Proving Causation in Malpractice: But-For and Substantial Factor Tests

Causation is one of the most contested battlegrounds in malpractice litigation, separating cases with clear negligence from those that can actually recover damages. Two principal legal tests — the but-for test and the substantial factor test — govern how plaintiffs must connect a defendant's conduct to the harm suffered. This page examines how each test operates, when courts apply one over the other, the doctrinal tensions between them, and the evidentiary requirements plaintiffs face in meeting each standard.

• Definition and Scope
• Core Mechanics or Structure
• Causal Relationships or Drivers
• Classification Boundaries
• Tradeoffs and Tensions
• Common Misconceptions
• Checklist or Steps (Non-Advisory)
• Reference Table or Matrix

Definition and Scope

Causation in malpractice law operates on two distinct layers: cause-in-fact (also called actual cause) and proximate cause (also called legal cause). The but-for test and the substantial factor test both address cause-in-fact — the threshold question of whether the defendant's conduct factually produced the plaintiff's injury.

The but-for test asks whether, absent the defendant's negligent act or omission, the harm would not have occurred. It is the default standard across all U.S. jurisdictions for single-cause scenarios. The substantial factor test applies in multi-cause scenarios where two or more independent causes, each sufficient alone to produce the harm, converge — making strict but-for analysis logically unworkable.

These standards are not interchangeable. Jurisdictions differ on which test governs in complex causation settings, and misapplying the wrong framework can defeat an otherwise valid claim. The elements of a malpractice claim framework requires plaintiffs to satisfy causation as a standalone element; a breach of the standard of care in malpractice law without proven causation does not support liability.

The Restatement (Third) of Torts: Liability for Physical and Emotional Harm (American Law Institute, 2010) addressed the substantial factor test directly in §27, criticizing its use as a general causation standard and reserving it for the narrow overdetermination scenario involving duplicative sufficient causes.

Core Mechanics or Structure

The But-For Test

The but-for test is expressed as a counterfactual conditional: "But for the defendant's negligent conduct, would the plaintiff's injury have occurred?" If the answer is no — the injury would not have happened without the negligence — causation is established. If the injury would have occurred regardless, the causal link fails.

In medical malpractice, expert testimony is the primary mechanism for satisfying but-for causation. An expert witness must typically testify to a reasonable degree of medical probability — a threshold courts in most jurisdictions interpret as greater than rates that vary by region likelihood — that the negligent act caused the specific harm. This is distinct from a scientific certainty standard; the legal probability threshold is its own measure. The role of expert witnesses in establishing this threshold is addressed in detail at expert witnesses in malpractice cases.

The Substantial Factor Test

Originally articulated in the Restatement (Second) of Torts §431–433 (American Law Institute, 1965), the substantial factor test asks whether the defendant's conduct was a material element contributing to the harm — not merely a trivial or theoretical contributor. Courts applying this test consider:

1. The number of other contributing factors
2. Whether the defendant's conduct was continuous or repeated
3. The lapse of time between the negligent act and the resulting harm
4. Whether the harm was foreseeable from the defendant's conduct

The substantial factor formulation emerged largely from cases involving concurrent independent causes — two fires, two sources of toxic exposure — where each cause alone could produce the full harm. In that narrow overdetermination context, the but-for test produces a logical paradox: neither cause alone satisfies it, yet both plainly contributed.

Causal Relationships or Drivers

Factual Background: What Creates Causation Complexity

The complexity of causation in malpractice cases derives from four principal sources:

Pre-existing conditions. A patient with a pre-existing disease may have suffered harm partly from the disease progression and partly from the practitioner's negligence. Courts apply the "thin skull" or "eggshell plaintiff" doctrine to hold defendants responsible for full harm to plaintiffs with pre-existing vulnerabilities, but that doctrine does not eliminate the need to prove the negligent act contributed to the specific worsening or new harm.

Multiple defendants. Where two or more practitioners each contributed negligently to a single injury — as occurs in surgical errors and malpractice or hospital liability and institutional malpractice contexts — courts must determine whether the concurrent negligence satisfies but-for individually or whether the substantial factor test applies.

Probabilistic harm. The loss of chance doctrine, detailed at loss of chance doctrine malpractice, modifies traditional causation analysis in cases where a defendant's negligence reduced a patient's statistical chance of survival or recovery. As of the 2023 edition of the Restatement (Third) of Torts, approximately many states had adopted some form of the loss of chance doctrine (American Law Institute commentary to §26).

Delayed manifestation injuries. Harm that materializes months or years after the negligent act — common in misdiagnosis and delayed diagnosis malpractice — requires bridging testimony linking the delay causally to a defined worsening in outcome.

Classification Boundaries

The but-for and substantial factor tests are not the only causation doctrines courts use. The full taxonomy includes:

The Restatement (Third) of Torts §27 explicitly limits the substantial factor test to the overdetermination category and recommends against applying it as a general-purpose substitute for but-for analysis. California, which retained substantial factor as its general causation instruction under CACI No. 430, is the most frequently cited jurisdiction using the broader version.

Tradeoffs and Tensions

Scientific vs. Legal Standards of Causation

The gap between scientific and legal causation standards creates persistent friction. Epidemiological and clinical science often expresses causation in population-level relative risks or confidence intervals — tools that do not map cleanly onto the individual-level probability demanded by tort law. A relative risk of 2.0 (doubling of population risk from exposure) is treated in some courts as the approximate threshold for "more likely than not" individual causation, but this translation is contested in the literature and in judicial gatekeeping under Federal Rule of Evidence 702 and Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993).

The Substantial Factor Problem

Using substantial factor as a general instruction rather than limiting it to overdetermination cases risks lowering the causation threshold below the preponderance standard. The American Law Institute criticized this expansion in the comments to Restatement (Third) §26, warning that it collapses causation analysis into a vague contribution test that undermines the but-for foundation.

Conversely, strict but-for analysis can produce unjust results in multi-cause toxic exposure cases or complex surgical scenarios involving 3 or more concurrent actors — each actor evades liability because no single actor's conduct alone satisfies but-for causation.

Loss of Chance and the Causation Threshold

The loss of chance doctrine directly conflicts with the traditional preponderance requirement. Where a patient had only a rates that vary by region survival chance before negligence and the negligence reduced it to rates that vary by region, the plaintiff cannot show it is more probable than not that but-for the negligence the patient would have survived. Jurisdictions adopting the doctrine either treat the lost chance itself as the compensable injury or proportionally reduce damages — each approach carrying distinct implications for malpractice damages: compensatory and punitive.

Common Misconceptions

Misconception 1: Proving a deviation from the standard of care automatically establishes causation.
Breach and causation are independent elements. A practitioner may have deviated from accepted practice in a way that had no causal connection to the harm that actually occurred. Courts require separate expert testimony addressing causation, not merely negligence.

Misconception 2: The substantial factor test is easier to satisfy than but-for.
In jurisdictions applying the Restatement (Third) approach, substantial factor is reserved for a narrow category and requires showing the conduct was a sufficient independent cause — not merely a contributing factor in a chain. In those jurisdictions, substantial factor may actually be a stricter standard than it appears.

Misconception 3: "Reasonable degree of medical probability" means scientific certainty.
The legal threshold of reasonable medical probability means greater than rates that vary by region likelihood as assessed by a qualified expert, not certainty to within a confidence interval required by research-based publication standards. These are entirely distinct evidentiary frameworks.

Misconception 4: Loss of chance is a causation theory recognized in all states.
As of the 2023 ALI commentary, loss of chance has been rejected outright in roughly some states, adopted in approximately many states in some form, and remains unsettled in the remainder. Applying it in a rejecting jurisdiction is a case-ending error.

Misconception 5: But-for causation can be satisfied by showing the outcome was bad.
A bad outcome — including patient death — is not itself evidence of causation. The but-for inquiry requires affirmative proof that a specific negligent act or omission changed the outcome from what it would otherwise have been.

Checklist or Steps (Non-Advisory)

The following represents the analytical sequence courts and parties typically apply when evaluating causation in a malpractice claim. This is a descriptive framework only, not legal guidance.

Step 1 — Identify the specific negligent act or omission
Causation analysis begins with a precisely defined act or omission, not a general characterization of poor care. Vague allegations produce vague causation testimony.

Step 2 — Determine the applicable causation standard
Assess whether the case involves a single cause (but-for applies by default), concurrent independent sufficient causes (substantial factor under Restatement Third §27 may apply), or probabilistic harm (loss of chance doctrine may apply depending on jurisdiction).

Step 3 — Identify the claimed harm with specificity
The harm must be defined with enough particularity that an expert can assess counterfactual outcomes. "Worse health generally" is insufficient; "delayed diagnosis of Stage II colorectal cancer resulting in progression to Stage IV" is the type of specificity courts require.

Step 4 — Secure qualifying expert testimony
Under Federal Rule of Evidence 702 and Daubert, expert causation testimony must rest on sufficient facts or data, apply reliable methods, and reliably connect the methods to the facts of the case. State courts apply parallel gatekeeping standards under either Daubert or the older Frye general acceptance test.

Step 5 — Address pre-existing conditions
Document the baseline condition before the negligent act. Causation testimony must distinguish harm attributable to the pre-existing condition from harm attributable to the defendant's conduct.

Step 6 — Assess concurrent actors and apportion causation
In multi-defendant cases, determine whether jurisdiction applies joint and several liability, proportionate fault, or modified comparative fault rules, as these affect how causation must be apportioned. See contributory and comparative negligence in malpractice.

Step 7 — Evaluate proximate cause (scope of liability)
After cause-in-fact is satisfied, assess whether the harm was within the foreseeable risk that made the defendant's conduct negligent — the proximate cause inquiry under Restatement (Third) §29.

Reference Table or Matrix

But-For Test vs. Substantial Factor Test: Comparative Framework

Causation Doctrine Application by Case Type

References

• American Law Institute — Restatement (Third) of Torts: Liability for Physical and Emotional Harm (2010)
• American Law Institute — Restatement (Second) of Torts §431–433 (1965)
• Federal Rules of Evidence, Rule 702 — Testimony by Expert Witnesses (U.S. Courts)
• Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993) — Cornell Legal Information Institute
• California Civil Jury Instructions (CACI) No. 430 — Judicial Council of California
• National Practitioner Data Bank — Health Resources and Services Administration (HRSA)
• Sindell v. Abbott Laboratories, 26 Cal.3d 588 (1980) — reference via California Courts
• Federal Rule of Evidence 702 — Advisory Committee Notes, 2023 Amendment (U.S. Courts)