and neurotic hostility. Research has asked ‘who is hostile?’, ‘How does hostility link to
stress?’ and ‘how does hostility link to illness?’.
Who is hostile?
Hostility is higher in men than women (Matthews et al. 1992), higher in those of lower
socio-economic status (e.g. Siegman et al. 2000) and seems to run in families (Weidner
et al. 2000). It seems to be more common in people whose parents were punitive,
abusive, interfering and where there was a lot of conflict (Matthews et al. 1996) and
Houston and Vavak (1991) have argued that it relates to feelings of insecurity and
negative feelings about others.
How does hostility link to stress?
As described above individuals vary in their physiological reactions to stress, with some
showing greater stress reactivity than others. Researchers have argued that hostility
may be the social manifestation of this heightened reactivity. To assess this, Guyll and
Contrada (1998) explored the relationship between hostility and stress reactivity and
reported that chronically hostile people showed greater reactivity to stressors involving
interpersonal interactions than non hostile people. In addition, Frederickson et al.
(2000) indicated that hostile people show larger and longer lasting changes in blood
pressure when made to feel angry. Therefore hostility and stress reactivity seem to be
closely linked. What are the implications of this for the stress–illness link?
How does hostility link to illness?
Much research has shown an association between hostility and coronary heart disease.
In particular, researchers have argued that hostility is not only an important risk factor
for the development of heart disease (e.g. Williams and Barefoot 1988; Houston 1994;
Miller et al. 1996) but also as a trigger for heart attack (Moller et al. 1999). However, it
may not be hostility per se that predicts heart disease but how this hostility is expressed.
Ramsay et al. (2001) and McDermott et al. (2001) explored associations between a range
of components of hostility and symptoms of coronary artery disease (CAD) in men with
CAD versus a control group of men attending a fracture clinic. Results at baseline and at
two year follow-up showed that the best predictor of CAD symptoms was not hostility but
anger expression. Similarly, Siegman and Snow (1977) argued that the expression of
anger and hostility might be a better predictor of stress reactivity and subsequent health
outcomes than the state or either anger or hostility on their own. So how might hostility
and the expression of hostility cause illness? The link between hostility and heart disease
illustrates a role for a physiological pathway with the associated heightened stress
reactivity leading to cardiac damage. However, research also suggests that hostility
may also impact upon health through two other pathways. First, hostility is linked to
unhealthy behaviours such as smoking, alcohol intake, caffeine consumption and poorer
diet (e.g. Lipkus et al. 1994; Greene et al. 1995). Second, hostility may be associated with
other moderating factors. For example, hostile individuals may avoid social support and
STRESS AND ILLNESS 277