with generating evidence for clinicians to guide their decisions on prevention and treatment. Study questions in this subfield include, for example, what is the optimal threshold for initiating treatment for high blood pressure to prevent or delay dementia? What are the test characteristics, such as sensitivity, specificity and predictive value, of new screening or diagnostic tests? Which new biomarkers can improve identification of at-risk groups for heart failure?
Table 1.2Example of common areas of interest in epidemiology
| By exposure: Clinical epidemiology Environmental epidemiology Genetic epidemiology Nutritional epidemiology Occupational epidemiology Pharmacoepidemiology Psychosocial epidemiology Social epidemiology |
| By outcome: Cancer epidemiology Cardiovascular epidemiology Infectious disease epidemiology Psychiatric epidemiology Epidemiology of ageing |
| By method or type of study question: Analytical epidemiology Descriptive epidemiology Experimental epidemiology Observational epidemiology Theoretical epidemiology |
Social epidemiology, in turn, focuses on the social determinants of health and disease such as gender, socioeconomic status, and race/ethnicity (Berkman, Kawachi, & Glymour, 2014). The original Whitehall study of British civil servants, which demonstrated a graded relation of higher income and occupational status with a lower risk of morbidity and mortality – counterintuitive at the time when executive workers were thought to be burdened by detrimental levels of stress – is widely hailed as one of the first convincing demonstrations of the socioeconomic gradient in health in contemporary society (Marmot, Shipley, & Rose, 1984; Marmot et al., 1991).
The first Whitehall study was established in the 1970s as an examination of clinical risk factors with questions on car ownership originally added to the baseline examination to assess physical inactivity (Reid et al., 1974). Unexpectedly, car ownership, rather than the lack of car, was associated with reduced mortality and...
