Medical Malpractice

Medical Malpractice and the Economics of Liability

Health economists treat medical malpractice as an incentive system as well as a legal wrong. The threat of a claim is meant to make providers bear the cost of harm caused by careless care, and the payment is meant to compensate the injured patient. US studies suggest the system does both jobs poorly, since few negligently injured patients claim and much of the money goes on lawyers and administration. Liability pressure also changes clinical behaviour, prompting tests and procedures ordered to avoid a lawsuit rather than to improve health, which is known as defensive medicine. In England the usual term is clinical negligence, and claims against the NHS are managed by NHS Resolution.

Negligence, causation and injury as the legal test

Kessler describes three elements that a successful claim in the United States usually has to show. The patient suffered an adverse event, the provider's action or inaction caused it, and the provider took less care than the average member of the profession in good standing would have taken in the circumstances. NHS Resolution explains liability in England in similar terms, as treatment that falls below a reasonable standard of competence and results in an injury that would likely have been avoided, or been less severe, with appropriate care.

The test separates malpractice from bad outcomes in general. The Harvard Medical Practice Study reviewed 30,121 New York hospital records from 1984 and found adverse events in 3.7% of hospitalisations, of which 27.6% were due to negligence. Scaled to the state, that meant 27,179 negligent adverse events among 2,671,863 discharges.

Two economic jobs for one liability system

Kessler names two objectives, compensating patients injured through negligence and deterring negligent practice. Mello and colleagues add corrective justice for the injured person as a third benefit. In the economic reading, liability attaches an expected price to careless care, so a provider weighing a precaution compares its cost with the expected liability it avoids.

Deterrence weakens when that price is rarely charged. Matching the Harvard study's records against statewide claims data, Localio and colleagues found that only 8 of 280 patients with negligent adverse events filed a claim, a weighted rate of 1.53%. The statewide ratio of negligent adverse events to claims was 7.6 to 1 (27,179 to 3,570). Localio and colleagues add that this ratio overstates the chance of a claim, because most claims in the sample did not involve negligent adverse events. Kessler adds that malpractice insurance is typically not strongly experience-rated, so premiums track claims history only loosely and the signal reaching each clinician is blurred. Mello and colleagues conclude that reliable evidence about the deterrent effect of the tort system does not exist.

The English scheme for hospital trusts, the Clinical Negligence Scheme for Trusts (CNST), meets its costs from members' contributions, while the trust remains the legal defendant. Contributions reflect the type of trust, its specialties and its clinical staffing, and members with fewer, less costly claims pay less.

How accurately claims match negligent harm

Studdert and colleagues had physicians review 1,452 closed claims from five liability insurers. They found no verifiable injury in 3% of claims and no error in 37%. Most claims without an error (72%) received nothing, while most claims involving injury from error (73%) were paid, and claims without errors made up 13% to 16% of the system's total costs. For every dollar of compensation, 54 cents went on administrative expenses such as lawyers, experts and courts, a transaction cost of resolving disputes through litigation.

Using one nationwide insurer's 1991 to 2005 data, Jena and colleagues found that 7.4% of physicians had a claim each year but only 1.6% had a claim that led to a payment. The yearly share facing a claim ranged from 19.1% in neurosurgery to 2.6% in psychiatry, and by the age of 65 an estimated 75% of physicians in low-risk specialties and 99% in high-risk specialties had faced one.

Defensive medicine and its cost-effectiveness

Mello and colleagues cite the most common definition, from the former US Office of Technology Assessment, under which defensive medicine occurs "when doctors order tests, procedures, or visits, or avoid certain high-risk patients or procedures, primarily (but not solely) because of concern about malpractice liability." Kessler distinguishes positive defensive medicine, which supplies unproductive or harmful care, from negative defensive medicine, which withholds beneficial care or drives clinicians from practice. He attributes positive defensive medicine to moral hazard from health insurance, because neither patients nor physicians bear most of the cost of care in any one case, while the lost time, stress and reputational damage of a suit stay uninsured.

In a 2003 survey of Pennsylvania physicians in six high-risk specialties, 93% reported practising defensive medicine and 92% reported assurance behaviour such as extra tests, procedures and referrals. Among those who described their most recent defensive act, 43% reported using imaging in clinically unnecessary circumstances.

Kessler and McClellan measured actual spending instead. Using data on elderly Medicare patients treated for serious heart disease in 1984, 1987 and 1990, they found that reforms which directly reduce liability lowered medical expenditure by 5% to 9% without substantial effects on mortality or medical complications. Kessler later reported that study's cost-effectiveness ratio for liability-induced intensive treatment as over USD 500,000 per additional one-year survivor, well above a commonly accepted value of about USD 100,000 for a year of life. In cost-effectiveness analysis terms, the extra spending bought survival at an ICER more than five times the value Kessler used as a cost-effectiveness threshold.

Counting the cost of the liability system

Mello and colleagues estimated annual US liability costs at USD 55.6 billion in 2008 dollars, 2.4% of health spending. Their components were indemnity payments of USD 5.72 billion, administrative expenses of USD 4.13 billion, defensive medicine of USD 45.59 billion and lost clinician work time of USD 0.20 billion. They left out insurance premiums, which already contain expected indemnity and defence costs, to avoid double counting. They rated the evidence for the defensive medicine figure as low.

For England, NHS Resolution received 15,236 new clinical negligence claims and reported incidents in 2025/26. Payments on clinical schemes totalled GBP 3,238.5 million, made up of GBP 2,427.9 million in damages, GBP 622.2 million in claimant legal costs and GBP 188.4 million in NHS legal costs. The estimated cost of harm for incidents in that year was GBP 4.8 billion, and the total clinical negligence provision was GBP 60.0 billion. Maternity accounted for 48% of that cost of harm and 58% of the provision. The schemes are funded on a pay-as-you-go basis, with members contributing enough each year to meet that year's liabilities.

In US physician payment, the professional liability insurance component of the Medicare fee schedule is based on the premiums clinicians pay for malpractice insurance. This is the Malpractice RVU, one of the three parts of each service's relative value unit weight, so liability costs feed directly into fee-for-service prices.

Worked example of the deterrence gap with illustrative numbers

The figures below are hypothetical and chosen for illustration. A hospital could add a monitoring step after surgery that costs GBP 40 per patient. The step cuts the probability of a negligent injury by 0.002, and each such injury causes a loss of GBP 50,000, assumed equal to the damages a court would award.

$$S = q \times \Delta p \times L$$

Here $S$ is the expected liability saving per patient from taking the precaution, $q$ is the probability that a negligent injury leads to a paid claim, $\Delta p$ is the fall in the probability of a negligent injury, and $L$ is the loss per injury.

Step 1 (social test). The expected harm avoided is $\Delta p \times L = 0.002 \times 50{,}000 = 100$ GBP per patient, which exceeds the cost of 40, so the precaution is worth taking and skipping it is the kind of care a negligence standard aims to deter.

Step 2 (full enforcement). If every negligent injury led to a paid claim, $q = 1$ and the saving is $S = 1 \times 0.002 \times 50{,}000 = 100$ GBP, so a provider facing the full liability would adopt the step.

Step 3 (claims rarely made). If only 2% of negligent injuries led to a paid claim, a little above the 1.53% filing rate in the New York records, which is generous since not every claim is paid, the saving is $S = 0.02 \times 0.002 \times 50{,}000 = 2$ GBP, far below the cost of 40.

Step 4 (break-even claim rate). Liability alone justifies the step only when the claim rate reaches

$$q^{*} = \frac{B}{\Delta p \times L} = \frac{40}{0.002 \times 50{,}000} = 0.4$$

where $B$ is the cost of the precaution per patient, so 40% of negligent injuries would need to end in a paid claim.

Step 5 (the opposite error). Now take a defensive test costing GBP 250 that adds 0.002 QALYs. Its ratio is $\mathrm{ICER} = 250 / 0.002 = 125{,}000$ GBP per QALY, far above an illustrative threshold of GBP 20,000 per QALY, yet a clinician who pays nothing for the test but bears the uninsured stress of a claim may still order it.

The example shows how one system can under-deter cost-effective precautions while encouraging care that fails a cost-effectiveness test. Tying each provider's contribution to its own claims record strengthens the first signal, which is why reform proposals often pair compensation changes with experience rating.

Where malpractice sits beside nearby ideas

Every negligent injury is an adverse event, but most adverse events involve no negligence, as the Harvard study's 27.6% shows. An error that causes no injury is not malpractice either, which is why most claims without a verifiable injury in the sample Studdert and colleagues reviewed (31 of 37) received no payment.

Defensive medicine overlaps with supplier-induced demand, since both add services that informed patients might not choose. Mello and colleagues note that when physicians respond to rising premiums by doing more, it is impossible to tell how much of the extra volume is defensive and how much is meant to raise revenue.

No-fault compensation is the main alternative to tort. New Zealand replaced its tort-based system in 1974 with a government-funded scheme and in 2005 extended eligibility to all treatment injuries. Kessler notes that no-fault and health court proposals generally aim to compensate more patients less generously, usually alongside stronger experience rating to protect incentives for precaution.

Limitations and common misreadings

Most of the evidence is from the United States, where liability rules and payment differ from the NHS. The defensive medicine estimates rest heavily on cardiac care for older Medicare patients, and Mello and colleagues note that two other studies could not replicate them for other conditions.

1. Claims measure harm. Most negligent injuries in the New York records never produced a claim, so claims data say more about who sues than about negligent harm.
2. Most claims are baseless. Over a third of reviewed claims involved no error, but most of these were not paid and they made up a minority of total costs.
3. Liability costs are premiums or payouts. Premiums already contain expected indemnity and defence costs, so adding them to payouts double counts, and in the US estimate defensive medicine was far larger than payouts.
4. Every compensation payment is a social cost. Mello and colleagues treat awards such as lost income as transfers that another party would otherwise have borne, so from a societal perspective they arguably should not be counted, although Mello and colleagues kept them in their estimate.

Sources

• Kessler DP. Evaluating the medical malpractice system and options for reform. Journal of Economic Perspectives. 2011;25(2):93-110. https://doi.org/10.1257/jep.25.2.93
• Kessler D, McClellan M. Do doctors practice defensive medicine? Quarterly Journal of Economics. 1996;111(2):353-390. https://doi.org/10.2307/2946682
• Brennan TA, Leape LL, Laird NM, et al. Incidence of adverse events and negligence in hospitalized patients. Results of the Harvard Medical Practice Study I. New England Journal of Medicine. 1991;324(6):370-376. https://doi.org/10.1056/NEJM199102073240604
• Localio AR, Lawthers AG, Brennan TA, et al. Relation between malpractice claims and adverse events due to negligence. Results of the Harvard Medical Practice Study III. New England Journal of Medicine. 1991;325(4):245-251. https://doi.org/10.1056/NEJM199107253250405
• Studdert DM, Mello MM, Gawande AA, et al. Claims, errors, and compensation payments in medical malpractice litigation. New England Journal of Medicine. 2006;354(19):2024-2033. https://doi.org/10.1056/NEJMsa054479
• Studdert DM, Mello MM, Sage WM, et al. Defensive medicine among high-risk specialist physicians in a volatile malpractice environment. JAMA. 2005;293(21):2609-2617. https://doi.org/10.1001/jama.293.21.2609
• Jena AB, Seabury S, Lakdawalla D, Chandra A. Malpractice risk according to physician specialty. New England Journal of Medicine. 2011;365(7):629-636. https://doi.org/10.1056/NEJMsa1012370
• Mello MM, Chandra A, Gawande AA, Studdert DM. National costs of the medical liability system. Health Affairs. 2010;29(9):1569-1577. https://doi.org/10.1377/hlthaff.2009.0807
• Bismark M, Paterson R. No-fault compensation in New Zealand: harmonizing injury compensation, provider accountability, and patient safety. Health Affairs. 2006;25(1):278-283. https://doi.org/10.1377/hlthaff.25.1.278
• NHS Resolution. Annual report and accounts 2025/26. Published 9 July 2026. https://www.gov.uk/government/publications/nhs-resolution-annual-report-and-accounts-2025-to-2026
• NHS Resolution. Clinical Negligence Scheme for Trusts (web page), accessed 14 October 2026. https://resolution.nhs.uk/services/claims-management/clinical-schemes/clinical-negligence-scheme-for-trusts/
• Medicare Payment Advisory Commission. Physician and other health professional payment system. Payment basics, revised October 2024. https://www.medpac.gov/wp-content/uploads/2024/10/MedPAC_Payment_Basics_24_Physician_FINAL_SEC.pdf