Causation in negligence - Factual causation: 'but for' test

Learning Outcomes

This article explains the concept of factual causation as a key element in establishing negligence. It focuses specifically on the 'but for' test, its application, and its limitations. For the SQE1 assessment, you need to understand how to apply the 'but for' test to determine if the defendant's breach of duty factually caused the claimant's harm. Your understanding will allow you to analyse scenarios and identify the correct legal outcome regarding factual causation in multiple-choice questions.

You should also be able to distinguish straightforward single-cause cases from complex situations where multiple potential or cumulative causes operate, appreciate the evidential burden and standard of proof, and recognise when the courts have adopted modified approaches (material contribution to damage, material increase in risk) in response to scientific or evidential uncertainty. You must be comfortable analysing lost chance in clinical negligence, and consecutive or concurrent causes, and know the special position for mesothelioma claims.

SQE1 Syllabus

For SQE1, you are required to understand the principles of factual causation in negligence, particularly the application of the 'but for' test, with a focus on the following syllabus points:

• The definition and role of factual causation within the elements of negligence.
• The 'but for' test as the primary method for establishing factual causation.
• Applying the 'but for' test to given factual scenarios.
• Recognising situations where the 'but for' test may present difficulties, such as cases involving multiple potential causes.
• Understanding the standard of proof required (balance of probabilities).
• Appreciating the “all or nothing” nature of proof in civil cases and its implications for causation.
• Knowing when courts have used modified approaches: material contribution to damage (e.g., cumulative industrial diseases) and material increase in risk (the Fairchild exception for mesothelioma, reinforced by statute).
• Identifying lost chance claims in clinical negligence and why sub‑50% chances do not satisfy the civil standard of proof.
• Understanding consecutive and concurrent causes, including multiple sufficient causes and how liability is treated in cases like Baker v Willoughby and Jobling v Associated Dairies.
• Recognising divisible versus indivisible injuries and the practical impact of apportionment and joint and several liability.

Test Your Knowledge

Attempt these questions before reading this article. If you find some difficult or cannot remember the answers, remember to look more closely at that area during your revision.

1. What is the primary test used to determine factual causation in negligence?
   1. The reasonable foreseeability test
   2. The material contribution test
   3. The 'but for' test
   4. The proximity test
2. In the context of factual causation, what standard of proof must the claimant meet?
   1. Beyond reasonable doubt
   2. Clear and convincing evidence
   3. Balance of probabilities
   4. Prima facie evidence
3. Which case famously illustrates the application of the 'but for' test where a doctor's negligence did not factually cause the patient's death?
   1. Donoghue v Stevenson
   2. Caparo Industries plc v Dickman
   3. Barnett v Chelsea & Kensington Hospital Management Committee
   4. Bolam v Friern Hospital Management Committee

Introduction

Once you have established that a defendant owed the claimant a duty of care and breached that duty by falling below the required standard of care, the next essential element in a negligence claim is causation. The claimant must prove that the defendant's breach of duty actually caused the damage suffered. Causation is divided into two main parts: factual causation and legal causation. This article focuses on factual causation, specifically exploring the primary test used by the courts: the 'but for' test. Understanding factual causation is essential for determining liability in negligence.

Key Term: Causation
The link between breach of duty and harm. It comprises factual causation (did the breach actually cause the harm?) and legal causation (is the harm too remote or has the chain of causation been broken?).

Key Term: Factual Causation
The principle that the defendant's breach of duty was, as a matter of fact, a cause of the claimant's damage. It addresses the question: did the defendant's actions actually lead to the claimant's harm?

Key Term: 'But For' Test
The primary test for factual causation which asks: 'But for the defendant's breach of duty, would the claimant have suffered the harm?'

Key Term: Balance of Probabilities
The civil standard of proof. The claimant must show it is more likely than not (over 50% probability) that the defendant’s breach caused the harm.

Factual Causation: The 'But For' Test

The fundamental principle of factual causation is to establish a link between the defendant’s breach of duty and the claimant’s harm. The main test the courts apply is the 'but for' test.

If the answer to the 'but for' question is 'no' (i.e., the harm would not have occurred without the defendant's breach), then factual causation is established. If the answer is 'yes' (i.e., the harm would have occurred anyway, regardless of the defendant's breach), then factual causation is not established, and the negligence claim will fail at this stage.

The burden of proving factual causation rests on the claimant. The standard of proof is the balance of probabilities, meaning the claimant must show it is more likely than not that the defendant's breach caused the harm. This often results in an “all or nothing” outcome in civil litigation: if the claimant cannot reach the 51% threshold for the breach having caused the harm, causation is not proved and the claim fails.

The general formulation comes from authority such as Cork v Kirby MacLean Ltd, where it was stated that if the damage would not have happened but for the fault, the fault is a cause; if the damage would have happened in any event, fault or no fault, that fault is not a cause.

Applying the 'But For' Test

The application of the 'but for' test is often straightforward in cases with a single, clear cause of harm. The classic illustration is Barnett v Chelsea & Kensington Hospital Management Committee [1969] 1 QB 428.

Worked Example 1.1

A patient attended a hospital casualty department complaining of severe stomach pains and vomiting after drinking tea. The doctor on duty negligently failed to examine the patient and sent him home, telling him to contact his GP. The patient later died from arsenic poisoning contained in the tea. Medical evidence showed that even if the doctor had examined and treated the patient correctly, the patient would have died anyway due to the severity of the poisoning. Did the doctor's negligence factually cause the patient's death?

Answer:
No. Applying the 'but for' test: but for the doctor's negligent failure to examine the patient, would the patient still have died? Yes, the evidence indicated death was inevitable due to the poisoning. Therefore, the doctor's breach was not the factual cause of death.

To apply the test correctly, identify the breach precisely and ask the counterfactual question. Avoid conflating causation with breach or with foreseeability and remoteness. In Barnett, the doctor unquestionably breached the duty of care; but factual causation failed because the death would have occurred regardless.

Worked Example 1.2

A factory worker develops lung disease after many years of exposure to dust. Some dust exposure ('innocent dust') is an unavoidable part of the job, even with precautions. However, due to the employer's negligence in maintaining extraction equipment, the worker was also exposed to excessive amounts of 'guilty dust'. It is impossible to say definitively whether the disease was caused by the 'innocent' dust, the 'guilty' dust, or the cumulative effect of both. How might the 'but for' test apply here?

Answer:
Applying the 'but for' test strictly is problematic. Can the worker prove that 'but for' the negligent exposure to 'guilty dust', they would not have developed the disease? It might be difficult if the 'innocent' dust exposure alone could have caused the illness. This highlights a limitation of the 'but for' test in multiple cause scenarios. (Note: In such cases, courts may consider alternative tests like 'material contribution to harm', which are beyond the scope of this specific article focusing on the 'but for' test).

Worked Example 1.3

A premature baby develops retrolental fibroplasia (a condition causing blindness). There are five possible causes for the condition; only one involves a negligent excess of oxygen administered by hospital staff. The claimant cannot show which of the five causes actually operated. Does the 'but for' test succeed?

Answer:
No. This is analogous to Wilsher v Essex Area Health Authority. The claimant bears the burden to show on the balance of probabilities that the negligent cause, rather than one of the non‑negligent alternatives, caused the harm. Where several independent possible causes exist and the negligent factor is only one of them, ‘but for’ causation is not proved.

Limitations of the 'But For' Test

While the 'but for' test is the starting point, it can be difficult to apply, particularly in situations involving multiple potential causes of harm. The law has developed limited exceptions or modified approaches in response to evidential and scientific problems.

Key Term: Material Contribution to Damage
Where harm results from cumulative causes acting together, causation can be established by showing the defendant’s breach made a more than minimal contribution to the damage, even if ‘but for’ cannot be proved.

Key Term: Material Increase in Risk
In narrow categories involving scientific uncertainty (most notably mesothelioma from asbestos), causation may be established by showing the defendant’s breach materially increased the risk of the claimant contracting the disease.

Key Term: Multiple Sufficient Causes
Situations where two or more causes would each be sufficient to bring about the harm (e.g., simultaneous negligent acts). The orthodox ‘but for’ analysis can yield contradictory outcomes unless responsibility is shared.

Key Term: Lost Chance
A claim that negligence reduced the claimant’s prospects of a better outcome. In clinical negligence, a chance below 50% generally does not satisfy causation on the civil standard for personal injury.

Key Term: Divisible Injury
An injury that can be scientifically or evidentially apportioned between causes (e.g., cumulative exposure over time). Damages may be allocated proportionately.

Key Term: Indivisible Injury
An injury that cannot be meaningfully divided between causes (e.g., a single broken bone from concurrent negligence). Each tortfeasor is liable for the whole damage to the claimant, subject to contribution between defendants.

In industrial disease cases, harm may be cumulative. In Bonnington Castings Ltd v Wardlaw [1956] AC 613, the House of Lords accepted causation where the employer’s breach materially contributed to pneumoconiosis, even though ‘but for’ could not be strictly proved against the guilty dust alone. This approach reflects that cumulative, combined exposures can each play a part in producing the disease.

By contrast, where there are multiple alternative possible causes (as in Wilsher), the claimant must still satisfy ‘but for’ causation on the balance of probabilities. Simply adding one possible cause to an existing set of risks is not enough.

In exceptional asbestos cases, the Fairchild v Glenhaven Funeral Services line of authority allows claimants to rely on material increase in risk due to scientific uncertainty about the causal mechanism of mesothelioma. Parliament reinforced this specific position in Compensation Act 2006, section 3, making responsible persons jointly and severally liable in mesothelioma claims. Sienkiewicz v Greif extended Fairchild to a single negligent defendant where environmental exposure also existed; the question remains whether the negligent exposure materially increased the risk.

Lost chance claims show the “all or nothing” nature of the civil standard. In Hotson v East Berkshire AHA, where evidence showed a 75% chance of the same outcome even with competent treatment, the claimant could not show on balance that negligence caused the disability; a 25% lost chance did not meet the standard.

Multiple sufficient causes and consecutive causes can also challenge simple application. In Baker v Willoughby, the original tortfeasor remained liable for continuing loss despite a later intentional shooting and amputation; in Jobling v Associated Dairies, a subsequent non-tortious disease reduced the damages period, as the later condition would have occurred in any event. These decisions reflect different policy considerations about supervening causes and responsibility.

Worked Example 1.4

A worker develops pneumoconiosis from dust exposure. Some dust is unavoidable; some arises from negligently maintained equipment. Medical science indicates the disease is cumulative. The worker cannot show which dust particles caused the disease. Can causation be proved?

Answer:
Yes, on a material contribution approach (Bonnington Castings v Wardlaw). The claimant shows the negligent exposure made a more than minimal contribution to the cumulative disease, even if a strict ‘but for’ analysis is impossible.

Worked Example 1.5

A child falls from a tree and sustains a hip injury. Hospital staff then negligently fail to diagnose vascular necrosis. Medical evidence shows a 75% chance that even prompt, competent treatment would still have led to the same disability. Is factual causation proved?

Answer:
No. As in Hotson v East Berkshire AHA, the claimant must prove on the balance of probabilities that the negligence caused the disability. A 25% lost chance does not satisfy causation for personal injury in clinical negligence.

Worked Example 1.6

A driver’s negligence causes a severe leg injury, with chronic pain and reduced earning capacity. Later, the claimant is shot in a robbery, resulting in amputation of the same leg. Does the original driver remain liable for the original loss?

Answer:
Yes, as per Baker v Willoughby. The original tortfeasor’s liability does not cease due to a subsequent intentional wrong; damages are assessed as if the supervening event had not reduced the original loss. Contrast Jobling v Associated Dairies, where a non‑tortious disease reduces recoverable loss because it would have occurred in any event.

Worked Example 1.7

A Rolls‑Royce suffers damage in collision A, necessitating a respray of the lower half of the car. Before it is repaired, collision B occurs and the same area is damaged, also requiring a respray. Can the claimant recover the respray cost from the second driver?

Answer:
No. In Performance Cars v Abraham, the second defendant did not cause additional damage; the respray was already required. Applying ‘but for’, the cost would have been incurred anyway. Recovery lies against the first tortfeasor.

Worked Example 1.8

A claimant develops mesothelioma. They worked for multiple employers, all of whom negligently exposed them to asbestos, and scientific evidence cannot identify which exposure initiated the disease. How is causation addressed?

Answer:
The Fairchild exception applies: each negligent employer materially increased the risk of mesothelioma and is liable. Section 3 Compensation Act 2006 renders responsible persons jointly and severally liable. In Sienkiewicz v Greif, this approach was applied even where there was only one negligent defendant alongside environmental exposure.

Worked Example 1.9

In a gas‑filled room, two negligent actors each light a match at the same moment. An explosion ensues. Asking “but for X’s negligence, would the harm have occurred?” yields “yes,” and the same for Y. Does ‘but for’ analysis alone resolve liability?

Answer:
No. This illustrates multiple sufficient causes: orthodox ‘but for’ produces contradictory results. Courts may treat both actors as factual causes of the same indivisible injury, with each liable to the claimant, subject to apportionment of responsibility between defendants under contribution principles.

Key Term: Joint and Several Liability
Where multiple defendants are liable for the same indivisible damage, the claimant can recover full damages from any one of them. Defendants then seek contribution between themselves.

Key Term: Civil Liability (Contribution) Act 1978
Statutory mechanism allowing a defendant who pays more than their share for the same damage to recover a just and equitable contribution from other responsible parties.

Exam Warning

For SQE1, ensure you can apply the basic 'but for' test correctly. Be aware that questions might present scenarios where causation seems likely but fails the 'but for' test (like Barnett), or where its application is complicated by multiple factors. Focus on whether the claimant can prove, on the balance of probabilities, that the harm would not have occurred but for the specific breach identified. Distinguish cumulative causes (material contribution may be relevant) from alternative causes (Wilsher scenarios where ‘but for’ must be proved). Remember lost chance below 50% does not satisfy causation for personal injury. For mesothelioma, note the statutory joint and several liability under Compensation Act 2006.

Key Point Checklist

This article has covered the following key knowledge points:

• Factual causation is the first stage in establishing the causal link between the defendant's breach and the claimant's harm in negligence.
• The primary test for factual causation is the 'but for' test.
• The 'but for' test asks: 'But for the defendant's breach of duty, would the claimant have suffered the harm?'
• If the harm would have occurred regardless of the breach, factual causation is not established.
• The claimant must prove factual causation on the balance of probabilities; causation in civil cases is an “all or nothing” threshold.
• The Barnett v Chelsea & Kensington Hospital case is a key example of applying the 'but for' test.
• The 'but for' test can be difficult to apply in situations involving multiple potential causes (e.g., Wilsher v Essex AHA).
• In cumulative disease cases, courts may accept material contribution to damage (e.g., Bonnington Castings v Wardlaw).
• Lost chance in clinical negligence (e.g., Hotson v East Berkshire) under 50% does not satisfy the civil standard of proof for causation.
• Multiple sufficient causes can defeat a simple ‘but for’ analysis; courts can treat both as factual causes of an indivisible injury, with contribution between tortfeasors.
• Consecutive causes are treated differently: original tortfeasors may remain liable despite later intentional acts (Baker v Willoughby), whereas non‑tortious supervening disease can limit damages (Jobling v Associated Dairies).
• Divisible injuries may be apportioned (e.g., Holtby v Brigham & Cowan), whereas indivisible injuries lead to joint and several liability, with apportionment between defendants under the Civil Liability (Contribution) Act 1978.
• Mesothelioma claims engage the Fairchild exception and Compensation Act 2006, s.3, making responsible persons jointly and severally liable; Sienkiewicz confirms applicability with a single negligent defendant.

Key Terms and Concepts

• Causation
• Factual Causation
• 'But For' Test
• Balance of Probabilities
• Material Contribution to Damage
• Material Increase in Risk
• Multiple Sufficient Causes
• Lost Chance
• Divisible Injury
• Indivisible Injury
• Joint and Several Liability
• Civil Liability (Contribution) Act 1978