Chapter 3: Causation

Question 1

Introduction: Nexus between breach and damage

Answer 1

Negligence is generally broken down into duty, breach, causation and remoteness. You have already considered duty and breach. Once breach is established, the claimant must then prove that this breach caused the damage. This link or ‘nexus’ between breach and damage is essential. Without it the claimant’s claim will fail, ie there will be no liability.

To determine if there is the necessary link two points must be considered:
(a) Factual causation; and
(b) Legal causation.

Question 2

2. Factual causation: The ‘but for’ test

Answer 2

The issue of factual causation is one of the most complex and challenging topics arising in the law of tort. The question that needs answering is: as a matter of fact, did the defendant’s breach cause the claimant’s loss? The ‘but for’ test is the starting point for answering this question

The ‘but for’ test: On the balance of probabilities, but for the defendant’s breach of duty,
would the claimant have suffered their loss at that time and in that way? If no, factual causation is satisfied. The claimant would not have suffered their loss were it not for the defendant’s breach. If yes, factual causation is not satisfied, and the defendant is not liable. The claimant would
have suffered their loss even without the defendant’s breach.

Question 3

Key case: Barnett v Chelsea and Kensington Hospital [1969] 1 QB 428

Answer 3

Facts: The hospital breached the duty of care owed to a patient in that a doctor failed to carry
out a proper examination. The patient died of arsenic poisoning after drinking poisoned tea. However, the evidence showed that he would have died even if the doctor had examined him. There was little or no chance that the antidote would have been administered to him on time even if he had been examined.

Held: The claim failed on causation: but for the defendant’s breach (failing to examine), would the claimant have died at that time and in that way? Yes. On the balance of probabilities, the claimant still would have died even if he had been examined by the doctor. Drinking the poisoned tea would have brought about his death whether or not he was treated by the doctor.

Question 4

But for Test: More than 50% chance

Answer 4

Remember: the ‘but for’ test must be proved on the balance of probabilities ie that there is a more than 50% chance that the defendant’s breach caused the claimant’s loss. So, a small chance that the accident might have happened anyway does not mean that the claim will fail.

Example 1: A window cleaner falls from a building because they had not been properly trained in cleaning this particular type of ‘sash’ window. ‘But for’ this lack of training, they would not have fallen and suffered injury.

Example 2
A driver is injured in a car accident caused by defective brakes in their own vehicle and the negligent driving of a lorry driver who collided with them. It is not possible here to say definitely that ‘but for’ the defective brakes the accident would not have occurred because we do not know,
without further information, whether or not the negligent driving of the lorry driver would have been enough on its own to cause the collision.

Question 5

Key case: Wilsher v Essex AHA [1988] AC 1074

Answer 5

Facts: The claimant was born prematurely and suffered a condition that caused him to go blind. There was evidence that his blindness could have been caused by any of five equally probable different factors, only one of which was tortious (ie an excess of oxygen given incorrectly). The others were all the natural result of his premature birth. The evidence suggested that these factors
did not operate together to cause the blindness: one factor was solely responsible for the loss

Question 6

Key case: Wilsher v Essex AHA [1988] AC 1074 Judgement

Answer 6

Held: The claimant had to prove that but for the defendant’s breach he would not have suffered the blindness. As the standard of proof in civil actions is the balance of probabilities, the claimant had to establish that it was more likely than not (ie more than a 50% chance) that the blindness was caused by the breach as opposed to any of the other four possible factors. The claimant was
unable to do this and so did not succeed. He could only prove that there was a 20% chance.

Question 7

2.1 Factual causation in clinical negligence where the breach is a failure to advise on risks

Key case: Chester v Afshar [2004] UKHL 41

Answer 7

The surgeon failed to disclose to the claimant the very small risk of paralysis resulting from
surgery. After the operation, the claimant suffered paralysis in one leg.

The defendant argued that causation could only be proved if the claimant could show that she would not have had the operation had she been warned of the risk. The court took a more relaxed approach: where the breach is a failure to advise on risk, the ‘but for’ test is satisfied if the claimant can prove on the balance of probabilities, that if they had been warned of the risk, they would not have had the operation or deferred it to a later date. The claimant proved that ‘but for’ the surgeon’s failure to warn, the injury would not have occurred when it did and the chance of it occurring on a subsequent occasion (if the claimant had the operation later) was very small.

Question 8

2.2 Summary

Answer 8

• Factual causation deals with establishing the link between the defendant’s breach and the
  claimant’s damage.
• The starting point for establishing factual causation is always the ‘but for’ test; but for the defendant’s breach, would the claimant have suffered their loss? If no, factual causation is
  satisfied.
• The claimant must prove the ‘but for’ test on the balance of probabilities ie there is a greater
  than 50% chance that the breach caused the damage.
• In Wilsher, the claimant was unable to prove factual causation using the ‘but for’ test. There were five equally probable causes of the claimant’s loss acting independently; only one of
  which was the breach. The claimant could therefore only prove that there was a 20% chance that the breach caused his loss.
• In clinical negligence, where the breach is a failure to advise on risk, the ‘but for’ test can be
  satisfied if the claimant can prove that they would not have had the treatment or would have deferred the treatment had they been told of the risk.

Question 9

3 Factual causation where the ‘but for’ test cannot be
satisfied: 3.1 Material contribution test (cumulative causes)

Answer 9

The starting point for establishing factual causation is the ‘but for’ test. If this is satisfied, then factual causation is established. However, even if the test cannot be satisfied, there are some exceptional circumstances in which the court will nonetheless find factual causation. The material
contribution test is applied to factual causation when there is more than one cause of the
claimant’s loss, and the causes were acting together (cumulatively) to cause the loss

Question 10

Key case: Bonnington Castings v Wardlaw [1956] AC 613

Answer 10

Facts: The claimant sued his employers in negligence for a respiratory disease (pneumoconiosis). Whilst it was clear that the cause of the disease was exposure to dust at work, only part of this
exposure was due to the defendant’s breach of duty. Some of the exposure to the dust was deemed to be a natural consequence of the work being carried out (and therefore non-tortious).
Here the tortious dust and non-tortious dust operated together to produce the disease as medical evidence established that the disease was cumulative, caused by a build-up of dust in the lungs.

The claimant could not satisfy the ‘but for’ test as he could not prove that but for the defendant’s breach (exposure to the tortious dust), he would not have suffered his disease at that time and in that way. Medical evidence could not say whether exposure to the tortious dust alone was enough to cause the disease.

Question 11

Key case: Bonnington Castings v Wardlaw [1956] AC 613 Judgement

Note: The material contribution test has been applied to clinical negligence claims

Answer 11

Held: The House of Lords deviated from the ‘but for’ test and introduced the material contribution test. If the defendant’s breach could be proved to have materially contributed to the claimant developing the disease, then the defendant would be liable for all the loss. By material contribution, the court meant a ‘more than negligible’ contribution to the loss. The claimant was therefore successful as he could prove that the tortious dust had made a more than negligible
contribution to causing his disease.

Question 12

Key case: Bailey v Ministry of Defence [2008] EWCA Civ 883

Answer 12

Facts: Due to the claimant’s weakness, she choked on her own vomit, causing brain damage. The weakness was caused by:
(a) the natural progression of the claimant’s condition (non-tortious); and

(b) the negligent lack of care by the defendant (tortious).
Medical experts could not say that ‘but for’ the negligent treatment, she would not have suffered the weakness and subsequent brain damage.
Held: Claim succeeded. The claimant could prove that the negligent treatment (the breach) made a material (more than negligible) contribution to her brain damage

Question 13

Lord Justice Waller (applies in sequential cumulative causes)

Answer 13

Subsequent case law has confirmed that the material contribution test applies to sequential cumulative causes (ie where one cause comes after the other: for example, the claimant has a fall and suffers a head injury, then later receives negligent medical treatment and the medical evidence is that the fall and the clinical negligence caused brain damage) as well as to simultaneous cumulative causes (ie causes operating at the same time: in Bailey the negligent treatment and underlying medical problems caused the weakness simultaneously).

Question 14

3.2 Material increase in risk Key case: McGhee v National Coal Board [1973] 1 WLR 1 (HL)

Answer 14

Facts: The claimant contracted dermatitis as a result of exposure to brick dust. The claimant worked with brick dust during his working hours and there was no breach in relation to this exposure (ie it was non-tortious). However, his employers were in breach of duty for failing to provide washing facilities for him to wash the brick dust off his skin at the end of the working day.
As a result, the brick dust was on his skin for an extended amount of time including his cycle journey home.

Question 15

Difference from Bonnington

Answer 15

The facts were therefore similar to Bonnington but with an important difference. In McGhee, medical evidence could not establish that dermatitis was a cumulative condition (ie it could not prove that the tortious and non-tortious dust operated together to cause the dermatitis).

The dermatitis could have been caused by a single exposure to the dust, so may have been caused by the non-tortious dust with the tortious dust making no contribution at all. Thus, the material contribution test could not be satisfied. The claimant could not show on the balance of probabilities that the tortious dust had made a more than negligible contribution.

Question 16

McGhee v National Coal Board [1973] 1 WLR 1 (HL) Judgement

Answer 16

Held: Medical experts could demonstrate that the tortious exposure to the dust materially
increased the risk of contracting dermatitis: the longer the dust was on the claimant’s skin, the
greater the risk of contracting dermatitis. The House of Lords therefore imposed liability on the defendant on the basis that their breach had materially increased the risk of the claimant’s injury.

This is arguably an easier test to satisfy than the material contribution as the claimant does not even have to prove that the breach made any actual contribution to their loss, just that it increased the chances of the claimant suffering that loss. This is perhaps why this test is only applied in very limited situations.

Question 17

Key aspect of material increase in risk

Answer 17

Currently, the material increase in risk test seems to be confined to industrial disease cases only
where there is scientific uncertainty over cause. For example, McGhee. The material increase in risk test is most widely applied in mesothelioma cases. Mesothelioma is a lung cancer caused by exposure to asbestos dust.

Question 18

Key case: Fairchild v Glenhaven Funeral Services Ltd [2003] 1 AC 32

Answer 18

Facts: The claimant worked for a number of employers at different times in the 1960s, all of whom exposed him to asbestos. Twenty-five years later, he developed mesothelioma. The scientific
evidence was unable to determine with certainty whether the disease was caused by a cumulative exposure to asbestos over time or by exposure to a single asbestos fibre (just like in McGhee with
the brick dust). As a result, it was impossible for the medical experts to say which exposure with which employer was the cause of his illness.

The material increase in
risk test is most widely applied in mesothelioma cases. Mesothelioma is a lung cancer caused by exposure to asbestos dust.

Question 19

Key case: Fairchild v Glenhaven Funeral Services Ltd [2003] 1 AC 32 Judgement

Answer 19

Held: The House of Lords applied the material increase in risk test and the claim succeeded. The
claimant could show that by exposing him to asbestos, the defendant had materially increased his risk of contracting mesothelioma. Although this might seem unfair to the employer, as they may have made no contribution to the mesothelioma at all, this had to be weighed against the
argument that people who suffer harm due to their employers’ breach deserve to be
compensated. In the case of mesothelioma, science is unable to provide the claimant with the
means to ever establish factual causation using the ‘but for’ test (unless the claimant worked for
only one employer who exposed them to asbestos, which is rare).

Question 20

3.2.1 Limits to the material increase in risk test (Single Agency Cases)

Answer 20

The material increase in risk test is applicable only to industrial disease, single agency cases.
Single agency means that there is only one causal agent (eg the dust In McGhee and Fairchild).
The medical evidence just could not prove that it was the tortious dust that caused or even contributed to injuries.

Question 21

Differences from Wilsher

Answer 21

In Wilsher , where it was also impossible to say whether the defendant’s harm contributed in any way to the claimant’s loss, the material increase in risk test could not be applied. This is because, unlike McGhee, the claimant’s loss in Wilsher could have been caused by any one of five equally probably different causes (only one of which was tortious). This is therefore a multiple agency case. The tortious cause may not have increased the risk of blindness at all. The courts in Wilsher relied on this to distinguish the case from McGhee.

Question 22

3.3 Loss of chance? Key case: Hotson v East Berkshire Health Authority [1987] AC 750

Answer 22

A child fell from a tree and broke his leg. The hospital was negligent in its treatment and the child was left paralysed. However, medical evidence indicated that there was a 75% risk that the broken leg would have left the claimant paralysed even without the negligent treatment.

The ‘but for’ test therefore failed.
Instead, the claimant argued ‘loss of chance’ ie that he had lost a 25% chance of recovery due to the defendant’s negligence and should be awarded 25% of his losses. The House of Lords rejected
this argument. The reality of the situation was that the child was most likely paralysed by the
original fall.

Question 23

Allied Maples Group v Simmons & Simmons [1995] 1 WLR 1602 (CA),

Answer 23

Although the ‘loss of chance’ argument does not seem to apply to medical negligence or personal injury cases, the courts have allowed the test in cases involving pure economic loss. In Allied Maples Group v Simmons & Simmons [1995] 1 WLR 1602 (CA), the claimant lost the chance to negotiate a clause in a contract as a result of the solicitor’s failure to advise. Causation was successful as the claimant proved that there was a real and substantial chance that the seller would have agreed to the clause.

Question 24

3.4 Summary

Answer 24

• The starting point for establishing factual causation is always the ‘but for’ test.
• Sometimes it is difficult, if not impossible, to establish what would have happened without the breach (where there is more than one potential cause).
• When this happens, if the multiple causes operated together to cause the claimant’s loss, the courts might depart from the ‘but for’ test and apply the material contribution test.
• In industrial disease single agency cases (namely mesothelioma and lung cancer caused by asbestos), the courts might depart from the ‘but for’ test and apply the material increase in risk test.
• Loss of chance can be argued where the loss is pure economic loss.

Question 25

4 Apportionment and multiple sufficient causes

4.1 Apportionment

Answer 25

Apportionment is a calculation to apply once factual causation has been established. Where
there are multiple tortious factors which are known to have caused part of the loss, the courts apportion liability between the defendants in a way that produces a practical result, providing compensation to the claimant while recognising the respective fault of the defendants.

Question 26

Key case: Fitzgerald v Lane & Patel [1987] QB 781

Answer 26

Facts: The claimant crossed the road at a pelican crossing when the lights were red for pedestrians. The first defendant driver collided with him and the claimant was thrown from the
bonnet of that car into the road, where he was run over by a car driven by the second defendant.
Both defendants had been negligent, and the claimant was equally at fault. It was impossible to say which of the two collisions had actually caused the injuries, or to what extent each had
contributed.

Held: each defendant was responsible for 25% of the claimant’s losses (50% of the claimant’s
losses were not recoverable, representing the claimant’s own negligence).

Question 27

Section 3 of the Compensation Act 2006,

Answer 27

However, note that in mesothelioma cases, under s 3 of the Compensation Act 2006, defendants are jointly and severally liable. This means that any or all of the negligent employers who exposed the claimant to asbestos will be liable to the claimant for the whole sum of damages, but
they can recover contributions from each other if necessary, to make the distribution of the loss
fair.

Question 28

4.2 Multiple sufficient causes: Is the original tortfeasor responsible?

Answer 28

In cases such as Bonnington, the claimant suffered one loss. What happens where the claimant
suffers damage as a result of the defendant’s negligence and then, some time later, a second
event occurs which causes exactly the same damage, or worsens the damage already caused? In these situations, the events are not linked and there are two or more distinct losses that can be attributable to distinct causes. There is more than one defendant each of whom pass the ‘but for’
test, but one action comes after the other (or if the later action is non-tortious, it has caused the
subsequent harm). Is the original tortfeasor responsible for the loss caused by the other?

Question 29

Key case: Performance Cars v Abraham [1962] 1 QB 33

Answer 29

Facts: A third party negligently collided with the claimant’s Rolls Royce. The Rolls required a
respray to remedy it. Two weeks later, a second collision between the same Rolls Royce and a
different car, driven by the defendant, caused similar damage which also required a respray to repair it.

Held: As the requirement for a respray already existed before the second collision, there was
effectively no damage arising from the second collision.

Question 30

4.2.1 Multiple sufficient causes: Contrasting case law

Key case: Baker v Willoughby [1970] AC 467

Answer 30

Facts: Due to the defendant’s negligence, the claimant suffered a leg injury (pain and stiffness). Subsequently, the claimant was shot in a robbery and his injured leg had to be amputated. The robbers could not be found.

Held: First defendant should continue to be liable for the original injuries to the leg, beyond the time of robbery (even though he no longer had pain and stiffness). The intervening tortfeasor
(robber) should compensate for any additional losses caused.

Question 31

Key case: Jobling v Associated Dairies [1982] AC 794

Answer 31

Facts: Due to the defendant’s negligence, the claimant injured their back and as a result suffered reduced earnings. Some time later, the claimant suffered a further back injury (non-tortious) arising from an illness unconnected to the accident. He was unable to work at all.

Held: the defendant’s liability ceased at the point that the further back injury developed. He did not have to compensate the claimant for the ‘vicissitudes of life’.

Question 32

4.3 Summary

Answer 32

• Where multiple tortious factors have caused the loss, the courts apportion liability between the
  defendants to produce a practical result, providing compensation to the claimant while recognising the respective fault of the defendants.
• Where there is more than one defendant, each of whom passes the ‘but for’ test for distinct separate losses, but one of their actions comes after the other, consider multiple sufficient causes. The later action may be non-tortious eg a natural event, but is the factual cause of the later loss.
• If the second defendant has not caused any additional damage to the claimant, they will not be liable.
• If the second event is tortious, the first defendant is liable for the original damage past the
  point of the second event. The second defendant is liable for any additional damage.
• If the second event is naturally occurring, the defendant is liable for damage only up to the
  natural event.

Question 33

5. Legal causation

Answer 33

Once factual causation has been established, it is then necessary to see whether, as a matter of
law, the defendant may be held liable. The defendant is not liable for absolutely everything that follows from their breach. A line has to be drawn and certain subsequent events that occur after the breach may break the chain of causation. This is the principle of novus actus interveniens ie an intervening act. Whether a subsequent event breaks the chain or not will depend on the nature
of what occurs.

5.1 Types of novus actusinterveniens. There are three types of intervening or novus actus events to consider:
(a) Acts of God or natural events
(b) Acts of third parties
(c) Acts of the claimant

Question 34

5.1.1 Acts of God or natural events

Answer 34

An act of God breaks the chain of causation if it is some exceptional natural event. For example, being struck by lightning, drowning in a flood, or the onset of certain disease.

Question 35

Key case: Carslogie Steamship Co Ltd v Royal Norwegian Government [1952]
AC 292

Answer 35

Facts: The claimant’s vessel was damaged in a collision with the defendant’s ship, for which the
defendant admitted liability. The repairs were not immediately necessary, and the vessel was
taken to the US for repair. On the way, the vessel suffered heavy storm damage. The initial repairs took 10 days to fix, the storm damage 51 days.

Held: The defendant was liable for damages arising from the first collision but not the storm
damage; the storm was held to be a novus actus interveniens. Natural events will not break the
chain of causation if they could have been foreseen and the defendant should have taken them into account as events that were likely to happen.

Question 36

5.1.2 Acts of third parties

Answer 36

Where the subsequent event is the act of a third party, the courts have viewed it as breaking the chain of causation if it was highly unforeseeable (something that was very unlikely to happen as
a result of the defendant’s negligence).

Question 37

Key case: Knightley v Johns [1982] 1 WLR 349

Answer 37

Facts: The first defendant caused a road traffic accident. Subsequently, a police inspector
negligently handled traffic control following the accident. This negligence led to the claimant, a police officer being injured (he had been ordered to travel down a tunnel against the flow of oncoming traffic).

Held: The first defendant successfully argued that the negligent handling by the police inspector broke the chain of causation between his negligence and the officer’s injury. The police inspector’s
actions were highly unforeseeable.

Question 38

Acts of third parties: Medical treatment

Answer 38

When the act of third party is medical treatment, the courts are reluctant to hold that this breaks the chain of causation. When a defendant causes injury, they take the risk that the claimant may not respond well to medical treatment or that the treatment may not be perfect. The medical treatment will not break the chain unless it is so gross and egregious as to be unforeseeable.

Question 39

Key case: Wright v Cambridge Medical Group [2011] EWCA Civ 669

Answer 39

Facts: The claimant was a child whom the defendant GP practice had negligently failed to refer to hospital promptly for diagnosis and treatment and who subsequently suffered permanent hip injury. When the patient was eventually referred to the hospital, the treatment she received was negligent.

Question 40

Key case: Wright v Cambridge Medical Group [2011] EWCA Civ 669

Answer 40

Held: The judges agreed that the hospital had been negligent in their treatment of the claimant
but that this negligence was not ‘such an egregious event, in terms of the degree or unusualness of the negligence […] to destroy the causative link between the defendants’ negligence and the
claimant’s injury’ (per Lord Neuberger).

Elias LJ in the same case went on to state that it could not be said that ‘the negligence in this case [of the hospital/third party], although serious, deserved to be characterised as gross or egregious so as to break the chain of causation […]’. The
defendant GP was liable for the full extent of the claimant’s loss, regardless of the negligence of
the hospital.

Question 41

5.1.3 Acts of the claimant

Answer 41

The legal test for an act of the claimant breaking the chain of causation is that the act must be
highly unreasonable. It is rare for the claimant’s unreasonable behaviour to break the chain of
causation as this would normally be dealt with under the defence of contributory negligence. This produces a fairer result, whereby the defendant is still liable for the claimant’s loss, but the
claimant’s damages are reduced to reflect the extent to which they contributed to their loss.

Question 42

Key case: McKew v Holland & Hanmen & Cubitts (Scotland) Ltd [1969] 3 All ER 1621

Answer 42

Facts: The claimant suffered a leg injury due to the defendant’s negligence, which resulted in
impaired mobility. Later, he descended a steep staircase without a handrail. His leg buckled, he fell and broke his ankle.

Held: The claimant acted very unreasonably and broke the chain of causation between the
breach and broken ankle.

Question 43

Key case: Wieland v Cyril Lord Carpets [1969] 3 All ER 1006

Answer 43

Facts: Due to the defendant’s negligence, the claimant wore a neck brace which restricted her
ability to use her glasses properly. As a result, she fell down some stairs and injured her ankle.

Held: The claimant had acted carefully (she had help from her son when descending the stairs). Her actions did not break the chain and the defendant was liable for her ankle injury.

Question 44

5.2 The effect of a novus actus interveniens (breaking the chain of causation)

Answer 44

The effect of a novus actus is, as we have mentioned, that it breaks the chain of causation. The defendant will still be responsible for any loss before the novus actus event but will not be responsible for any loss after it

Question 45

5.3 Summary

Answer 45

• The defendant is not always liable for absolutely everything that follows from their breach.
  Certain events that occur after the breach may break the chain of causation. This is the
  principle of novus actus interveniens.
• There are three types of novus actus interveniens: acts of God, acts of third parties and acts
  of claimant.
• Acts of God break the chain of causation if they are exceptional natural events.
• Acts of third parties break the chain of causation if they are highly unforeseeable.
• If the act of third party is medical treatment, this will only break the chain of causation if it is so gross and egregious as to be unforeseeable.
• Acts of claimant break the chain of causation if they are highly unreasonable. It is rare for the claimant’s unreasonable behaviour to break the chain of causation as this would normally be dealt with under the defence of contributory negligence.