Abstract
In Britain, the NHS spends millions of pounds a year compensating patients injured during medical treatment. Compensation is paid if the patient can demonstrate that treatment was supplied negligently. However, concern over the cost, effectiveness and administrative efficiency of this approach has led jurisdictions like Sweden, New Zealand and some US states to alter the basis for compensation, and the Department of Health has now published proposals for reform in England. We assess the current approach in England and provide costings for some key alternatives to have featured in the latest policy debate. We draw lessons for reform from international experience. Sustaining an injury related to medical care can have serious economic and health consequences and such injuries are a growing concern in many countries due to a perception that the costs they give rise to are rapidly increasing. Public policy in this area can be regarded as having to address two key objectives: providing compensation to those who have suffered injuries and providing incentives to practitioners to supply an appropriate standard of care. However, different countries have adopted widely varying strategies in pursuit of these objectives, and in particular have adopted very different approaches to the question of liability for injury. In the UK, victims of medical injuries can claim compensation by filing a legal claim against the alleged perpetrator, such as a hospital or individual practitioner. In order to secure an award of damages, the claimant must prove that the defendant is at fault for the injuries cited, or, in the current UK terminology, that the tort of clinical negligence has occurred. This negligence liability approach, however, is not found in all jurisdictions. For example, in Sweden, New Zealand and some American states, it is sufficient mainly to prove that the defendant caused the injuries in order to receive damages. Because it is not necessary to prove negligence, these schemes are often termed ‘no-fault’ schemes, and are typically thought to have lower administrative costs due to the reduced burden of proof.1 These contrasting approaches to medical injury have many distributional and other consequences: for example, under negligence liability, if injurers take efficient care the victims bear all losses, whereas under no-fault the injurers bear all losses irrespective of their care decision. However, research attention has focused on two particular areas of interest: first, assessing the respective cost and efficiency characteristics of negligence liability and no-fault schemes and, second, seeing whether other composite models can be created; for instance, models in which some element of fault must be proven but in which the cost, delay and complexity involved in obtaining compensation are reduced by streamlining legal procedures and/or transferring liability from individual physicians to their employers. The UK's fault-based approach to compensation has received significant criticism for a number of years and has been accused of failing either to provide fair compensation or to create incentives for deterrence. Most recently, the National Audit Office (NAO, 2001), a major public inquiry (Kennedy, 2001) and the House of Commons Public Accounts Committee (PAC, 2002) have each proposed radical reform. The principal charges are that the system is costly to the NHS, imposes long delays on patients seeking redress, is administratively inefficient (in that the legal and administrative costs of cases regularly exceed the value of the damages at stake), and engenders a culture of secrecy and cover-up in which colleagues are unwilling to ‘blow the whistle’ on bad practice and is therefore inimical to improved standards of care. In support of these charges, it has been noted that in 2000/01 total NHS provisions for future settlements of negligence claims stood at £4.4 bn; that the average time to resolution of the 23,000 cases outstanding at the end of March 2000 (excluding the longest, most complex, cases) was over five years; and that in 65% of settlements below £50,000, legal costs exceeded damages.2 Recognising these concerns, the Secretary of State for Health established an Advisory Group in 2001 to examine possible reforms, and in June 2003 a set of proposals was published for consultation (DoH, 2003). Whilst it rejects a general move to no-fault compensation, a stated aim of this consultation document is to ‘move the role of tort from its current central position to the outer perimeter of the NHS’ (Chapter 8, paragraph 10). In its place would be a ‘Redress Scheme’ containing a fast procedure for resolving ‘small claims’ (those with value below £30,000) – ultimately involving a relaxed standard of care – and a no-fault procedure for compensating birth-related neurological injuries (typically those injuries with the highest cost and complexity). Under these proposals, a person would retain the right to sue through the courts but (except for neurologically impaired babies) with a presumption that they had first applied to the NHS Redress Scheme. Those accepting packages of care (where possible to be provided by the NHS) and compensation under the scheme would be required to waive their right to go to court on the same case. It is clear that the Department of Health's proposals constitute potentially significant reforms to the way clinical injuries are compensated by the NHS (and, given the sums involved, to resource allocation within the NHS). Yet the empirical basis for the debate to which they contribute has not been strong in the UK.3 The purpose of this paper is to present findings from new research aimed at casting fresh light on this debate. The research was commissioned to inform the deliberations of the Secretary of State for Health's Advisory Group; see Fenn et al. (2002). The paper is structured as follows. Section 1 reviews briefly the economic theory underlying the objectives of patient compensation schemes. In Section 2 we describe the institutional arrangements within which the English clinical negligence scheme operates,4 discuss its annual cash cost and assess its deterrence effects. Section 3 summarises what is known about the structure and performance of several overseas no-fault schemes in order to draw lessons for reform in England and identify data requirements for modelling such reform. In Section 4 we estimate, for the first time, the cost of possible reforms, paying particular attention to ‘small claims’ fault-based schemes of the type proposed in DoH (2003) and to a commonly cited model for no-fault compensation in England: the Swedish no-fault scheme. Section 5 concludes the paper.