Editorials

Hostility and the heart

The links between personality and coronary heart disease have challenged investigators ever since Friedman and Rosenman first suggested that the disease was more common among people with a type A personality—time pressured, competitive, and aggressive.¹ The research has tried to improve clinicians' ability to predict who is most vulnerable to coronary heart disease and so to identify people who are most likely to benefit from prevention measures. One result of this research was recently summarised in a meta-analysis by Miller et al which showed that hostility is an independent risk factor for coronary heart disease.²

Hostility is a broad concept. It has connotations of anger, aggression, and a chronic negative outlook and so encompasses feelings, overt actions, and thoughts or attitudes.³ The cognitive components (those involving thoughts and attitudes) may include cynicism or mistrust, a desire to oppose others, or a wish to do them harm.⁴ A distinction should be drawn between the experience of hostility and its expression.⁵ The experience of hostility is subjective, perhaps including angry feelings or suspicious, cynical thoughts. Expressive or behavioural hostility implies observable acts of aggression, which may be verbal (for example, insults) or physical (for example, punching).⁵

Though these different components of hostility are conceptually distinct, they are difficult to distinguish clearly on research measures.^{2 5} This is partly because the various aspects are correlated: the experience of anger is often related to its expression. In addition, each separate measure may assess more than one aspect of hostility.³ For instance, a measure may address both the extent of a person's cynical attitude and his or her experience of angry feelings. Instruments measuring hostility do, however, vary in their emphasis on particular components. Measures such as the Cook-Medley hostility scale⁶ centre on the experiential, cognitive, cynical aspects, as do many other self report questionnaires. Assessments of hostility based on a standard interview seem, by contrast, to emphasise the observable, expressive, elements.⁵

These overt aspects of hostility have generally had the strongest association with coronary heart disease.² In keeping with earlier reviews,⁷ Miller et al found that expressive hostility was independently associated with objective evidence of coronary heart disease such as confirmed myocardial infarction.² Measures of experiential (cognitive) hostility had slightly weaker correlations with coronary heart disease and were more often associated with subjective evidence of coronary heart disease such as self reported chest pain. Hostility has been found to have associations with other cardiovascular risk factors such as cigarette smoking and alcohol consumption,⁸ but the results of the meta-analysis by Miller et al remained the same after controlling for these factors.² Furthermore, the statistical techniques used allowed for weighting of studies according to the strength of design with greater weight given, for example, to prospective population based studies.

In general studies of the extent of the effect of hostility on the outcome of acute myocardial infarction (fatal and non-fatal) have been in line with those reported in 1995 by Barefoot et al, who found a multiple adjusted relative risk of 1.56 on acute myocardial infarction for an increase in the hostility score of two standard deviations.⁹ Comparisons with physical risk factors are difficult—as is shown by the diversity of study designs and statistical procedures. However, Barefoot et al used definitions of disease and statistical methods⁹ similar to those of Jensen et al, who reported a (multiply adjusted) relative risk of acute myocardial infarction in the highest versus the lowest quartiles of plasma cholesterol concentrations as 2.8.¹⁰ For systolic blood pressure the comparable risks were 3.1 (untreated) and 2.0 (treated), and for smoking 30 g of tobacco/day versus non-smoking the relative risk was 2.8. Recently, Kawachi et al reported a relative risk 2.66 for the effect of anger on coronary heart disease, though the outcome measure included angina pectoris as well as acute myocardial infarction.¹¹

In their meta-analysis Miller and colleagues acknowledged that this current work investigating the links between hostility and coronary heart disease has its foundations in the research of Friedman and Rosenman into type A behaviour.¹ This concept was studied extensively from the 1960s onward and became very well known to the general public as well as to research workers. Gradually, however, the emphasis has moved to hostility since the results have shown repeatedly that it was the hostility element of the type A pattern that was the "toxic" component related to coronary heart disease.^{7 12 13 14} The question that now has to be answered is which component of hostility is most strongly and consistently associated with coronary heart disease.^{2 7}

If we are to advance our understanding of the effects of hostility on coronary heart disease we shall have to pay closer attention to definitions and measurements and also to underlying physiological mechanisms.^{2 4} In the meantime, trying to avoid coronary heart disease gives us another good reason to be nice to one another.

Martha C Whiteman, Research associate,^a F Gerald R Fowkes, Professor of epidemiology,^a Ian J Deary, Professor of differential psychology ^b

^a Wolfson Unit for the Prevention of Peripheral Vascular Diseases, Department of Public Health Sciences, University of Edinburgh, Edinburgh EH8 9AG, ^b Department of Psychology, University of Edinburgh, Edinburgh EH8 9JZ Martha Whiteman is supported by the British Heart Foundation

1. Friedman M, Rosenman R. Association of specific overt behavior pattern with blood and cardiovascular findings. JAMA 1959;12:1286-96.
2. Miller TQ, Smith TW, Turner CW, Guijarro ML, Hallet AJ. A meta-analytic review of research on hostility and physical health. Psychol Bull 1996;119:322-48. [Medline]
3. Barefoot JC. Developments in the measurements of hostility. In: Friedman HS, ed. Hostility, coping and health. Washington DC: APA, 1992.
4. Smith TW. Concepts and methods on the study of anger, hostility and health. In: Siegman AW, Smith TW, eds. Anger, hostility and the heart. Hillsdale, NJ: Erlbaum, 1994: 23-42.
5. Siegman AW. From type A to hostility to anger: reflections on the history of coronary-prone behavior. In: Siegman AW, Smith TW, eds. Anger, hostility and the heart. Hillsdale, NJ: Erlbaum, 1994: 1-21.
6. Cook W, Medley D. Proposed hostility and pharisaic-virtue scales for the MMPI. J Appl Psychol 1954;38:414-8.
7. Booth-Kewley S, Friedman HS. Psychological predictors of heart disease: A quantitative review. Psychol Bull 1987;101:343-62.
8. Whiteman MC, Fowkes FGR, Deary IJ, Lee AJ. Hostility, cigarette smoking and alcohol consumption in the general population. Soc Sci Med 1997;44:1089-96.
9. Barefoot JC, Larsen L, von der Lieth L, Schroll M. Hostility, incidence of acute myocardial infarction, and mortality in a sample of older Danish men and women. Am J Epidemiol 1995;142:477-84. [Abstract/Free Full Text]
10. Jensen G, Nyboe J, Appleyard M, Schnohr P. Risk factors for acute myocardial infarction in Copenhagen II: Smoking, alcohol intake, physical activity, obesity, oral contraception, diabetes, lipids and blood pressure. Eur Heart J 1991;12:298-308.
11. Kawachi I, Sparrow D, Vokonas PS, Weiss ST. Symptoms of anxiety and risk of coronary heart disease: the normative aging study. Circulation 1996;5:2225-9.
12. Dembroski TM, Costa PT. Coronary-prone behavior: components of the type A pattern and hostility. J Pers 1987; 55:211-35.
13. Deary IJ, Fowkes FGR, Donnan PT, Housley E. Hostile personality and risks of peripheral arterial disease in the general population. Psychosom Med 1994;56:197-202. [Abstract/Free Full Text]
14. Johnston DW. The current status of the coronary-prone behavior pattern. J Roy Soc Med 1993; 86:406-9.