Q J Med 2003; 96: 463-464
© 2003 Association of Physicians
Editorial |
Stress, disease and ‘joined-up’ science
Medical Research Council Unit
Southampton General Hospital
Southampton
With the news that Glenn Roeder, manager of West Ham United, was admitted to hospital following the novelty of a win from his relegation-haunted side, stress and cardiovascular disease are once more on the front pages of our newspapers. It seems that football management is as bad for your health as smoking or drinking,1 and the anguish of standing on the touchline, but being unable to intervene in the drama of the game, seems to be the cause. Studies of 60 top managers have apparently shown that their pulse and blood pressure responses during a match are higher than when the same people ran themselves to exhaustion on a treadmill.
Stress is of undoubted economic importance. The Confederation of British Industry estimates that businesses lost £10.2 billion in 1998 through lost productivity caused by stress, and figures from the Health and Safety Executive suggest we lose some 13.4 m working days per year to stress-related disease. We also employ over 50 000 stress counsellors. It seems odd, therefore, that conventional medicine remains shy of the concept of stress-related disease. A glance at any standard text on general medicine reveals only an occasional mention of stress either in relation to coronary heart disease, hypertension or the metabolic syndrome, which are the prime contenders for stress-related pathologies. The specialist literature is little better.
Undoubtedly, part of the problem is that stress remains a somewhat nebulous concept. It has been said that defining it is rather like defining pornography: we all know what it is, but it remains difficult to specify a clear boundary. The problem of developing a coherent approach is also heightened by the multiplicity of different specialities involved in stress research. Indeed, there are almost industrial demarcations in the scientific groups and their approach to the problem. While endocrine systems have remained the province of neuroendocrinologists and autonomic physiologists, studies of the central response to stress have remained largely the domain of psychologists. However, developmental biologists, sociologists and epidemiologists, each with their own specialist journals, meetings and even different scientific languages, also contribute, making the field very difficult for outsiders.
The scientific idea behind stress-related disease is simple, and was formulated by Hans Selye over 60 years ago. He recognized that physiological systems activated by stress not only served to protect the individual and allow them to cope with both internal and external threats, but were also capable of damaging the body.2 This idea is strongly supported by a large body of animal evidence, which shows that exposure to external stressors activates the hypothalamic-pituitary-adrenal axis and autonomic stress responses, which in turn raise blood pressure, blood glucose and lipids, and mimic other features of the metabolic syndrome. The problem, and source of controversy, has been the difficulty in showing that these systems are of clinical importance in human populations.
Because psychosocial factors appear to be the predominant stressors in modern society, epidemiological studies have tried to construct a measure of stress based on the psychological characteristics of an individual’s position in the social hierarchy. This idea has been strongly supported by research with non-human primates, which has suggested that the incidence of atherosclerosis is related to the position in the dominance hierarchy. The human equivalents, such as lack of control, job strain and poor self-esteem, are indeed linked with high blood pressure and the progression of atherosclerosis.3 However, most of these studies depend on self-reported psychological measures, potentially leading to spurious associations, because people reporting higher stress levels tend to report more symptoms of disease.4 Against this, however, biological markers of stress such as plasma cortisol concentrations and indices of sympathoadrenal activity are linked with a range of cardiovascular outcomes.5 These findings suggest that the indices of self-reported stress used in these epidemiological studies do not really capture the stresses to which individuals are exposed.
The physiology of the systems involved in stress is complex and only partly understood; increasingly, we are realizing that the biomarkers used to assess activation of the stress system are only indirectly related to what is happening at the tissue level. Another difficulty, noted in the earliest days of stress research, is that there are very substantial individual differences in response to the same external stressor. Although this relates in part to the way in which a threat is perceived, there is increasing evidence that the quality of nurturing and the early environment also have profound influences on the biological response to stress in later life. As a result, stress studies need to adopt what has become known in epidemiology as a ‘lifecourse’ approach, recognizing that the response to a stressor depends not only on the nature of the stress offered, but on the ‘stress history’ of the individual which may start during early fetal life. Some animal experiments even suggest that the effects of stress are transgenerational.6 If so, then the stress history of the parents may need to be taken into account.
Better methods are required for evaluating how people respond to stressful stimuli, as well as assessing the stressors to which people are exposed. Much stress-testing has depended on measuring peoples’ reactions to a series of mental stress tasks, which are frequently computer-based. The ability to carry out these ‘stressful’ tasks will depend on the subject's level of education, motivation and ability to concentrate (and probably familiarity with computers), and must therefore detract from their primary purpose of assessing stress responsiveness. Improved techniques need to be integrated with appropriate clinical studies where there is adequate data to assess the current impact of the stressor as well as information on the lifetime ‘stress history’. We also need to re-evaluate how we assess hormonal and autonomic responses in the light of recent developments in our understanding of these systems.
Above all we need a clear answer for our patients who remain convinced that stress is a potent cause of their ills. To do this is clearly going to be a challenging task, requiring ‘joined-up’ science, in which a wide range of different disciplines, including neuroscientists, developmental biologists and epidemiologists, work together to strengthen the evidence base. If we fail, we risk relegating the relationships between stress and disease to fringe science. We also risk that our patients (including our football managers) will turn away from us and seek help from alternative medicine.
References
2. McEwen BS. Protective and damaging effects of stress mediators. N Engl J Med 1998; 338:171–9.
3. Hemmingway H, Marmot M. Psychosocial factors in the aetiology and prognosis of coronary heart disease: systematic review of prospective studies. Br Med J 1999; 1460–7.
4. MacLeod J, Davey Smith G. Commentary: stress and the heart, 50 years of progress? Int J Epidemiol 2002; 31:1111–13.
5. Brunner EJ, et al. Adrenocortical, autonomic and inflammatory causes of the metabolic syndrome. Circulation 2002; 106:2659–65.
6. Francis D, Diorio J, Liu D, Meaney MJ. Nongenomic transmission across generations of maternal behavior and stress responses in the rat. Science 1999; 286:1155–8.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||