Abstract
There has been a continuing interest in recent years in the relationship between stressful experiences and illness both psychological and medical. The relationship is pivotal in relation to litigation for liability for psychiatric disorder. Psychiatrists can become the expert ‘hired guns’ in the subsequent litigation, some there to establish that a stressor was significant, that a psychological disorder ensued and that there was a causal link, others to refute such an argument. Concerns about this process have been expressed by members of the judiciary and discussed in the press.
Because of the nature of psychological disorder and the somewhat limited evidence basis for our understanding, it becomes easy to fill the void in knowledge of causality with a range of plausible but sometimes spurious psychological explanations which can be the basis for widely differing psychiatric opinions. Judicial decisions may at times thus be unduly influenced by the relative strength of clinical conviction, the plausibility of psychiatrists' arguments or their apparent expertise. Although personal clinical judgement is indeed most important, the fact that empirical data is increasingly available in assessing first the severity of stressors and, second, the links to psychological disorder, requires that clinical opinion should whenever possible be supported by the best available empirical evidence, from these domains.
Psychological impairment
The assessment of psychological disorder now has considerable reliability (both interrater and retest) with the development of diagnostic criteria such as DSM-IV and ICD-10. Judging severity of emotional distress or disorder is thus usually reliable provided patient reports are truthful and not due to malingering or embellished for other reasons. The assessment of the consequent social impairment, however, may be less reliable because criteria are more poorly defined [1] and, in the medico-legal setting, because these patient self-reports of impairment may be easier to embellish and thus more influenced by legal proceedings [2], be this with conscious intent or otherwise. Depressive disorders provide the best example for discussing the assessment of stressors in a medico-legal context, because depression is the most common of stressor-related psychological disorders and because it is the commonest reason for psychiatrists' testimony in litigation.
The significance of stressors
Any evaluation of the psychological impact of an adverse and stressful experience should be largely based on common sense: namely the expectation that most individuals of similar background would be likely to experience a similar adverse emotional response. Indeed there exists generally good agreement among people in the community as to what may be significantly stressful [3]. Furthermore, there is acceptable agreement about the severity of most stressful experiences when populations are studied cross-nationally [4]. This is not to say that social, cultural and individual differences do not significantly influence the perceptions, or indeed impact, of some experiences [5], but most individuals and groups can agree on a range of life stressors which would be universally stressful. More important perhaps, mental health workers such as psychologists, psychiatrists and social workers in a research setting show strong agreement in judging severity of most life events [6].
In judging the personal impact of a stressor, an extensive appraisal of the individual's psychological and social vulnerability is, of course, important. A critical issue, however, emerges when one moves from the more objective aspects of an individual's stressor experience to the patient's subjective appraisal of it and his or her self-reported emotional response to it; problems can clearly arise. In the context of medico-legal proceedings in particular, the final psychiatric evaluation may be based on quite distorted information. First, a patient will perceive an experience as far more stressful when depressed or distressed, than when not [7]. Second, this may be especially important as there is evidence that symptoms including functional somatic symptoms [8], lower back pain [9] and posttraumatic stress disorder are more severe during legal proceedings although this is also disputed [10]. Third, the patient's report of the impact of a stressor may be maximized when the stressor is compensable while the reporting of coexisting non-compensable stressors or their impact may be minimized, leaving a distorted view overall of the recent experience of stressors. Using a ‘common sense’, population-based approach as a foundation for a more individual clinical approach adds a useful dimension and may reduce some of the potential biases in assessment of stressors.
Evidence from research studies of both ‘normal’ and clinical populations shows one can reliably judge the severity of a stressor using this common sense approach and examples are provided to demonstrate this (see [11]). This method for stressor assessment involves the taking of a detailed life-stressor history, separately identifying each acute or chronic stressor, and describing and assessing the wide-ranging social context in which it occurs. These social context variables have been themselves empirically linked to psychological disorder and include gender [5], age [12], social status [13], finances [13], employment [14], number and age of dependants [15], social supports/isolation [16], social relations [17], physical health [18] and education status [16]. Any stressors which could have been caused by psychological disorder are excluded from consideration. This type of interview and rating of stressors is reliable, contemporary, personalized and yet remains relatively objective. As it does not actively embrace the individual's own self-report of his or her emotional response to the stressor, it avoids major subjective bias and has a more normative and objective basis. In the research setting, furthermore, the life-stress interview and the psychiatric interview are administered separately. This approach, perhaps modified, could be a useful adjunct to the more usual individual/subjective clinical method, which can embrace ‘vulnerability hypotheses’ which may have no empirical basis. It could perhaps be used as a starting point for which any significant deviation(s) would require very strong justification.
This common sense approach at times may seem incompatible with the medico-legal process given the different objectives of the litigating parties. This adversarial system encourages the plaintiff to demonstrate his or her particular personal vulnerability to a stressor (which may not be significant according to any common sense approach) while the defence aims to refute this argument (even in circumstances when a stressor could be significant). Psychiatrists, like barristers, may be influenced by who is paying the piper, and indeed some psychiatrists practice solely as either plaintiff or defendants' expert. Their objectivity is therefore questionable. In the assessment process either a sympathetic or more inquisitional approach may be used, and, as a consequence, particular findings may be emphasized or minimized depending on which party is paying the fee.
The relationship between stressor and illness
Assessing the relationship between stressors and psychological disorder in any clinical or medico-legal setting is distorted by several factors. In psychological illness, seldom is there obvious cause and effect as one commonly finds in trauma-induced physical injury. Because the findings can commonly be confounded, relevant principles of causality used in research might also usefully be applied even in the assessment of individual cases. These four causal principles include temporal sequence, the strength of the association, the consistency (or replicability) of association and, finally, coherence of findings. The first, establishing the true temporal sequence of stressor and depression, is not always simple. First, depression commonly has an insidious onset and may not initially be recognized by the patient; only the added burden of a traumatic experience, as a final straw, may trigger either the recognition of disorder, and/or a decision to present or seek help [19, 20]. Second, a covert depression or depressive prodrome may, in fact, be the unrecognized cause of some stressful events; this is especially apparent with interpersonal stressors. Such stressors can then further reciprocally generate depression [21] or indeed other stressful events.
The second principle of causality relates to the real (rather than ‘apparent’) strength of the association; in particular, could the relationship one observes be confounded, at least in part? For instance, those with personality disorders or significant traits are more likely both to perceive stressors as significant [22] and to generate stressors [23], and also be constitutionally prone to depression [24]. A common (non-compensable) factor can cause both the stressor and the disorder. Even when personality factors generate stressors, which then trigger or contribute to depression, the stressor may remain an important mediating and probable causal variable, but may be less relevant in relation to liability of others. There is similarly evidence that the same genetic factors which predispose to depression in women [25], men [26] and adolescents [27] may also predispose to the occurrence of stressors. In this case the relationship between stressor and depression might be in part spurious, indeed one twin study indicated approximately 30% of the relation between stressor and depression was in fact genetically determined [28]. Distinguishing whether a stressor is spurious or truly causal may prove difficult, indeed it is even possible that the same stressor could both have in part a spurious (‘genetic’ or personality-based) relationship to depression and in part a true causal association. In addition, these same subjects (who may appear to have some ‘compensable stressor’) are also more likely to have generated other (‘non-compensable’) stressors which may also be contributing to their disorder. Finding the degree to which a disorder can be attributed to the mix of potentially compensable and non-compensable stressors may prove exceedingly difficult.
Individual vulnerability to stress and special pleading
It is recognized that all healthy, well-adjusted individuals are vulnerable to the risk of significant (usually transient) psychological disorder following catastrophic stress [29]. Equally, it is rare for healthy, well-adjusted individuals to suffer a sudden onset of clinical depressive disorder of any significant duration or severity following relatively trivial stressors [30]. Notwithstanding this, the individual's perception of and/or vulnerability and response to stressors, manifest as the ‘fragile egg shell syndrome’, often becomes pivotal in litigation and may be fuelled by some degree of collusion between patient, lawyer or clinician. Disproportionately more of these ‘disputed’ vulnerable cases or ‘outliers’ are indeed contested in court; those where the stressor was obviously severe are less often contested.
In the litigation setting one should be especially cautious where there seems to be undue or excessive vulnerability, that is, the stressor does not appear particularly severe in usual common sense terms and a ‘fragile egg shell syndrome’ is proposed. Any justification for this vulnerability should be supported when possible by other objective data, rather than by non-validatable, ‘psychological hypothesis generation’ alone. For example, if special vulnerability is proposed, it would be important to identify earlier episodes of significant and similar disorder occurring in response to similar ‘low severity’ stressors and particularly in litigation-free settings. A first occurrence of vulnerability to a ‘non-severe’ stressor in a litigation context should be viewed with considerable caution since mild and moderate stressors are common in everyday life; it begs the question of why the patient has not decompensated in response to earlier and similar mild-to-moderate stressors. This, to some extent, reflects a third causal principle, namely consistency of association or ‘replicability’ within an individual's life history. Similarly, one should be able to argue that other subjects with a similar vulnerability and experiencing a similar stressor would also decompensate; this would demonstrate some consistency of association across individuals. This underpins the issue of ‘probability’ or likelihood; not only is it a statistical concept, it clearly has clinical and legal implications.
In relation to non-compensable stressors, it is highly likely that other (non-compensable) stressors have also occurred (given the ubiquitous nature of life stress) and it may prove difficult to distinguish their separate contribution to the disorder, especially in ‘vulnerable’ individuals. This issue of relative attribution of different stressors becomes even more complex when one considers that once a stressor-induced episode of depression appears to have occurred, a further episode can be provoked by a subsequent less severe stressor [31, 32]. When either one of these stressors is compensable and the other not, determining the degree of causality for the ‘compensable’ component of the illness episode is problematic. Furthermore, personality disorder or traits, as discussed earlier, may further compound the above problems.
Finally, it is important to consider the fourth relevant causal principle, that of coherence of explanation: the explanation proposed by the clinician should fit readily with other accepted understandings of aetiology of this disorder. The approach using these notions of causality as described could lend considerable strength to clinical evidence presented.
Opinion
Significant biases can arise in the medico-legal assessment of stressors and psychological disorder. These include patient's reports distorted in some cases by selfjustification or the prospect of financial gain, by anger and frustration in others and by distress in most.
Notwithstanding who pays the fee, the psychiatrist supposedly is the agent of the court, yet one finds that their different agendas can influence their interpretation of what may be a very complex relationship between stress and disorder: it is easy to understand the resulting conflict in psychiatric testimony and why at times the objectively implausible can appear plausible.
Psychiatrists have a number of roles in the court setting. First, clinicians are there to present expert evidence of findings concerning the presence of psychological illness, and its associated impairments. Second, they may be required to present evidence concerning the likely psychological impact of recent stressful events. Third, they may need to provide good clinical and other evidence for that individual suggesting a causal association, support this with empirical evidence whenever possible and, finally, apply relevant causal principles. The problems discussed might be further minimized if the psychiatrist first, truly evaluates his or her position about possible bias, second, is aware the relationship of stressor and disorder is likely to be complex, and third, has taken a critical stance in relation to possible ‘special’ vulnerability.