Primary or secondary victim?

The Law Commission (1995) has stated:

It is indeed a conceptually confusing area and this distinction gives rise to real difficulties in practice. A primary victim – someone who suffers psychiatric injury due to his or her own injury or the threat of injury – can claim on proof of the same and that it was caused by negligence: no more is needed. A secondary victim has a whole separate set of requirements to discharge, as set out below. All this contributes to the intricacy of the legal maze, but two definitions given by Lord Oliver in Alcock v Chief Constable of the South Yorkshire Police [1992] are sufficient for present purposes:

• • a primary victim is someone ‘who is involved either mediately or immediately as a participant in an accident’

• • a secondary victim is someone who is ‘no more than a passive and unwilling witness of an injury to another’.

Alcock is the case of the Hillsborough Stadium football ground disaster. Relatives of some of the 96 Liverpool football fans who were crushed to death unsuccessfully sought damages for the psychiatric injuries which they suffered. Each of the claimants had either been present at the stadium, but remote from the terrace where the fatalities occurred, or had witnessed the event live on television or in later news broadcasts.

It should be noted that the claim of a secondary victim is not a derivative. The claim of the secondary victim does not depend on proof that the primary victim suffered physical or psychiatric injury. It may succeed in the absence of the same, so long as the defendant was in breach of a relevant duty to the secondary victim and that victim suffered relevant damage (again, see below).

The historical context

The jurisprudence relating to psychiatric damage generally, and secondary victim cases in particular, is all judge-made. It exemplifies classically the common law's approach of developing jurisprudence by accretion – an approach which has both the advantages and the disadvantages of flexibility and of the capacity to change to reflect developments in societal attitudes.

This area of the law evolved from cases in the 19th century which were burdened by an understanding of mental ill health fundamentally irreconcilable with the modern understanding. In Coultas v Victorian Railway Commissioners (1886) the Privy Council held that a nervous or mental shock ‘unaccompanied by any actual physical injury’ could not attract damages in an action for negligence. In this case it was the negligence of the Melbourne railway crossing keeper that resulted in a near miss as James and Mary Coultas crossed the railway line in their horse-drawn buggy, causing Mary Coultas to suffer ‘severe nervous shock’; she suffered a miscarriage and some months of physical illness.

There has been historic mistrust of psychiatric evidence on the part of the courts. As Lord Bridge observed in McLoughlin v O'Brien [1983]:

For a more detailed historical account see Mullany and Handford [Reference Mullany and Handford2006].

The control mechanisms

In Alcock, Lord Oliver identified several elements which had been found in the reported cases to be the essential criteria for a successful secondary victim claim, including most fundamentally (as recently emphasised in Liverpool Women’s Hospital NHS Foundation Trust v Ronayne [2015], hereafter referred to as Ronayne) that the claimant should have suffered frank psychiatric illness or injury. These have been described as ‘control mechanisms’ (see Box 1). They are so called because, to quote His Honour Judge Denyer QC in Morgan v Somerset Partnership NHSFT [2016]:

As Mrs Justice Swift DBE stated in Shorter v Surrey and Sussex Healthcare NHS Trust [2015]:

If the application of these control mechanisms results in what seems to be, or is, unfairness, this is recognised by the courts:

However, this does not mean that the courts will not try to be fair. As Lord Justice Ward said in North Glamorgan NHS Trust v Walters [2002]:

What is a frank psychiatric illness or injury?

One of the leading secondary victim cases is that of McLoughlin. It is important to recall its facts, which were extreme (Box 2). As Lord Wilberforce commented, these circumstances were capable of producing an effect going well beyond that of grief and sorrow. The significance of this point lies in the judgment of Lord Bridge in this same case:

As Lord Oliver put it in Alcock:

Herein lies the requirement for the courts to be assisted by experts in psychiatry. As Lord Steyn put it in White v Chief Constable of South Yorkshire [1992], the unsuccessful case brought by police officers who had suffered PTSD caused by their experiences in tending primary victims or otherwise dealing with the bodies of primary victims at the Hillsborough Stadium disaster:

Although the use of ICD or DSM diagnoses is not strictly speaking essential, the requirement that there should be significant distress and impairment of functioning means that the legal test effectively requires a diagnosis recognisable to the psychiatric profession. (Note that even such a diagnosis will only be sufficient if it is of a condition which satisfies the requisite features in terms of its causation – essentially, that it was caused by the shocking nature of the event to which the individual was exposed – see further below). Although ‘recognisable’ does not mean the same as ‘recognised’, it is worth noting the approach of the court in Hussain v The Chief Constable of West Mercia Constabulary [2008]:

It is also worth noting the advice of Marshall et al (2012):

This is illustrated by the case of Rorrison v West Lothian College (1999). The pursuer, in a negligence case, suffered severe anxiety, panic attacks, loss of confidence and loss of self-esteem, identified by a clinical psychologist, but because she ‘had not pleaded any disorder that was recognised in DSM-IV’ and had not been diagnosed by a psychiatrist as suffering from any recognised psychiatric disorder, her case was dismissed.

What is shock?

As we have already made clear, it is not sufficient to prove the existence of positive psychiatric illness or recognisable psychiatric harm. That illness or harm must have arisen in a particular way. It must have arisen as a result of what the law terms ‘nervous shock’ (Box 3). The term has been described as unfortunate because at first glance it seems to refer to the actual experience of being shocked (Rix Reference Marshall, Kennedy and Azib2011). The courts recognise this:

So the issue is not just as to diagnosis as such, but also as to aetiology and causation.

The essential ingredients are the suddenness of the experience, its horrifying nature and its direct perception through senses such as sight, hearing or touch. ‘Shock’ here has a specific meaning in law, not dissimilar to the exceptionally threatening or catastrophic nature of the threshold criterion in PTSD, and its meaning is not the same as the colloquial meaning:

An event of which the person has some warning and gradually cumulative events will not satisfy the suddenness criterion. In Sion v Hampstead Health Authority [1994] the claimant failed to recover damages when his son died 14 days after a road traffic accident. He stayed at his son's bedside, he watched his deterioration, he saw him fall into a coma and die. He was witness to a process that continued for some time and, when death occurred, it was not surprising but expected. Likewise, in another unsuccessful case, where a 14-year-old boy died following a road traffic accident, his parents experienced ‘a dawning consciousness that they were going to lose him’ (Taylorson v Shieldness Produce Ltd [1994]). In Alcock it was held that nervous shock was yet to include psychiatric illness caused by the accumulation over a period of time of more gradual assaults on the nervous system.

A distressing experience that is not sufficiently horrifying to agitate the mind violently will not satisfy the horrifying criterion. In Ward, where the claimant's daughter died as a result of medical negligence, the judge found that the claimant's descriptions of what she witnessed did not strike him as shocking at the time, although undoubtedly they were distressing. The experience had to be wholly exceptional:

Thus, in Ronayne (see further below) the court found that:

Furthermore, what is ‘horrifying’ has to be judged by objective standards and by reference to persons of ordinary susceptibility (Owers v Medway NHS Foundation Trust [2015]). What Mr Ronayne saw was not horrifying by objective standards.

A similar issue arose in Shorter, where the claimant was a senior sister in neurointensive care. Mrs Justice Swift DBE said:

Hearing about the event in a telephone call will not satisfy the direct perception criterion. In Brock v Northampton General Hospital Trust [2014], where the claimant's daughter died as a result of a negligently treated overdose, the claim failed because the claimant said that it was when the telephone call was received that she realised that her daughter was indeed going to die and ‘it is one of the fine distinctions that a telephone call giving bad news cannot found liability to secondary victims'. As established in McLoughlin, shock must come through sight or hearing of the event or of its immediate aftermath.

The frank psychiatric injury element and the nervous shock element go together to form the requirement, first, that the event should be not just sufficiently sudden and shocking but that it should in fact cause the psychiatric injury and, second, that it should do so because of its sudden and shocking nature. The Royal College of Psychiatrists regards the ‘shock-induced' requirement as causing serious problems and has observed:

As a matter of analysis, both of us – psychiatrist and lawyer – agree whole-heartedly with this sentiment. However, this is the law.

Proximity

The law also requires there to be a sufficiently proximate relationship between the tortfeasor, i.e. the person who has caused the wrong, and the secondary victim. Proximity in the law of negligence generally describes the relationship between parties which is necessary in order to found a duty of care owed by one to the other. This explains the failure of a bystander to recover damages when she heard the noise of a fatal motorcycle accident; it was accepted that she had suffered ‘nervous shock' but the motorcyclist owed her no duty (Bourhill v Young [1943]). But in secondary victim cases, the term proximity is also used in a different sense to mean physical proximity in time and space to an event. In this sense it operates as one of the control mechanisms. In a secondary victim case, physical proximity to the event is a necessary, but not sufficient, condition of legal proximity.

In Taylor v A. Novo (UK) Limited [2013] a woman was injured at work when shelving fell on her. She suffered an apparently minor head injury and an ankle injury. She attended hospital and was discharged after a day or so. Some 3 weeks later she suddenly collapsed and died as a result of an unexpected pulmonary embolism caused by a deep vein thrombosis attributed to the original injury. This happened in the presence of her daughter, Ms Taylor, who suffered PTSD as a result. The psychiatric evidence was not in issue. There was clearly a relationship of legal proximity between Novo and their employee. However, the Court of Appeal found that there was no such relationship between Novo and her daughter: or, to put it another way, she was not within their contemplation as someone who might suffer damage as a result of its actions. Her claim as a secondary victim failed as a result. That is because she was not present at the scene of her mother's accident at work or any scene that might sensibly be thought to be part of its immediate aftermath. If she had been allowed to recover damages for what happened 3 weeks after her mother's accident, it would logically have followed that she would have been able to recover damages for psychiatric illness even if her mother's death had occurred months, and possibly years, after the accident (assuming she could prove causation); the court considered it unreasonable to stretch so far the concept of proximity to a secondary victim. Thus, the court decided that allowing Ms Taylor to recover damages would extend the scope of liability to secondary victims considerably further than had been done hitherto and the policy reasons articulated by Lord Steyn in White to confine the right of action of secondary victims by means of strict control mechanisms militated against any further substantial extension.

There is no doubt that, to some, this judgment seeks to draw absurd distinctions: since the shocking event did not take place until 3 weeks after the original tort, why should that delay be taken into account when considering whether the secondary victim shocked by that event has a claim? It can only sensibly be understood if seen through the prism of the requirement for legal proximity. This in itself may by some be thought to raise a question about the application of that requirement to situations such as this. In any event the case is clearly illustrative of the apparently arbitrary way in which the control mechanisms operate.

Reasonable foreseeability and vulnerability

In Page v Smith [1996] (note that this is a case of psychiatric injury suffered by a primary victim – i.e. as a result of perceived danger to himself) Lord Lloyd specifically defined reasonable foreseeability as a control mechanism. It is thus an inherent control mechanism in the tort of negligence more generally (in the context of duty and remoteness of damage). This means that, in order for there to be a breach of duty, the circumstances must be such that a person of what the law terms ‘normal fortitude' or ‘ordinary phlegm' might suffer psychiatric injury by shock. In other words, it must have been reasonably foreseeable that a person of normal fortitude would be affected by what occurred. However, this does not mean that someone who is not of normal fortitude or ordinary phlegm (someone with, to adapt a phrase from the law relating to physical injuries, an ‘eggshell mind’) cannot recover damages (whether as a primary or as a secondary victim):

Thus, the person of normal fortitude test is used to establish the threshold in terms of breach. If the threshold is reached, a person who is not of normal fortitude may have a claim, albeit that but for their vulnerability they might not have suffered psychiatric injury – and if they have suffered more injury as a result of their vulnerability their damages are not limited to those which a person of normal fortitude would have been awarded.

North Glamorgan NHS Trust v Walters [2002]

When he was aged 10 months, Mrs Walters’ son became unwell and was admitted to hospital. He was mistakenly diagnosed as suffering from hepatitis A. In fact, he was suffering from acute hepatitis, which led to liver failure. The NHS trust responsible for the relevant hospital admitted that he had not been properly diagnosed or treated and that, if he had been, he would have been given a liver transplant and would probably have lived. In the event, the baby was kept in hospital while various tests were carried out, but was allowed home at weekends. Box 4 sets out the subsequent events that culminated in the baby's death.

The psychiatrists who gave expert evidence agreed that Mrs Walters had suffered a recognised psychiatric illness, namely pathological grief reaction. They also agreed that, absent the events that she witnessed, experienced and participated in over the period of her son's illness, her pathological grief reaction would not have occurred.

The question for the trial judge was whether what happened was a sudden appreciation by sight or sound of a horrifying event, rather than an accumulation over a period of time of more gradual assaults on the nervous system, and whether it was that sudden appreciation that caused the pathological grief reaction. Judgment was given in Mrs Walters' favour and the defendant hospital appealed.

In the Court of Appeal, judgment was again given in favour of Mrs Walters. Lord Justice Ward found that:

This case established the principle that, if the necessary sudden shock is there, it does not prevent recovery of damages that the events themselves were drawn out over 36 h.

This case is also significant in that it is one of probably only two successful ‘hospital cases'. Hospitals are in a very particular position in relation to the psychological responses of individuals due to the illnesses or injuries of others even before one gets to the secondary victim issues. The day-to-day work of hospitals involves people who are injured or ill or, in the case of obstetrics, healthy but vulnerable. They go to hospital followed by their loved ones. Normally that will be because of an injury or illness that occurred elsewhere, whether actionable or not. Sometimes it is because of something that happens in hospital that will sometimes be actionable on the part of the primary victim. But it does not follow even then that any secondary victims have a claim too. First, in hospitals the nature of the triggering event – illness/injury (whether or not caused by negligence, clinical or otherwise) – may well be qualitatively different in terms of whether sudden shock is involved: the event itself may not be perceived at all, for example infection, haemorrhage, etc.; and in any case the trigger is less likely to be sudden: more likely gradual onset. Second, it is, bluntly, in the nature of hospital visits that you can expect bad news: you can expect to see drips, monitors and sick people. This is important and often overlooked. The general point is that experiences of this sort in hospital may be deeply distressing or upsetting, and may well give rise in some cases to psychiatric disorder, but they are less likely to be a ‘sudden assault on the nervous system’ than something shocking that happens in the street or at work. Thus, if the requirements for ‘sudden shock’ etc. are absent, perhaps inevitably because of the nature of (say) the clinical negligence concerned, then that will indeed make recovery of damages that much harder.

It is for this reason that, contrary to what one might optimistically call popular belief, and notwithstanding that historically some settlements have been achieved without going to trial, there are in fact very few ‘hospital cases’ in which secondary victims have been successful at trial. One of these was Walters (see above). Another was Tredget v Bexley Health Authority [1994] – where a baby was born alive but the negligence which inflicted the ultimately fatal injuries on the baby took place during labour in front of the father in circumstances described as ‘chaos’ and ‘pandemonium’ in the delivery room. Both of these cases clearly involved wholly exceptional facts, which explain their respective outcomes (see the observations of Lord Justice Tomlinson as to Walters in Ronayne, cited below).

Young v MacVean [2015]

In this case (Box 5) it was held that Mrs Young sustained psychiatric injury, but hers was of the nature of an illness caused by the accumulation over a period of time, albeit a relatively short time, of more gradual assaults on the nervous system. She suffered no shock as a result of seeing the aftermath. She had become worried but worry is not the same as shock. Although she witnessed the aftermath, it was not the aftermath which caused the shock. What caused the shock was what she was told, so there was no direct appreciation of the event through sight or sound.

Owers v Medway NHS Foundation Trust, Secretary of State for Health [2015]

Mr Owers witnessed a deterioration in his wife's condition following a stroke which was not diagnosed and treated sufficiently promptly. The psychiatric evidence is set out in Box 6. The court found that the hospital's breach of duty caused his PTSD and that, but for the breach, he would have suffered depression and adjustment disorder in any event. However, his claim was unsuccessful on the basis that although his experience was ‘very distressing [there] was no sudden appreciation of a “horrifying” event’.

Liverpool Women's Hospital NHS Foundation Trust v Ronayne [2015]

The details of this case are set out in Box 7. A clinical psychologist called by the claimant diagnosed Mr Ronayne as suffering from PTSD. A psychiatrist called by the defendant was initially of the opinion that he had suffered no psychiatric disorder, only anger and stress falling short of psychiatric illness, but at the trial maintained that, if his functioning was impaired, the diagnosis was one of a mild adjustment disorder. The trial judge decided that the claimant was not suffering from PTSD. He did not want to label the condition, but said that if he had to do so he would find that it was an adjustment disorder. He found for Mr Ronayne on the basis that he had suffered a frank psychiatric disorder.

The effect of this judgment at first instance in Mr Ronayne's favour – as it stood – was significantly to extend recovery of damages by secondary victims, in particular in the hospital context. This was because the claim succeeded (a) simply in the context of a psychological reaction (unspecified in terms of diagnosis) to distressing events, (b) absent any sudden and shocking event, and (c) absent any evidential basis for a finding that Mr Ronayne's psychiatric illness was as a result of the sudden shock of that event.

The defendant hospital appealed. The issues on appeal were (a) whether the events concerned were of a nature capable of founding a secondary victim case, i.e. whether they were, in the necessary sense, ‘horrifying’, and (b) whether in fact the sudden appreciation of that event or those events, i.e. shock, caused the claimant's psychiatric illness. To put it another way, perhaps more pithily: were the events here objectively of a sufficiently sudden and horrifying nature and, if they were, was it in fact that sudden and horrifying nature that caused the illness?

On appeal it was held that the trial judge:

Lord Justice Tomlinson decided that the claim fell at the first hurdle, so it was unnecessary for the court to decide on the causation issue – i.e. whether the trial judge was justified in finding that it was the appearance of his wife on these two distinct occasions, as opposed to his wife's ill health, which caused the adjustment disorder. However, he felt it right to record that he was very doubtful about the trial judge's conclusion in this regard. He then went on to analyse the expert psychiatric and psychological evidence as to causation in more detail (Box 8). This analysis illustrates the particular challenge for such expert witnesses.

The main problem here is of course that, in theory at least, the court needs to know whether the psychiatric injury was caused simply by the shocking nature of the events, or by other factors, or by some combination. That obviously raises the question of what happens if (as is in reality highly likely in most cases) it is a complex aetiology which is the result of a combination of the shocking nature of events and of other factors as well.

The authorities do not provide any answer to this question, and each case is currently resolved on its own facts and merits. However, one possible answer for the future which seems to us to make sense would be for the law to adopt and apply here the concept of ‘material contribution' (familiar in clinical negligence cases when medical science can identify the fact that a tort has caused damage but has insufficient data to identify the extent of that damage). In this event the claimant would recover damages if he or she can show that the sudden shock of the objectively shocking event made a material contribution to his or her psychiatric condition, even where it is impossible to disentangle the strands of the complex aetiology concerned. However, this is simply speculation by us.

Shorter v Surrey and Sussex Healthcare NHS Trust [2015]

Mrs Shorter brought a claim after her sister died as a result of a negligently managed stroke. She was herself a senior sister working in neurointensive care. She had a number of experiences culminating in her witnessing her sister's death (Box 9). Her claim was unsuccessful, the trial judge holding that:

Like the judgment in Ronayne, this is also a judgment that illustrates the relevance of psychiatric evidence to causation (Box 10). It illustrates the key role of visual experiences such as flashbacks, the need for a forensic analysis of the different accounts that the claimant gives of their experiences, the need to take into account the claimant's reaction, or otherwise, to previous shocking experiences and the reliance placed on observation of the claimant's mental state when giving evidence in court as to their experiences.

Morgan v Somerset Partnership NHSFT [2016]

On 26 October 2011, Mrs Morgan's husband attempted suicide and was assessed by one of the defendant trust's employees. On 1 November 2011, she had gone to work, she came home and there was a note on the table from her husband indicating that he was going to commit suicide. She went into the garage and found him there. He had cut his wrists. There was a significant amount of blood. He was subsequently taken to hospital; happily, he survived. Mrs Morgan suffered an adjustment disorder consequent on finding her husband in that state. She claimed negligence against the NHS trust whose employee had attended him on 26 October. However, the claim failed; the judge found that there was ‘simply not the regular degree of proximity in terms of time and space between the original tort, vis-à-vis the Defendants and the Claimant's husband, the primary victim, and the loss or rather the damage suffered by the Claimant […] that is necessary in order to get this type of claim off the ground'.