Criminal Causation and the Careless Doctor

Transcription

1 Crirninul Causation and the Cureless Doctor John E. Stannard * Issues of causation rarely crop up in criminal cases. The reason for this is obvious. Where causation is in issue, a defendant will be trying to disclaim responsibility for certain consequences of his actions. However, a person accused of serious crime will not, as a general rule, be answerable for consequences unless he intended them or was, at the very least, reckless as to whether they would occur. This means either that the consequences must have been at least foreseen by the defendant, or at any rate that they must have constituted an obvious risk arising from the conduct in question. Where this is proved, it will be virtually impossible to argue that the consequences were not caused by the defendant. Where it is not, the charge will fail because of the lack of mens rea, and it will therefore not be necessary to worry about causation. In this context, homicide is the exception that proves the rule. A defendant can be convicted of manslaughter or even murder in respect of a death that was neither intended, nor foreseen, nor even an obvious consequence of his conduct.? Indeed, he can be held liable for manslaughter even in cases where he foresaw no harm to the victim at all.) This being the case, an argument based on causation may become highly relevant. One of the most common situations of this sort is where D (the defendant) has caused injury to V (the victim), so that V requires medical treatment. If at the end of the day V dies, can D argue that the death would not have occurred had it not been for the medical treatment (or lack of it)? In the past, this defence has rarely been successful in English law.4 The recent decision of the Court of Appeal in Cheshire5 shows that this situation is not likely to change. The Facts of Cheshire The defendant, Cheshire (D), was involved in an argument with another man, V, in a Greenwich fish and chip shop early in December D produced a pistol and fired it at the ceiling, following which a struggle took place in which D shot V, wounding him severely in the thigh and stomach. V was subsequently taken to hospital where he was kept in intensive care and given extensive surgical treatment. In particular, following respiratory problems, a tracheotomy was carried out and a tube inserted into V s neck, where it remained in position for some four weeks. Following the removal of the tracheotomy tube in the middle of January 1988, V s condition began to improve to some extent. However, he continued to complain of difficulty in breathing and, during the second week in February, was seen by *Lecturer in Law, Queen s University of Belfast. I Strict liability is. of course, an exception to this. An intention to do grievous bodily harm is enough for murder; D need not contemplate any danger to life: Cunningham [I All ER 863. Where death occurs as a result of D s unlawful act. D will be guilty of manslaughter if the act was one which any sober and reasonable man would realise was likely to cause harm: Church [ 1966) I QB 59. DPP v Newbury and Jones [ 1976) 2 All ER 365. Jordan (1956) 40 Cr App Rep 152 is the exception. [ All ER

2 The Modern Law Review [Vol. 55 several doctors of varying experience. However, an X-ray failed to reveal any lung trouble and the problem was put down to anxiety. On the evening of 14 February, V again complained that he could hardly breathe and was seen by a junior house surgeon, J. J was concerned about V s condition and noted that his breathing was rasping and laboured (a condition technically known as stridor ). Shortly afterwards, V took a turn for the worse and, despite attempts to resuscitate him, he died. It was not until the post-mortem that the true problem became clear. Following the removal of V s tracheotomy tube, V s windpipe had become obstructed: a rare but not unknown complication of such treatment. Indeed, the windpipe had become so constricted that even a small amount of mucus could block it and cause asphyxiation. No doubt this problem could easily have been alleviated had it been spotted in time, but it had not. D was subsequently charged with murder. Not surprisingly, he sought to argue that he was not criminally responsible for the death and that the negligent medical treatment sustained by V had broken the causal link. During the trial, the court was confronted with a direct conflict of expert testimony on this issue. According to the pathologist who had conducted the post-mortem, the death was caused by cardio-respiratory arrest due to gunshot wounds of the abdomen and leg. 6 But D s expert witness testified that death was due to the failure [of the doctors] to recognise the reason for his [sic] sudden onset and continued breathlessness after the 8th February [and the] severe respiratory obstruction, including the presence of stridor. Given this conflict of evidence, how was the jury to be directed? In a sense, it is hard not to sympathise with D. After all, (1) the original shooting had taken place over two months prior to the death, (2) the injuries received in the shooting were on the mend and no longer threatened V s life,s and (3) there seemed to be at least prim facie evidence of some negligence on the part of the doctors. Certainly, it seems that V would have made a full recovery had it not been for the doctors failure to spot the obstruction in his windpipe: an obstruction which, moreover, had been brought about by their own treatment. However, the trial judge was uncompromising in his direction to the jury. Even if the medical treatment was incompetent and negligent, D s responsibility for the death was not affected. The only situation in which the chain of causation could be broken and D absolved would be if the medical treatment was reckles~, ~ ie if the doctors had acted with indifference,i0 not caring less as to the consequences of their acts. On this direction the jury convicted D, and it is indeed hard to see how they could have done otherwise. The defendant appealed, arguing that the trial judge had been far too restrictive in his direction as to when careless medical treatment could break the chain of causation. Indeed, he had virtually withdrawn the issue from the jury,il given that on his interpretation careless medical treatment could virtually never break the causal link. The question for the Court of Appeal was whether the trial judge had correctly applied the principles of causation. 6 ibid at p ihid at p ihid at p ihid at p I I This is the meaning adopted in Surnum (1983) 78 Cr App Rep 149. and suggested by the present writer at (1985) 101 LQR 540, at pp I All ER at p

3 Principles of Causation Causation is said to be a question of fact for the jury, but which is to be decided on the basis of guidance given by the judge.12 The criminal law has traditionally insisted on D s act being both a factual and a legal cause of V s death before D can be held legally responsible for it. The first issue, that of factual causation, involves the prosecution in proving that the death would not have occurred when it did but for D s acts,l3 and that the acceleration of death brought about by D was more than merely trivial.i4 This issue is unlikely to arise in the type of case presently under di~cussion,~5 given that the medical treatment in question will rarely, if ever, prove to be necessary in the absence of an injury inflicted by D. Legal causation gives rise to more problems. As a general rule, it can be said that (assuming D s acts to be a factual cause of V s death) D will be held criminally responsiblenh if one of three criteria are satisfied. The first principle is that, assuming factual causation to be present, D will remain liable despite intervening causes if, at the time of V s death, the original wounds or injuries inflicted by him were still an operative cause of that death. This principle is illustrated by a line of cases in which D was convicted of murder or manslaughter, despite the death having been brought on by V having been badly treated or not at all. In Smith,In D was charged with murder after having wounded V in a barrack room brawl. In his defence, D argued that death would not have occurred if the wound had been properly treated: as it was, V had been dropped several times on the way to the dressing station, was neglected for some time when he arrived, and was finally given thoroughly bad 19 medical treatment which may have destroyed his chance of recovery. Despite all this, D s conviction was upheld on the ground that the original wounds were still an operative cause of V s death. Again, in Bluue,? ) the injuries inflicted by D necessitated a blood transfusion, which V refused on religious grounds. One of the grounds2 for upholding D s conviction for manslaughter was that V s decision, however unreasonable, did not alter the fact that the original wound was still a substantial and operating cause of V s death. And in Mulcherek,22 D was not allowed to argue that the true cause of V s death was the decision of the doctors to disconnect the life-support machine on which V had been placed following complications arising from the original injuries inflicted by D. It was D, said the court, and not the doctors who were on trial, and there was no denying that the original injuries remained an operative cause of the death. Even where the original injuries are no longer operative, D will still not escape responsibility for the ensuing death if it was reasonably foreseeable that death would Pagett (1983) 76 Cr App Rep 279. In Malcherek [I All ER 422, the trial judge was held to be justified in withdrawing the issue of causation from the jury. If causation is a question of fact, this is contrary to DPP v Stonehouse [ AC 55. Often called but-for causation or causa sine qua non : see White [I KB 124. Smith and Hogan, Criminal Law (6th ed, 1988) p 316. It may, of course, be crucial in the converse case where it is the liability of the doctors that is in issue: see Barnett v Chelsea and Kensington Hospital Management Committee [ I QB 428; Hotscm v East Berkshire Health Authority [ All ER 909. Depending, of course, on his degree of fault. Smith and Hogan, op cit pp QB 35: contrast Jordan (1956) 40 Cr App.. Rep i1959j 2 QB 35. at p All ER 446. Comoare Holland (1841) 2 Mood & R 351. For the other, see below p.570. [I All ER

4 The Modern LAW Review (Vol. 55 result from his acts. This is the well-known body on the seashore principle23: if D renders V unconscious and leaves him lying on the seashore, he will be legally responsible for V s death if he is subsequently drowned by the incoming tide, but not if V is struck by lightning or swallowed up in an earthquake. Why? Because the former result is foreseeable as likely to occur, but not the latter. To put it another way, leaving V unconscious on the seashore increases the risk of his being drowned, but does not increase the risk of his being struck by lightning or swallowed up in an earthquake. Thus, in V was found dead in the river after being involved in a scuffle with D on the bank, in the course of which V s wallet was stolen by D. At D s trial for murder, D argued that V had leapt into the river of his own accord and had not, as the prosecution had asserted, been pushed. The jury was directed that even if D s story were true, D would remain liable if V had jumped into the river to escape a well-grounded fear of violence: not that there had to be no other way of escape, but it had to be something a reasonable man might do. This principle may also be used to explain Huyw~rd,~~ where V (who had a heart complaint) died of fright after being threatened by D, and Pugett,26 where D used V as a human shield in the course of a gun battle with police in which V was killed. In both cases, it was reasonably foreseeable that V would die in that way as a result of D s acts. For an example of the foreseeability principle in relation to medical treatment, we can look at M~Znfyre,~~ where V died from choking on a restorative, and Forrest,28 where V died from infection contracted in the hospital where he was being treated. In both cases, death was a foreseeable consequence of D s actions. The final principle is that D must take his victim as he finds him. This comes from Bl~ue,~~ where, it will be remembered, the original wound inflicted by D was followed by V s refusal of a blood transfusion on religious grounds, so that V ultimately died. Upholding D s conviction for manslaughter, Lawton W said that it was no defence for D to argue that V s decision was an unreasonable one, since it has long been the policy of the law that they who use violence on other people must take their victims as they find them.30 He added that this, in his judgment, meant the whole man, not just the physical man 3 : hence, V s religious beliefs provided no excuse for D. This, as we have seen, is a case where the original wound was still operative, but Lawton LJ s principle is capable of wider application: say, for instance, V is injured by D and then, because of some nervous debility, kills himself.32 On this principle, D would be liable for the death. It is only where none of these principles apply that D can escape liability on the ground that the original injury, though the factual cause of death, was not the legal cause. Such cases are unlikely to arise, but one example is Here, D was convicted of murder after stabbing V. However, the conviction was quashed by the Court of Criminal Appeal on new evidence being given that, at the time of the death, the original wound had almost healed and that death was brought on by the continued administration of a drug to which V was known to be allergic, treatment See Perkins at (1946) 36 J Cr L & CR 393. (1842) Car & M 284. (1908) 21 Cox CC 692: cf Towers (1874) 12 Cox CC 530. (1983) 76 Cr App Rep 279. (1874) 2 Cox 379; see Hart and Honor& Causation in the Law (2nd ed) p 355. (1886) 20 South Australia LR 78; Hart and Honor& lor cit. [ All ER 446; above p 569. ibid at p 450. ibid. cf Pigney v Pointers Transport Services Ltd [ All ER 807. ( 1956) 40 Cr App Rep 152.

5 that was described as palpably wrong and not Here, the original wound was no longer operative (it had almost healed) and the intervening treatment could not be described as reasonably foreseeable (it was not normal ). Nor was this a case where one could apply the principle that D must take the victim as he finds him. So the chain of causation can sometimes be broken in cases of this sort. But this will only happen in the most exceptional circumstance^.^^ The Decision in Cheshire It will be seen in the light of the above principles that D s argument in Cheshire, to the effect that the careless medical treatment given to V broke the chain of causation between the shooting and V s subsequent death, was a pretty hopeless one. The conclusion reached by the Court of Appeal was that, while the trial judge had been wrong to mention recklessness (so directing the jury in terms of the culpability of the doctors rather than the consequences of their treatment), the summing up taken as a whole had done no injustice to the appellant. It was only in the most extraordinary cases that medical treatment of an injury inflicted by D would break the causal chain: namely, where it was so independent of D s acts, and in itself so potent in causing death, as to render the contribution made by D legally insignificant. Though the medical treatment in the present case was not as good as it might have been, it was far from being grossly abnormal and, accordingly, was not sufficient to break the causal link. The appeal was therefore dismissed. A number of points of principle were stressed by Beldam LJ in his discussion of the relevant issues: (1) Causation, whether in homicide or elsewhere, is a question of fact for the jury, which is to be decided in accordance with legal principles to be explained to the jury by the judge.j6 (2) In most cases, ajudge need do no more than tell the jury that D s act need not be the sole cause, or even the main cause, of V s death, provided that it contributed significantly to that (3) Sometimes, the issue is whether the actions of a third party can be regarded as a nevus actus interveniens, relieving D of criminal responsibility. The essence of a novus actus interveniens is not simply the intervening act of another: rather, the term implies something so independent of the act of the accused that it should be regarded in law as the cause of the victim s death, to the exclusion of the act of the accused. 3E (4) There is a clear distinction to be drawn between the blameworthiness of conduct and its causative effect. It is therefore unhelpful when causation is in issue to direct juries using figures of speech more appropriate for conveying degrees of fault or blame (as in a civil case where apportionment is at issue). In particular, the mere fact that medical treatment falls below the 34 ibid at p Lord Lane CJ in Mulcherek described Jordan as a very particular case depending on its own exact facts : [I98112 All ER 422, at p 428. Compare Purchase v See/ye (1918) 231 Mass 434 (employer held not liable for injuries sustained by plaintiff when doctors confused him with another patient and performed a totally unnecessary operation, without treating the original injury at all) All ER at p 674: see Malcherek All ER 422; Pageir (1983) 76 Cr App Rep 279. at 288 per Robert Goff W; above p ibid approving the words of Robert Goff W in Pageii, supra. 38 ibid. 58 1

6 The Modern Law Review [Vol. 55 standard expected of the competent medical practitioner does not of itself break the causal chain; such treatment is unfortunately only too frequent in human experience for it to be considered abnormal in the sense of extraordinary. 39 (5) In cases where the argument is that bad medical treatment broke the chain of causation, the jury should be told that, even though the negligent treatment was the immediate cause of V s death, they should not regard it as excluding D s responsibility unless it was so independent of D s acts, and itself so potent in causing death, that they regard the contribution made by D s acts as insignificant.40 Where, then, does Cheshire fit into the principles discussed above? At some points in his judgment, Beldam LJ seems to regard the case as one where the original wound was still the operative cause of the death. Thus, some stress was placed on an Australian case, Evans and Gurdiner (No 2).41 Here, V was stabbed in the stomach and given surgical treatment. Nearly a year later, he died from stricture of the small bowel, a not uncommon complication of the operation carried out to deal with stab wounds of the type suffered by V. As in Cheshire, D argued that the failure of the medical team to diagnose V s complaint broke the chain of causation. The court began by applying the test whether the original felonious act was still an operating and substantial cause of the death. They concluded that it was: the failure of the medical practitioners to diagnose correctly V s condition, however inept or unskilful, was not the cause of death. It was the blockage of the bowel which caused death, and the real question for the jury was whether that blockage was due to the stabbing.42 While one cannot cavil with the decision, the reasoning used by the court in Evans and Gurdiner (No 2) is questionable. A wound is not an operative cause of death merely because it causes some other condition which in turn brings about the death: if this were so, there would be no need for the second principle, that of foreseeability. Indeed, it would be possible to categorise a wound as still operative in such cases even if it had completely healed, which seems contrary to Jordan.43 It would seem far more sensible to treat Evans and Gurdiner (No 2) as an illustration of the foreseeability principle: given that stricture of the bowel was a not uncommon complication of the treatment necessitated by D s original wound, it was clearly foreseeable that V might die in this way. It is quite clear, from the references of Beldam LJ to the fact that bad medical treatment was unfortunately only too frequent in human experience to be considered abnormal, and from his reliance on Pugeft, that Cheshire involved an application of the foreseeability principle. A man who leaves his unconscious victim on the seashore where he is drowned by the incoming tide is responsible for the death, not because the blow was the operative cause of death (though the blow certainly caused the drowning) but because, in the circumstances, it was reasonably foreseeable that V might die in that way. D increased the risk of him dying in that manner. In the same way, if D wounds V who is operated on and later dies of a complication resulting from the operation, he is responsible for V s death if the complication All ER at p ibid at p I VR ihid at p (1956) 40 Cr App Rep 152; above pp (1983) 76 Cr App Rep

7 was reasonably foreseeable. Even if the medical treatment is faulty, it will provide no defence; since bad medical treatment is not an uncommon phenomenon, it will be within the bounds of reasonable foreseeability. The only case where it will excuse D is where it is not only bad, but so appallingly incompetent as to be outside the bounds of what one would normally expect. Doctors may occasionally fail to make the correct diagnosis, as in Evans and Gardiner (No 2) and in Cheshire itself. But one very rarely sees them giving drugs to a patient who is known to be allergic to them, as in Jordan. The competence of the treatment may be relevant in so far as mildly incompetent treatment is foreseeable, whereas grossly incompetent treatment is not. But, in the end, the question is one of foreseeability. Conclusion When, in the light of Cheshire, will medical treatment of any injury inflicted by D break the chain of causation? The answer must be, very rarely indeed. The principles seem to be as follows. (1) Medical treatment will not break the chain of causation simply because V would not have died but for the bad treatment. The injuries inflicted by D need not be the sole cause, or even the main cause, of the death, provided they made a significant contribution to it. (2) Where the original injuries can be categorised as being still operative at the time of the death, the causal chain will not be broken. It will not matter if the medical treatment was incompetent, or even grossly abnormal. Provided the wound is the direct cause of the death, D will remain responsible for it. (3) Where the original injuries are no longer operative, the test will be whether the bad medical treatment was reasonably foreseeable. Treatment falling within the normal band of incompetence will be regarded as foreseeable for the purposes of this rule. But where the treatment is so palpably bad as to be outside the bounds of reasonable contemplation, it may break the chain of causation. (4) Where the death occurs as a result of the treatment of the injury (or lack of it) by V himself, D may still be responsible on the ground that he must take his victim as he finds him. This may be so even though the original injuries are no longer operative and though the treatment is so grossly wrong as to be unforeseeable. The cases where D is not caught by any of these principles are bound to be few and far between; even if it is not the law that you take your victim as you find him, it will not be often that the original wound will have ceased to operate AND the subsequent treatment be palpably abnormal. The argument that it wouldn t have happened if the doctors hadn t made such a mess of it has never been one with much chance of success in a criminal case and, after Cheshire, the odds against its succeeding must be even greater. 583