Consensus Conference: Stress and Challenge, Health and Disease
Brisbane, Australia
9, 10, 11 February 1998

Professor Beverley Raphael
Professor Philip Morris
and Dr Alex Bordujenko

Executive Summary

A consensus conference entitled Stress and Challenge, Health and Disease was convened in February 1998 by the RMA as part of the investigation into the possible causative role of psychosocial stress (particularly war or service related stressors) in the development of ischaemic heart disease (including coronary atherosclerosis), hypertension, cerebrovascular accident and psoriasis (gazetted 22/1/1997) and in respect of post traumatic stress disorder and hypertension and post traumatic stress disorder and ischaemic heart disease (gazetted 23/4/1997).

The RMA was established in 1994 under the provisions of the Veterans’ Entitlements Act 1986 (the Act). Its primary function is the determination of Statements of Principles (SOPs) under section 196B of the Act. These SOPs apply to a particular kind of injury, disease or death and list the factors which must be related to war or defence service. The RMA is mandated by Australian law to determine the presence or absence of causal relations between service related factors and disease, injury or death. In this case the factors of interest are psychosocial stress (particularly war or service related stressors) and post traumatic stress disorder. The RMA is required to use the applicable criteria for assessing causation currently applied in the field of epidemiology.

The RMA recognized that the literature examining the effects of psychosocial stressors and the perception of stress needed review in a broad contextual process. This is particularly in respect to the sound medical-scientific evidence concerning both the positive and negative effects of stress on the human organism.

To respond to this challenge the conference was held to examine the effects of stress on psychiatric illness and cardiovascular disease. It brought together eminent Australian and overseas scholars and researchers acknowledged for their expertise in examining associations between stress and psychiatric illness and cardiovascular disease. The conference was opened by the Minister for Veterans’ Affairs, Bruce Scott and co-chaired by Professors Beverley Raphael and Philip Morris. The veteran community was represented at the conference by observers from various ex-service organisations.

The three main questions addressed by the conference were:

How can we establish causes, measure and set doses that explain the association between exposure to stressors, psychiatric illness and cardiovascular disease?
Is there an association between exposure to stressors, psychiatric illness and cardiovascular disease?
What are the potential mechanisms for the association between stressors and disease, with special emphasis on psychiatric illness and cardiovascular disease?

The formal conference program combined reviews of the published literature with presentations of new data sets from Dr Terry Keane, Professors George Vaillant, Jake Najman and Gerard Byrne. Dr Trevor Anderson provided a personal and evocative presentation of the human perspective of positive and negative responses to stressor challenge and Professor Lars Weisaeth outlined the Norwegian system of military compensation and the United Nations perspective in the recognition, management and prevention of abnormal stressor responses in military and peace keeping personnel. The counterpoint of individual and population based data focused the conference participants in their tasks. The structure of the conference demanded considerable syndicate group consideration of the central issues, one of the key underlying issues being the definitions of stressors and stress which would have utility for the RMA.

The utility and role of modern epidemiological methods in the assessment of causal relationships between psychosocial stressors and psychiatric illness and cardiovascular disease were canvassed. A consensus was reached that use of the science of epidemiology was appropriate and necessary in the process of examining potential causal associations between stressors, stress and disease.

The process of causal inference was recognised as complex and subject to interindividual variation. It was considered that the use of causal criteria such as those attributed to Sir Austin Bradford Hill assisted in clarifying observed associations from causal associations. The Bradford Hill criteria are:

Strength of association
Consistency of association
Specificity of association
Temporality of association
Gradient of effect (dose)
Biological plausibility/coherence
Experimental evidence
Analogy

It was recognised that in assessing causal associations between stressor experiences and many of the psychiatric and cardiovascular disorders a number of the criteria would not be met fully or at all. However, it was felt that in the consideration of the body of evidence, the Bradford Hill or similar criteria should be used as an important part of evaluating causality between stressor experiences and disease.

It was considered that once causality had been inferred, a dose, based on the available sound medical scientific evidence, could be determined. In some situations the dose would take the form of a threshold effect in that only after a certain exposure would a negative health consequence emerge.

Outline of Broad Issues and the Consensus Findings

How can we establish causes, measure and set doses that explain any association between exposure to stressors, psychiatric illness and cardiovascular disease?

A number of differing models and definitions of stress were considered. Stress may be referred to as a cause or as an effect and the term ‘stressor’ is gaining greater acceptance as a representation of the cause. Stress then refers to the psychological and physiological responses which result from experiencing a stressor. For the purposes of considering potential causal associations the term stress was considered ambiguous and difficult to quantify and the term ‘experiencing a stressor’ was preferred by the conference participants. The conference canvassed a number of opinions about the nature of stressors and the essential elements of stressors that lead to adverse outcomes in terms of psychiatric illness or cardiovascular disease.

The ex-service representatives provided the stimulus for the development of a definition of military stressor. This definition was formed on the DSM-IV definition of the stressor criteria required for the diagnosis of post traumatic stress disorder, the advice of the expert participants, (including that arising from systematic studies and data analyses, in particular Dr Terry Keane), and input from the ex-service representative observers at the conference. A stressor relevant to military service was defined by the conference participants as:

The person experienced, witnessed or was confronted with an event or events that involved actual or threat of death or serious injury, or a threat to the person’s or other people’s physical integrity that might evoke intense fear, helplessness or horror.

In the setting of service in the Defence Forces, or other service where the Veterans’ Entitlements Act 1986 applies, situations that qualify as stressors include:

Engagement with the enemy; or
Witnessing casualties or participation in or observation of casualty clearance, atrocities or abusive violence; or
Acute or chronic threat of serious injury or death; or
Prolonged experience of malevolent environments.

It was recognised that this definition of stressor was at the more severe end of the spectrum of psychosocial stressors that individuals can be faced with either in military service or in civilian life.

The participants recognised that less obviously severe stressors might contribute to morbidity but chose not to deal with data concerning such effects because of lack of consistent systematic definitions for such stressors in the scientific literature and the very diverse methodologies and outcome measures which allowed little opportunity for pooling of data or comparison.

Is there an association between exposure to stressors, psychiatric illness and cardiovascular disease?

The formal presentations, which included detailed literature reviews by Professors Tennant, Don Byrne, Esler and West as well as results of primary research findings, syndicate and group discussions demonstrated a range of opinion and highlighted the need to consider the potential for bias and particularly confounding in the available data sets. Professor Vaillant’s 50 year prospective data sets demonstrated that much retrospectively collected material may support mere association and not causation, and demonstrated the importance of known confounding variables such as alcohol and cigarette consumption when considering cardiovascular disease outcomes.

Consensus was reached on a number of psychiatric illnesses and cardiovascular diseases where associations between exposure to specified stressors and the illness was agreed to be causally related.

Psychiatric illnesses that may be associated with exposure to stressors are: post traumatic stress disorder, acute stress disorder, panic disorder, major depressive disorder, dysthymic disorder and alcohol dependence.

Cardiovascular diseases that may be associated with acute exposure to stressors are: sudden cardiac death and cardiac arrhythmias. The body of evidence regarding the association between hypertension and exposure to stressors was not considered indicative of a causal association.

Panic disorder was considered as potentially associated with certain ischaemic heart disease end points, most particularly sudden cardiac death.

A number of other associations were considered. However, high levels of co-morbidity and the potential for confounding weakened the case for causality and necessitated further detailed and critical analysis. Post traumatic stress disorder was considered of particular interest by the participants. However, the limited published data available, and the frequent co-morbidity with substance use and other disorders, precluded the attribution of any specific causal associations between PTSD and cardiovascular outcomes.

What are the potential mechanisms for the association between stressors and disease, with special emphasis on psychiatric illness and cardiovascular disease?

Conference participants discussed potential mechanisms explaining associations between stressors and psychiatric illness and cardiovascular disease.

In addition to the general propositions that stressors involving threat may lead to anxiety spectrum conditions and stressors involving losses may produce depressive spectrum conditions, the range of theoretical constructs underlying associations were considered and potential psychoneuro-endocrine and psychoneuro-immune pathways were acknowledged particularly in the contributions from Professors Allan Husband and Alexander McFarlane.

Professor Scott Henderson provided a thought provoking contribution on the role of genetics in both the likelihood of stressor exposure and potential disease outcome. Genetic and other vulnerability and the immediate biological response to stressor exposure may explain associations between stressors and certain psychiatric illnesses. Professor Najman’s research highlighted aspects of social adversity, lifestyle risk factors and stressor experiences. Dr Terry Keane’s re-evaluation of the US National Vietnam Veteran Readjustment Study data set provided support to the view that war zone stressors are associated with a range of psychiatric disease end points and that gender contributes to a differential pattern of disease outcomes.

The potential mechanisms invoked to explain associations between stressors and acute cardiovascular end points were discussed. Acute stressors may be linked to certain cardiovascular disease end points directly, or indirectly through certain forms of psychiatric illness. An example of a direct link is the onset of serious cardiac arrhythmias closely following the experience of a severe life stressor. Experimental stress models demonstrate high secretion of noradrenaline and increased sympathetic stimulation of the heart under conditions of mental stress. Noradrenergic and sympathetic stimulation of this sort can trigger arrhythmias in subjects who have pre-existing cardiac disease. An example of an indirect pathway is where an individual exposed to a stressor develops panic disorder and a panic attack can then lead to a serious arrhythmia or sudden death. The proposed mechanism here is that the panic attack causes excessive sympathetic stimulation of the heart which then leads to the arrhythmic event. Consensus was reached that a number of direct and indirect pathways may link acute stressors with certain psychiatric and cardiovascular outcomes. Overall, however, it was clear that the psychoneuro-endocrine responses associated with stressors are complex and are subject to considerable intra and inter-individual variability and are not fully elucidated at this time.

Conclusion

The RMA Consensus Conference, Stress and Challenge, Health and Disease provided a valuable opportunity for an informed debate about the relationship between the stressor experience, stress and psychiatric illness and cardiovascular disease.

Consensus was reached about a number of general proposals. It was agreed that modern epidemiological methods and the Bradford Hill criteria are appropriate to assess associations between certain stressor exposure and these disorders. A list of psychiatric and cardiovascular conditions were identified that could be causally associated with stressor exposure. A number of other psychiatric illnesses and specific cardiovascular end points were nominated for the RMA to investigate further.

Delegates from the following organisations attended the conference:

Repatriation Medical Authority, Australia
Australian National University, Canberra, ACT
Baker Medical Research Institute, Melbourne, VIC
Boston University School of Medicine, Boston, USA
Brigham & Women’s Hospital, Boston, USA
Department of Military Psychiatry, Oslo, Norway
Peter MacCallum Hospital, Melbourne, VIC
Prince Charles Hospital, Brisbane, QLD
Royal North Shore Hospital, Sydney, NSW
Royal Brisbane Hospital, Brisbane, QLD
The Queen Elizabeth Hospital, Adelaide, SA
University of Queensland, Brisbane, QLD
University of Sydney, NSW

Representatives from the following organisations attended the conference as observers:

Australian Federation of Totally & Permanently Incapacitated Ex-Servicemen & Women
Australian Veterans and Defence Services Council
Korean Veterans Assoc. of Australia
Legacy
Regular Defence Force Welfare Association
Returned and services League of Australia Limited
Vietnam Veterans’ Association of Australia
Vietnam Veterans’ Federation of Australia
War Widows’ Guild of Australia
Department of Veterans’ Affairs, Australia