TY - JOUR
AU1 - Heywood,, Rob
AB - Abstract The law of negligence, as it applies to General Practitioners (GPs), is underexplored in the literature. There has been no substantial research undertaken that has penetrated deeper into claims that have actually reached court in order to analyse judicial reasoning pertaining to both breach of duty and causation. Given the increased pressures that GPs now face, these are important questions to consider. It is against this backdrop that this article seeks to present the findings of an empirical investigation into a number of reported clinical negligence claims brought against GPs. This analysis provides an original contribution to the developing academic discussion surrounding the changing nature of the doctor–patient relationship, and how it has come to be viewed in the eyes of the law. It also assesses the extent to which judges have become more receptive to protecting patient rights through the law of negligence, engaging in the expanding discourse concerning judicial deference to medical decision-making. It is argued that judges should sometimes show a greater propensity to question expert medical testimony in support of GPs, because some of the issues GPs typically face are less complex than in other clinical negligence cases involving technical areas of medicine, and that causation does not appear to be such a key factor in defeating patient claims. The work also provides useful guidance for GPs and their advisers in respect of where liability is most likely to be founded and how behaviour can be modified accordingly to reduce the chances of being sued. Breach of Duty, Causation, Communication, GPs, Negligence, Referrals I. Introduction Unless there is a serious emergency, or out of hours incident, General Practitioners (GPs) are usually the first port of call for patients when they begin to experience symptoms of illness. The job of a GP, however, is not an easy one.1 Her role in the provision of modern healthcare is expanding and now demands difficult judgement calls to be made in respect of allocation of scarce resources and, in some instances, the performance of minor surgical procedures.2 Other more routine functions performed by GPs include such things as prescribing antibiotics, providing general advice and information to patients about how to deal with illnesses and, crucially, referring patients to hospital.3 The latter may be required immediately owing to a patient presenting with symptoms that are consistent with a medical emergency or, alternatively, a referral may be made on a non-urgent basis for further diagnostic procedures or a specialist consultation. Claims against GPs do not usually make it as far as court, but occasionally they do, and it is here where, from a legal perspective, a number of important considerations have to be borne in mind. Faced with a high volume of patients, under considerable resource and time constraints, a GP must reach a conclusion very quickly about how best to deal with the list of symptoms exhibited. Thus, where a claim reaches court, a judge must remain cognisant of these difficulties when seeking to determine the appropriate legal standard of care, for if the threshold for establishing a breach of duty against a GP is set at too low a level, it may unreasonably increase her legal exposure. Equally taxing questions exist in relation to causation, because the pathology of certain illnesses may render it incredibly difficult to prove that any breach of duty on the part of a GP caused or materially contributed to the harm complained of by the patient. The law of negligence, as it applies to GPs, is underexplored in the literature. An existing study has investigated the volume and type of claims made against GPs,4 and an empirical investigation reflecting legally on the altruistic tendencies of GPs has also been published.5 However, there has been no substantial research undertaken that has penetrated deeper into claims that have actually reached court in order to analyse judicial reasoning pertaining to both breach of duty and causation. Despite only a modest number of claims being made against GPs,6 the increased pressures that they now face mean that these are important questions to consider. 7 It is against this backdrop that this article seeks to present the findings of an empirical investigation into a number of reported clinical negligence claims brought against GPs. From a theoretical legal perspective, this analysis provides an original contribution to the developing academic discussion surrounding the changing nature of the doctor–patient relationship and how it has come to be viewed in the eyes of the law.8 It also assesses the extent to which judges have become more receptive to protecting patient rights through the law of negligence,9 engaging in the expanding discourse concerning judicial deference to medical decision-making.10 On the basis of the findings, it is argued that judges should sometimes show a greater propensity to question expert medical testimony in support of GPs, because some of the issues GPs typically face are less complex than in other clinical negligence cases involving technical areas of medicine, and that causation does not appear to be such a key factor in defeating patient claims. The work also has practical significance, providing useful guidance for GPs and their advisers in respect of where liability is most likely to be founded and how behaviour can be modified accordingly to reduce the chances of being sued. The article begins by setting out the legal framework underpinning the study. The narrative then outlines the methodology employed and explains the method of data analysis. Thereafter, the work proceeds to discuss the substantive research findings and concludes by offering some tentative observations about how they may inform future clinical practice. II. Setting the Scene: The Legal Framework A. Breach of Duty As a result of the Bolam test, claimants traditionally faced great difficulty when attempting to pursue a negligence claim against a GP.11 The test stated that, provided a medical practitioner acted in accordance with a responsible body of medical opinion, she was not liable in negligence.12 It mattered not that there was a body of opinion that existed which would have acted to the contrary.13 Under conventional Bolam, judges were prevented from preferring one body of medical opinion to another, 14 the consequence of which was that they were prohibited from executing their usual function in professional negligence cases, which is to objectively scrutinise competing bodies of expert testimony in order to ascertain what is reasonable.15 Amidst the range of options and uncertainties faced by a GP when deciding how best to treat a patient, it would nearly always be possible for her to adduce a body of expert opinion to support her adopted course of action. Given that there was effectively no scope for a judge to scrutinise, or indeed question, the rationale and thought process underpinning that evidence, it was rare for a judge to ever hold that the body of opinion supporting a defendant was, in fact, unreasonable.16 However, subtle legal developments then began to take shape. In Hucks v Cole, an early GP case, it was suggested by Sachs LJ in the Court of Appeal that ‘when the evidence shows that a lacuna in professional practice exists by which risks of grave danger are knowingly taken, then, however small those risks, the courts must anxiously examine that lacuna – particularly if the risks can be easily and inexpensively avoided’.17 This delivered a clear message that, in some instances, it was appropriate for judges to question the basis of medical decision-making and, where necessary, to condemn as negligent any course of action that exposed a patient to an unreasonable risk. Building on Hucks, the decision of the then House of Lords in Bolitho is sometimes said to be the case that finally redressed many of the difficulties created by Bolam.18 In his judgment, Lord Browne-Wilkinson confirmed that judges did retain the power to exercise independent objective scrutiny of expert medical testimony in clinical negligence cases.19 Significantly, in the wake of Bolitho, any purported responsible body of medical opinion could only be held to be such if it was capable of withstanding the logical scrutiny of a judge. In determining the logicality of any expert opinion, judges were directed to give their minds to the relative risk versus benefit ratio of the course of action advocated.20 In the context of claims against GPs, after Bolitho, there ought to be greater scope for a judge to question any expert testimony lending support to a defendant’s case,21 yet, whether or not these risk versus benefit calculations are carried out in the way that they should be remains a moot point.22 An important question that this study therefore seeks to explore is the impact of Bolitho on cases against GPs.23 Is there a willingness from judges to scrutinise expert testimony in a way that operates to the benefit of claimants, or is there still a tendency to classify a body of medical opinion in support of a GP as being responsible, merely because one such opinion exists? In addition to Bolitho, the law of clinical negligence has recently undergone a change in mindset. The decision of the Supreme Court in Montgomery v Lanarkshire has arguably redefined the doctor–patient relationship in the eyes of the law.24 With regard to a doctor’s duty of disclosure, Lords Kerr and Reed held that a doctor is under a duty to take reasonable care to ensure that the patient is aware of any material risks involved in any recommended treatment and of any reasonable alternative or variant treatments. The test of materiality was defined as whether, in the circumstances of a case, a reasonable person in the patient’s position would be likely to attach significance to the risk, or the doctor is or should reasonably be aware that the particular patient would be likely to attach significance to it. This was subject to what is known as the therapeutic privilege, which entitles a doctor to withhold information from a patient if she reasonably considers that its disclosure would be seriously detrimental to the patient’s health.25 While the case concerned risk disclosure, the judgment itself transcends that single issue and is symbolic because of the way in which it elevates the position of patient rights within the law of negligence. The notion of patient empowerment through the exercise of self-determination occupied centre stage in the reasoning of their Lordships, and they distanced themselves from any historical judicial attitudes that were accepting of unqualified deference to the medical profession.26 The direction of travel of medico-legal jurisprudence is now patient-facing as opposed to doctor-led which, ultimately, may lead to a subtle change in emphasis in the reasoning of judges in the lower courts.27 It is anticipated that the impact of Montgomery in respect of claims against GPs will be minimal. First, it is still too early to assess with any accuracy the true effect that the judgment will have on claims of this type; it is only after a body of case law has built up over time that it will become possible to undertake any rigorous analysis. Secondly, the majority of claims made against GPs will not involve allegations of negligent disclosure and so the case is not squarely on point. Nevertheless, the importance of Montgomery should not be underestimated.28 In the aftermath of the judgment it is possible that judges may be more inclined to approach cases with a sympathetic eye towards patients, especially in the light of the recognised asymmetry of position between the two parties.29 Quite apart from that though, there are occasions in primary care where information disclosure may become more of a substantive legal issue in its own right. Where, for example, GPs perform minor operations, in the future, Montgomery may well be relevant in terms of the consent processes that are undertaken before those minor procedures are performed. Thus, post-Montgomery, it will be interesting to assess whether there is any evidence of a general change in attitude from judges which appears to be more patient-oriented, and whether there are any cases against GPs that focus directly on information disclosure. B. Causation Running parallel to the legal issues concerning breach, there are also pertinent questions relating to causation. The situations in which it ought to be incumbent upon a GP to refer a patient to hospital immediately will often involve aggressive, rapid progression illnesses. One problem with illnesses of this kind is being able to prove that, had a referral been made, things would have turned out differently for the patient. It may well be impossible to prove that had an earlier hospital referral been made, any subsequent damage would have been avoided because, no matter what the hospital would have done, the long-term harm caused by the destructive illness would already have occurred.30 Further problems are also encountered whereby the conduct of both a hospital and a GP is called into question. Where, upon receiving a patient, a hospital is also alleged to have been negligent in its diagnosis and treatment, it may become difficult to pinpoint whose negligence was the main contributing cause of any damage to the patient. Was the late, or lack of, referral by a GP a sufficient causal factor to the underlying harm, or, irrespective of that negligence, did the patient still arrive at hospital at a point at which non-negligent care would have rendered the harm avoidable? If the negligence of the hospital failed to prevent harm at a stage where it could still have been avoided, it provides a GP with an opportunity to argue that the responsibility should shift onto the hospital.31 Since the Court of Appeal decision in Wright (A Child) v Cambridge Medical Group (A Partnership), this particular avenue ought to have been closed off.32 Where both a GP was negligent in providing a late referral, and the hospital was also then negligent in its treatment, it was held that the hospital’s treatment of the patient was not such an egregious event, in terms of the degree of unusualness of the negligence, or the period of time for which it lasted, to defeat or destroy the causative link between the GP’s negligence and the patient’s injury.33 Moreover, a causal link existed between the late referral and the eventual harm suffered, as it deprived the patient of the opportunity to receive prompt and appropriate medical care in hospital, which would have meant there was more time to treat the patient effectively.34 The extent to which this ruling has since been relied upon by both claimants and defendants in GP cases is, thus, fertile ground for exploration.35 GPs will also encounter progressive illnesses that evolve at varying rates, often concealing the deleterious effect they may be having on the patient until it is too late.36 In a situation where a GP negligently fails to spot a potential symptom of cancer and, as a result, the patient then receives a late diagnosis, the patient will be likely to argue that, had an appropriate diagnosis and subsequent referral been made at the correct time, they would have stood a much better chance of survival. In Gregg v Scott, a delayed cancer diagnosis reduced the claimant’s chance of a cure from 42 per cent to 25 per cent.37 He had, thus, lost a 17 per cent chance of a cure.38 The majority of their Lordships in the House of Lords found against the claimant on the basis that he could not prove that had there been no breach of duty on the part of his GP, he would have had a better than 50 per cent chance of a cure.39 While it has since been suggested by the Court of Appeal that recovery in loss of chance claims should not be considered completely foreclosed in clinical negligence actions, it was said to be an extension to the law that could only be made by the Supreme Court.40 At present, recovery for a loss of a chance in clinical negligence is not permissible, and if this line of argument is pursued frequently in claims against GPs, claimants will fail to recover damages. III. The Study A. Context Clinical negligence claims made against GPs, for a variety of reasons, are not as high as the number of claims made against healthcare professionals in secondary care.41 As Brazier and Cave accurately assert: ‘many common complaints about general practice do not tend to be the kind of grievances that make litigation with its expense, pomp and ceremony worthwhile’.42 The Medical Defence Union (MDU) has previously identified that common complaints against GPs include such things as poor communication or attitude, rudeness, substandard out of hours care, difficulty in procuring a home visit or where a patient feels they have been unfairly removed from a GP’s list.43 Conduct of this nature, while irritating for those affected by it, will rarely cause harm of the magnitude that warrants full-scale litigation, although it may still have other professional ramifications.44 Other types of complaint identified by the MDU were less prevalent and included allegations of breach confidentiality and inadequate consent.45 What has been established though is that a GP’s legal exposure is at its highest in relation to inappropriate prescribing46 and, more frequently, in respect of misdiagnosis and/or lack of appropriate referral.47 Even so, the chances of these allegations being pursued as far as court are still incredibly remote and, with this in mind, over the course of history, GPs ought not to have been overly concerned about becoming the subject of a lawsuit; nowadays, they should worry more about sanctions that could be imposed by the General Medical Council (GMC).48 Nonetheless, increased pressures on general practice, alongside an evolving legal landscape, bring the relevance of this study more sharply into focus.49 Existing research has highlighted that the quality of care provided by GPs is inconsistent, varying significantly across different geographical regions and from practice to practice.50 It is perhaps unsurprising then that, in recent times, legal claims against GPs have become more common,51 which ought to be a concern not only for GPs but also for medical defence organisations.52 Set against these considerations, Brazier and Cave’s assertion that ‘changes in general practice which alter the more personal relationship between GP and patient may increase the GP’s vulnerability to claims for clinical negligence’, is one possible explanation as to why claims have become more frequent.53 However, important questions remain, which are in need of further exploration. Previous studies identifying which of a GP’s activities are most likely to become the subject of litigation are limited in scope.54 It is equally, if not more, important to investigate how judges reason cases heard in court and to pinpoint where liability is most likely to be founded. This type of investigation provides a greater opportunity to generate guidance that can be relied upon not only to inform and improve future clinical practice, but also to help GPs avoid the threat of legal action. B. Methodology A search of Westlaw was performed to provide the data set for this study. The search was confined to those cases that were reported between the dates of 13 November 1997 and 13 November 2017. The justification for limiting the search in this way was that 13 November 1997 was the date on which the then House of Lords handed down its landmark judgment in Bolitho.55 After that case, judges were allowed to scrutinise expert medical testimony to a greater extent, instead of demonstrating excessive deference to defendant testimony, as was the case under Bolam.56 The initial search term used in Westlaw was ‘general practitioners’, which was a specific term available in a key word search. The additional search term ‘negligence’ was then introduced within those primary results. Filtering the results in this way narrowed the findings and made the data set more accurate and manageable. In the initial search, that relied solely on the term ‘general practitioners’, a significant number of cases unrelated to negligence were returned and concerned such things as employment disputes and fitness to practice proceedings. Thus, ‘negligence’ was chosen, because it was the most appropriate secondary term to pare down the original results while at the same time remaining sufficiently broad so as to encompass all the cases in which breach of duty and causation were live issues. Despite its introduction, there was still a need to sift out more irrelevant cases. At this stage, however, the data set was smaller and the final honing was performed manually during a thematic analysis. The limitations of this approach should be acknowledged. The search did not account for unreported cases against GPs nor did it capture settled claims or those which were abandoned.57 Yet, any attempt to obtain detailed, accurate information about settled and abandoned claims would be incredibly difficult, and so, the value of the data gathered here was that it provided a snapshot of the reported cases involving allegations of negligence against GPs over a defined 20-year period. Given that clear-cut cases usually settle, the majority of hits returned represented finely balanced claims in which both sides considered that there was a reasonable prospect of success. The data revealed not only the types of GP conduct that commonly gave rise to legal action, but also how those disputes were argued by both parties to the litigation and how delicately poised disputes were approached, reasoned, and determined by judges. C. Analysis Between the designated dates of 13 November 1997 and until 13 November 2017, the search revealed a total of fifty-six cases. A further manual analysis discounted a number of cases on the basis that they were not directly relevant. These included psychiatric injury claims against GPs for allegations of abuse, evidential disputes, quantum hearings, fitness to practice cases, and administrative law issues. There were also three cases that were duplicated in the findings, because they began life in the High Court and then proceeded to the Court of Appeal. These cases were counted as one hit, the relevant overall decision on liability being taken from the Court of Appeal judgment. Finally, one further case was removed from the findings on the basis that it involved criminal gross negligence manslaughter. After discounting the unrelated returns, the data set comprised thirty-five cases. These judgments were then read, and a thematic analysis was conducted with the intention of identifying recurring themes from within the transcripts. Four core themes were identified. These were: diagnosis and referral—relevant considerations; balancing of risks versus benefits; communication and information exchange; and causation: inimical to success? These themes form the focal point of the proceeding discussion. IV. Discussion A. Headline Figures While this project adopts an interpretivist approach to the data analysis, it is nonetheless useful at this stage to highlight some of the headline figures. In the thirty-five cases under investigation, claimants succeeded in establishing liability against GPs in 45.7 per cent (n = 16). Claimants lost in 54.3 per cent (n = 19) of the cases, meaning that defendant GPs were held not to be liable in just over half of the judgments. Of the cases, 82.9 per cent (n = 29) related to allegations of misdiagnoses and inappropriate subsequent referral to hospital. Of those 29 cases involving misdiagnosis and referral, claimants were successful in 48.3 per cent (n = 14) of them, whereas they lost in 51.7 per cent of the cases (n = 15). 11.4 per cent (n = 4) of the cases concerned an allegation of negligent treatment, and in those cases, claimants lost 100 per cent of the time (n = 4). Of the judgments, 5.7 per cent (n = 2) related to an allegation of negligent communication and advice, in which the claimants had a 100 per cent success rate (n = 2). Existing research conducted upon a random sample of claims contained in the Medical Protection Society and MDU databases found that 50 per cent of those claims were a result of diagnostic delay.58 Major themes identified within those claims also centred on lack of knowledge and skills, diagnostic difficulties in newborns and children, and an insufficient level of suspicion regarding signs and symptoms of rare but life-threatening diseases.59 The figures here, therefore, support some of those previous findings. That earlier study, however, did not provide an in-depth exploration of the judicial reasoning behind those cases, and it is here where this research provides a more detailed and searching critique of the reasons for, and against, liability and whether, on closer examination, they are justified. The narrative now progresses to explore the key themes identified within the analysis. B. Diagnosis and Referral—Relevant Considerations Whether or not the need for a hospital referral is a medical necessity is a difficult judgement call for a GP to make. From a lay perspective, there may be a propensity to answer this question in over-simplistic terms. That is, if there is ever any doubt in a GP’s mind, she ought to be obliged to refer the patient to hospital, and the law should hold her to account if she fails to do so. Nonetheless, this ‘better safe than sorry’ attitude overlooks the complexities of a referral decision. The difficult nature of the task was underlined by the comments of Maurice Kay LJ in Doy v Gunn,60 where he identified that often the form of an illness can be ‘very rare’ but that the history of symptoms can be ‘very common’.61 GPs will see a number of patients on any given day and, not infrequently, the symptoms complained of by those patients will share many similarities. Add to this the fact that very often the early signs and symptoms of certain serious illnesses bear close resemblance to those exhibited by more trivial conditions, then the challenges encountered by GPs in identifying those rare cases in which common place symptoms may be camouflaging a more sinister condition become readily apparent. It was, therefore, prudent of Holland J in Vance v Taylor not to impose liability on a GP after he conducted a home visit and subsequently failed to refer the patient to hospital immediately after that assessment. At the time of the home visit, it would have been almost impossible to recognise that the patient was suffering from a virulent infection with ‘features in terms of history, presentation and extent that were unusual almost to the point of being unique’.62 There is a further aggravating factor. Time constraints are inimical to a GP’s ability to diagnose with complete accuracy and precision. A typical consultation will last somewhere between 9 and 15 minutes, so there are limits to what a GP can be expected to discover in that timeframe.63 A standard GP’s surgery will not usually have at its disposal enhanced diagnostic imaging machines nor will it possess the capabilities to perform more invasive diagnostic techniques. It follows that the main way in which a GP will reach the diagnosis informing her referral decision is by assessing the presentation of symptoms, questioning the patient about the history of those symptoms and undertaking a physical examination. Any physical examination conducted will not be as precise as a specific diagnostic procedure performed in hospital and so a conclusion has to be reached on the basis of a swift consultation that has an air of generality to it. In view of these various diagnostic challenges, it is understandable that certain conditions may be missed, and it would go beyond negligence’s core concept of reasonableness for a judge to hold a GP in breach of duty every time something went undetected and a referral was not made. Nevertheless, that does not mean that the law should excuse lax conduct where any examination is patently substandard. In Dainton v Powell, Swift J was justified in her conclusion that the defendant fell below the standard of care that the law reasonably expects of a GP in failing to detect developmental dysplasia of the hip in a child.64 A more conscientious investigation should have been conducted, which would not have created any extra burden on the GP in terms of the time and effort that was needed to complete it. On the exercise of reasonable care, it ought to have been obvious that something untoward was present, which warranted urgent referral to an orthopaedic specialist.65 There are also important questions to consider about the role of experts in GP cases. Scholars such as Teff and Jones have previously recognised that one problem with the judgment in Bolitho is that it has a tendency to focus a judge’s mind on the internal consistency of a particular expert’s testimony. This detracts from considering the expert opinion in totality, set against the broader question of what may or may not have been reasonable in the circumstances.66 It is an inescapable consequence of the practicalities of litigation that judges will have to place considerable emphasis on the credibility of both factual and expert witnesses, but, in doing so, they should not allow themselves to become too heavily influenced by what appears to be the most pre-eminent expert. If this happens, they may be too quick to accept those opinions as being responsible, when objectively viewed that conclusion may appear incongruous. In GP cases, this problem is exacerbated, because the defendant and claimant experts will often agree on many things, perhaps only disputing one key point that then becomes pivotal to determining the question of breach.67 Thus, where both experts agree on the uncertainties associated with a particular presentation and identify the ensuing diagnostic difficulties that flow from it, but then only disagree as to what would have been acceptable conduct when faced with that situation, there may be an understandable inclination from judges to be more sympathetic to the defendant’s witness. Implicit in this is a recognition that, in some situations, either course of action would have been reasonable and that it was just unfortunate, but not negligent, that the conduct of the defendant resulted in harm on the occasion in question. It therefore seems evident that judges are prepared to give a certain amount of leeway to GPs when it comes to assessing their conduct in respect of diagnosis and referral. This is a sensible attitude, based partly on recognition of the medical and scientific uncertainty associated with diagnosis, and partly on an awareness of the practical constraints that GPs have to operate within when attempting to reach an accurate determination about how best to treat a patient’s condition.68 However, this amount of latitude should not be afforded to a GP in every type of referral case nor should judges be completely deterred from deeming defendant expert testimony to be irresponsible, where what is being argued for seems unreasonable. There may be some instances where a particular destructive illness is suspected, which could result in devastating consequences for the patient. If, as a result of a missed or delayed diagnosis, those consequences transpire, provided that they could have been avoided by a GP taking a simple and trouble-free precaution, then the argument for imposing liability where she fails to do so gains traction. The thematic analysis exposed a cluster of cases concerning what are best characterised as aggressive and rapidly developing illnesses. The pathology of these conditions is slightly different to that of slow progression complaints, which may remain asymptomatic until the later stages of development. Their aggressive nature often means that time is of the essence, because any delay in accurate diagnosis and treatment could result in catastrophic harm. One such condition is meningitis. Of the cases under investigation in this study, 27.7 per cent (n = 9) related to meningitis. Of those nine cases, claimants lost 55.6 per cent (n = 5) of the time and won 44.4 per cent (n = 4) of the time. While these findings demonstrate that judges were, on occasion, prepared to hold GPs liable, claimants still lost in the majority of cases. The circumstances in which claimants were unsuccessful in meningitis claims turned mainly on evidential disputes as to what actually happened in the course of a consultation and to what extent the patient actually exhibited certain symptoms at the time they were assessed.69 The issue, therefore, frequently turned on credibility, which, as noted earlier, is a crucial aspect of any case and something that judges have to play close attention to amidst the throes of litigation. Trial judges have the benefit of hearing and assessing witnesses first hand and of observing how they perform under cross-examination. Appellate court judges do not enjoy these advantages to the same extent, but they do still have access to full expert reports and the daily trial transcripts. Thus, the law reports do not always give a complete picture of the finer points of a case, and so, a degree of caution has to be exercised when attempting to critique judges’ decisions. Nonetheless, based on what is available in the reported transcripts, a number of observations can be made about the reasoning of judges in certain GP cases that is suggestive of the fact that the bigger and more holistic picture of reasonableness in the circumstances is sometimes missed.70 In Knott v Leading, the claimant alleged that the defendant was negligent in failing to refer her to hospital when she had observable symptoms that were consistent with her suffering from meningitis.71 During the relevant consultation, the patient, a child, presented with symptoms of a sore throat and earache, a temperature, and a history of vomiting in the night. Moreover, a rash was observed across the upper part of the child’s chest.72 In the opinion of the GP, this was an urticarial rash and not a meningeal rash; such was his confidence in that assessment that he did not apply pressure to the rash to check if it remained or disappeared.73 As he had witnessed this type of presentation frequently, the GP moved directly to a diagnosis of viral illness, even though it was conceded by him that he should not have completely ruled out a differential diagnosis.74 Despite its exact classification being the subject of some uncertainty, Davies J found that the rash was not meningeal in nature at the time of the consultation and held that the GP was not negligent in failing to refer the child to hospital.75 Notwithstanding that the judge may have been more persuaded by the GP’s evidence over that of the claimant’s parents as to her actual presentation at the relevant time, here was a child that had a collection of symptoms that ought to have raised at least some concern in the mind of the GP. Admittedly, the presence of any one of those symptoms alone may have been insufficient to raise the necessary level of alarm, but perhaps those symptoms combined with a rash, whatever its true nature, ought to have been enough to persuade the GP to err on the side of caution and to refer. What is required, as stated by Cox J in Large v Waldron, is a ‘careful assessment of the totality of the signs and symptoms that exist, even if each symptom individually can be regarded as non-specific’.76 It was accepted in Knott that the ‘threshold for suspicion’ in meningitis cases was a ‘low one’ and that, as ‘the consequences of a misdiagnosis can be dire’, a GP has to be confident that she can exclude it before deciding not to refer.77 Considering that meningitis is known particularly to affect children and that, in some cases, those who are very young may not always be able to communicate their symptoms other than through the voice of their parents, when dealing with a child it is not beyond the realms of reasonableness to expect a higher degree of caution from a GP in making her assessment of whether or not to refer. In line with the expert testimony provided in Coakley v Rosie, which suggested that a GP ‘is expected to suspect the diagnosis, not to diagnose’,78 judges should perhaps be more inclined to endorse a body of medical opinion that supports a suspicious and vigilant attitude. It follows that where a suspicion of meningitis can reasonably be said to exist, or indeed any other such type of aggressive condition, given the speed at which they develop, and their potentially severe consequences, there should be a greater expectation from judges of the need to refer. Zarb v Odetoyinbo concerned another destructive condition, cauda equina syndrome. The GP was held not to be liable for failing to make a same day surgical referral for a patient who had previously suffered a disc protrusion. The patient presented to him with pain that radiated down both legs, and she was experiencing severe difficulty in walking.79 As with all the cases analysed in this study, it is possible that the medical complaint would have been dealt with differently when the facts actually occurred. The incident in Zarb, for example, took place in 2001, and the advice on how best to deal with a suspected case of cauda equina syndrome may well have differed back then from the present position. Nowadays, certainly, GPs are advised to treat any suspicion of the condition extremely seriously, because time is of the utmost importance.80 It is only natural then that one has to be wary of assessing judgments with the benefit of hindsight, but even so, certain things can still be said about Zarb given the widespread acceptance, even at the time, of cauda equina syndrome being a medical emergency with potentially disastrous consequences.81 The patient exhibited symptoms that ought to have alerted the GP to the very real danger of a serious neurological complaint. Given that she had previously suffered a disc protrusion, which was known to be located in a very delicate position near the spinal cord, the presenting symptoms plus that known history should have been enough to trigger a level of suspicion that met the threshold for a same day referral.82 Given that cauda equina syndrome can often lead to paralysis and permanent loss of bladder and bowel control, the slightest warning signs should demand urgent referral, both then and now. In these aggressive high-risk conditions, it is surely open to question Tugendhat J’s declaration that it may well be prudent and sensible for a GP to make a same day referral, but that a GP who ‘trusts his own judgment’ and decides not to will not act in breach.83 Where simple steps can be taken to avoid a catastrophic degree of harm, it should not be beyond the wit of a judge to hold that any course of action that fails to take those steps, and any testimony in support of it, is unreasonable in the circumstances and, therefore, negligent. The danger here is that this line of argument could be interpreted to suggest that the cautious approach ought to be treated as the ‘gold’ standard in every case and that a GP should automatically be held in breach for adopting any other course of action. That is not the argument; the point is that a GP, who does not typically participate in intricate and risky medical procedures and who, therefore, is not in an analogous position to that of many of her secondary care colleagues, is sometimes in a position where she can avoid catastrophic harm being caused to a patient by taking a straightforward precaution. Where that precaution is not taken, a judge ought to be more willing to make a finding of breach against her. C. Balancing of Risks versus Benefits The findings revealed that, in certain situations, judges were actually prepared to depart from expert medical testimony led in support of a defendant’s conduct.84 They appeared more inclined to do this though, where the issue focused on referral, advice, and communication, evidenced by the fact that claimants never succeeded in treatment cases. In line with its characterisation as a ‘gloss’ on Bolam, Bolitho was only mentioned in passing in the majority of cases and a line of reasoning that centred solely on the judgment itself was never really invoked. Nevertheless, there were some cases in which the balancing of risks and benefits did feature.85 The Bolitho balancing exercise is a useful arrow in a judge’s quiver because, where there is some degree of risk attendant upon a particular course of action, it is useful to offset that risk against any potential benefits in order to gauge whether the riskier approach was in fact responsible. On the balance point, judges have been seen to hold GPs liable for failing to take basic precautions, notwithstanding that there was a body of medical opinion in support of what the GP did.86 This is because where a very obvious precaution could have been taken, recourse to Bolitho is not necessary as any assessment of the risk versus benefit ratio would be so one sided that any ‘balancing’ of the two would be superfluous. Hence, the argument in Langdon v Williams that it was acceptable not to rouse a sleeping baby in order to assess accurately its level of responsiveness because ‘it is an instinct if you see a baby asleep to leave it’, was clearly unsatisfactory.87 This typifies a situation in which a patient was exposed to an unreasonable risk without any discernible benefit, where that risk could easily have been avoided. Where, however, the issue relates to more complex treatment decisions, any balancing exercise becomes more challenging for judges to undertake.88 It is in these situations where they may be more hesitant to interpose their own views over those held by preeminent experts. This is understandable to a degree, but it should not always be used as an excuse for abdicating any objective assessment of the evidence in its entirety. Take, for example, Maguire v North West Strategic Health Authority.89 The case is complicated but, in essence, involved a GP and a hospital effectively blaming each other for failing to detect the recurrence of a leak of cerebrospinal fluid from the patient’s nose, which ultimately caused him to suffer brain damage.90 One of the key questions that fell to be decided was whether, if the patient had been offered an earlier follow-up appointment after his initial discharge from the hospital, the consultant should have performed what is known as a ‘tilt test’ in order to identify any recurrence of the leak at that earlier stage.91 Despite this component of the case not focusing on the GP’s conduct, it neatly illustrates some of the problems faced by judges in the realms of assessing treatment decisions. Undoubtedly, there would have been some benefits to undertaking a tilt test in that it could have exposed the underlying problem earlier, meaning that there would never have been any need for the GP to be called into action at a later date, but there were also very clear risks attached to that course of action. If there was immature healing of the fistula from which the leak initially occurred, a tilt test could have caused increased hydraulic pressure, which would have disrupted the healing process.92 In respect of the risk versus benefit ratio, while the test itself may have been simple and non-traumatic insofar as its execution was concerned, the risk that could flow from it was potentially extremely severe. In view of this, it would have been very difficult to justifiably classify any medical testimony in support of not performing the test as being irresponsible.93 In contrast, judges ought not to be too reticent to intervene in relation to every treatment issue. In Thompson v Bradford, the main allegation of negligence centred on the fact that a GP failed to recognise that a perianal abscess in a small child was symptomatic of an acute illness rather than a localised infection.94 He then reassured the child’s parents that there was no reason why an immunisation against Polio should not go ahead, which was subsequently administered. Shortly afterwards, the abscess had to be lanced under general anaesthetic, and, as a consequence of that operation, the patient was left with an open wound. The Polio vaccine, as a live virus, then infiltrated the wound and tragically the child contracted the very condition that the vaccine sought to guard against.95 It was accepted that a recurrent and perianal abscess is an unusual presentation in general practice and that it may well have required surgery in the future.96 Arguably, these symptoms alone ought to have prompted a more careful thought pattern from the GP as to whether something more sinister was occurring. Irrespective of that, whatever the true diagnosis, it is submitted that caution ought to have prevailed. There were clearly benefits to postponing the vaccination to assess the true nature of the illness, which would not have exposed the child to any increased risk, save that the vaccination would have been slightly delayed.97 The patient was, therefore, exposed to a disproportionate and unreasonable risk; he was subjected to a live virus when the doctor ought to have foreseen the possibility of a future open wound, when compared with the much more discernible benefit of postponement, which ultimately would have avoided the consequences of that severe and unnecessary risk. This is the paradigm example of a situation in which the judge should have engaged with the expert opinion to a greater extent, perhaps questioning why the defendant’s expert endorsed behaviour which, despite being technically in accordance with Green Book Guidelines on common immunisation practice at the time, ignored any consideration of the need to proceed with extra diligence in the light of the especial peculiarities of the child’s condition.98 Remaining sensitive to a unique condition in a young child, and thereby suggesting a delay in the vaccination, surely portrays the expert testimony presented on behalf of the claimant in a more respectable and reasonable light.99 While it may be more challenging for a judge to weigh the risks and benefits where the dispute involves medical questions concerning appropriate treatment, it may not necessarily be as difficult where the focus is on referral. These issues sometimes demand less in the way of technical medical expertise, and the risk versus benefit ratios are not as finely balanced. Indeed, the findings here are indicative of varying attitudes from judges in regard to the balancing of risks and benefits in a patient’s favour in referral cases. Foskett J in Ganz v Childs suggested that where there was ‘some’ chance that a child may be suffering from pneumonia, it would be imperative to discover that quickly.100 As such, there could be no logical basis for delaying the relevant hospital tests. Where a child was concerned, his approach to the balancing exercise was that if the tests ‘proved negative, nothing would be lost and pneumonia would have been excluded. If the tests were positive, suitable treatment could be put in place quickly.’101 Given that children are perhaps more susceptible to certain conditions, and that they may not be able to communicate their symptoms as clearly as adults, a low threshold for suspicion is necessary. An additional layer of caution was therefore recognised by Foskett J as being essential when dealing with children. Nonetheless, it is not entirely fair of him to suggest that ‘nothing would be lost’ in terms of mandating a referral. This gives the impression that the balancing exercise is wholly one-sided, which is a misnomer. Referrals take time, cost money and, in some instances, will expose the patient to unnecessary levels of discomfort. There are also questions concerning allocation of resources. This was identified by Leveson LJ in Ministry of Justice v Carter, where he pointed out that specialist referrals have ‘implications in relation to the pressure of work on this service’.102 At various junctures in the judgment, guidance from professional regulatory bodies was also alluded to, which actively discouraged the practice of routinely over-investigating or making inappropriate referrals.103 Some of those guidelines have since been updated and now focus more on when to refer, rather than when not to.104 However, an interesting question is nonetheless raised about whether or not there is a hidden agenda behind actively discouraging referrals. In Carter, there was no suggestion of a financial reward for the doctor in not referring the patient, yet, a more recent article in the Guardian claimed that some GP practices were offered financial rewards for not referring patients to hospital,105 which is worrying. It is one thing to encourage a GP to rely on her own clinical judgement and to back her own diagnosis after a careful and robust examination of a patient, and it is quite another to suggest that a culture of referrals should be suppressed because of the additional cost and strain it may be placing on the National Health Service (NHS). Any incentives not to refer, financial or otherwise, must form part of a holistic balancing process in the mind of a GP when deciding whether or not to refer, and this approach should equally form an integral part of a judge’s reasoning when determining the legal question of breach. The balancing exercise should include not only consideration of the presentation of symptoms but also the severity and consequences of any suspected illness, any previous complaints suffered by the patient, the range of alternative diagnoses, and any characteristics particular to the patient that may increase the risk factor of any suspected condition.106 Resource questions are, of course, relevant factors that should also be accounted for in any balancing process and, in some instances, they may even prove to be the prevailing consideration. Yet, concerns over resources should not be allowed to dictate entirely the decision-making process in respect of referrals, especially where there is a combination of compelling factors, which are indicative of the patient’s health being compromised unreasonably by any indecision on the part of a GP. D. Communication and Information Exchange The theme of communication and information exchange permeated throughout many of the cases in this study.107 The importance of good communication between doctors and patients has been frequently identified.108 A breakdown in communication is often perceived to be a key factor that leads to complaints, which in turn may then escalate into formal legal action.109 Honing communication skills is, therefore, seen as a way of reducing complaints and resultant legal action.110 The findings illuminate some rather basic errors in communication, which led to GPs being held liable. In Cutting v Islam, for instance, the failure to communicate effectively a simple warning to the patient to return to the surgery if his rectal bleeding persisted caused the GP to be held liable.111 In conjunction with rudimentary communication errors, patient questioning was also a prevalent issue. It would seem that GPs sometimes fall into the trap of communicating in a one-dimensional manner. That is, a GP communicates with a view to reaching an accurate diagnosis in as short a time as possible, becoming over reliant on standard questions in order to rule out certain conditions and to identify others. Given the time pressures, this practice is perhaps understandable, but approaching patient questioning in this way injects a degree of objectivity to the interaction, and it is here where things can sometimes be missed. It was highlighted by Cox J in Large v Waldron that ‘some people are naturally more confident and assertive than others and that they will require little prompting in a doctor’s surgery. Others are less forthright and require more careful questioning to enable them to articulate and to elaborate upon the important information that the doctor requires.’112 It is essential then that a GP tailors her communication to the wants and needs of the individual patient, for questioning each patient in the same way, in terms of both style and substance, may lead to important diagnostic information being omitted in any conversation. The need for strong communication skills is enhanced where children are involved; very young children may not be able to speak at all, and even those who may be slightly older, may still struggle to convey the true nature of their symptoms. Accordingly, a GP must glean whatever she can from a child, but any gaps must be filled in by the parents. On this point, a worrying message was conveyed by HHJ Cotter QC in Maguire.113 For him, a common observation in litigation against GPs was ‘the patient’s parents having a real and reasonable concern that there may be a serious problem and being wrongly assured that little was amiss after a clearly inadequate assessment by a visiting GP’.114 This is where it becomes especially important to view communication as something more than just a one-way conversation. A GP must listen as well as question and be encouraged, wherever possible, to initiate a dialogue with patients and/or parents.115 She needs to recognise that she only sees a ‘brief snapshot’ of what is happening, as opposed to her patient, or her patient’s parents, who get to see ‘the bigger picture’.116 A GP should thus adopt an attitude that encourages a more rounded conversation between herself and her patient, because taking the time to explore any concerns in greater depth may reduce the potential for diagnostic errors and consequential legal action.117 Communication must also be viewed as a multifaceted concept. It is not just about a GP conversing with her patient; she must also communicate with her colleagues. The cases unveiled a frequent trend of patients being seen by a number of different GPs on a number of different occasions,118 and trouble stemmed from a lack of adequate consideration from one GP as to what had occurred previously in a consultation with another colleague.119 There was scarce evidence of joined-up thinking, and symptoms were often considered in isolation instead of any real thought being given to the history behind the current presentation and to any advice and treatment that had been provided at an earlier stage. Verbal communication, however, should not be the sole focus; written communication is of equal importance. In many of the claims against GPs, the contemporaneous notes made by the doctor played a big part in resolving factual and evidential disputes,120 yet it was sometimes acknowledged that those notes were either incomplete or perhaps not as detailed as they should have been.121 Given that so much will hinge on the quality of the notes should a case against a GP ever proceed to court, the advice has to be to make them as thorough and meticulous as possible. The main aim of the notes is to chart a comprehensive history, which can then be read by other medical colleagues; they must appear in an accessible and decipherable format in order to avoid the problem of a GP not adequately considering what has gone before when attempting to reach an accurate diagnosis. From a legal perspective, the need for comprehensive notes is pivotal as having a detailed record of what transpired during the course of a consultation may provide a stronger defence for a GP against a lawsuit. It stands to reason then that the importance of good practice in note taking should be heavily underlined in primary care. In respect of information disclosure, the true impact of the Supreme Court judgment in Montgomery v Lanarkshire on GPs remains to be seen.122 It has been suggested that the primary care relationship between GPs and patients ‘may not lend itself as easily to the Montgomery model of risk disclosure as a GP’s knowledge … may not be as thorough or detailed as one would expect of a specialist’.123 GPs may be concerned that they do not have ‘access to the full range of information or tools to enable them to discharge the standard of duty that Montgomery would seem to require’.124 While it may be unfair to expect the same level of disclosure of a GP than of a consultant specialist involved in intricate and high-risk surgery, it is not unreasonable to expect the former to disclose certain types of information to patients. There was no case that focused exclusively on risk disclosure, although the earlier mentioned decision in Thompson v Bradford is tangentially relevant.125 In addition to the allegation of negligent treatment, there was also an ancillary allegation of negligent disclosure. In the High Court, Wilkie J held the GP liable for failing to inform the child’s parents that a recurrent perianal abscess was unique in that particular GP’s experience, was extremely unusual, and that there was the very real prospect of the child having to undergo surgery in the near future should the problem not resolve after a course of antibiotics.126 The Court of Appeal later overruled this decision.127 This is exactly the type of case in which Montgomery should bite. The fact that it may not have been foreseeable to the GP that there was a specific risk of contracting Polio from the very vaccine that sought to guard against it, and that, therefore, he was under no duty to disclose it, should have been no excuse for his failure to divulge the other more general risks that may have equally operated on the minds of the child’s parents when deciding whether or not to proceed with the immunisation.128 Had they been informed of the illness’s highly unusual nature, and of the potential need for future surgery with the associated risk of an open wound, they may have been likely to postpone the vaccination until the child was in a more stable condition. While not specific to the risk that eventually materialised, these were still crucial pieces of information that the GP omitted to inform the parents about, which would have been likely to influence their decision. Arguably, the reasoning of the Court of Appeal in Thompson would not survive the ruling in Montgomery, as this was significant information that would have affected the decision-making process of most reasonable parents, and, if not, almost certainly those particular parents.129 Mindful of the modern developments in primary care, especially in the light of minor surgical procedures now being performed in primary care settings, the advice to a GP has to be to keep a watchful eye on the developing post-Montgomery case law and to remain diligent in terms of what she discloses to patients.130 E. Causation: Inimical to Success? The findings do not lend support to the view that causation is a troublesome obstacle to patients in GP cases. Where patients lost, the majority of claims actually failed because the patient was unable to prove breach. The question of causation, therefore, became academic. Nonetheless, judges would often proceed to speculate as to what would have been the situation had it been a live issue and, where breach was not established, they usually indicated that causation would also have failed.131 What become clear was that causation was not an impediment to success in that it did not operate to defeat claims that had succeeded in proving breach; once a patient established breach, a favourable finding of causation usually followed. In fact, where breach was proven, there was only one instance in which a claim subsequently failed on a point of causation.132 Where the question centred on the lack of, or late, referral, judges frequently endorsed medical testimony in support of the fact that earlier treatment would have avoided the harm that eventuated.133 Reliance by patients on the reasoning in Wright was thus never required in order to succeed. The judgment was actually only raised once, and by a GP not a patient, in an attempt by the former to argue that the chain of causation had been broken by the subsequent negligence of a hospital.134 However, as the question of breach had already been decided in favour of the GP, Jay J did not analyse the submission to any great extent, save for suggesting that, irrespective of the applicability of Wright, he would have found that the GP’s negligence was an effective cause of the patient’s injury had he been required to rule on it.135 From a different perspective, where a claimant’s primary argument in respect of causation was disputed, HHJ Oliver-Jones QC stated in Coakley v Rosie that, even if her ‘but for’ argument did not succeed, he would still have found for her on her alternative argument, which was based on the ruling of the Court of Appeal in Bailey v Ministry of Defence.136 Here, it was held that where cumulative causes of harm exist, and where the current state of scientific knowledge precludes a finding that the negligent act, as one of those causes, was in fact the most probable cause of the harm if a claimant can still prove that the contribution of the negligence was more than negligible, then the ‘but for’ test should be modified and her case ought to succeed.137 Given that Bailey was a case that involved hospital treatment, its apparent endorsement in respect of primary care negligence is curious. This is especially so in the light of it representing a modification to the conventional ‘but for test’ that is not without critics.138 Irrespective of that, it would seem even though some patients may encounter difficulty in establishing conventional ‘but for’ causation in a referral case against a GP, where scientific uncertainty exists, it may still be possible for them to argue that the medical condition itself, plus any delay in treating it based on an inadequate referral, represent cumulative causes of any harm. Provided it can be said that the contribution of the negligent referral was more than trivial, then judges may be willing to treat such arguments with a sympathetic eye. A further argument was also illumined. It was identified that, in certain circumstances, an urgent referral from a GP would lead to a patient being ‘fast-tracked’ through the hospital admissions system.139 It was pointed out by Jay J in FB v Rana that ‘there is a material difference between a child arriving at A&E, even by ambulance, because parents are concerned that she might require urgent attention, and a child arriving there … because a trained medical professional, a GP, harbours concern’.140 This is effectively saying that, where a patient is specifically referred by a GP, she is more likely to receive swifter treatment from a hospital and, thus, enjoy a greater chance of any illness being brought under control than she would have experienced had she arrived at the hospital by other means. This improves the chances of a patient being able to establish that a lack of referral was a key causative factor, but it is a slightly skewed line of reasoning. It is perhaps plausible that a patient who presents to hospital of her own volition may not be taken as seriously as she would have been had she arrived on the back of a GP referral. However, it is hoped that any obviously sick patient, especially a child arriving with concerned parents, and certainly one that arrives by ambulance, would be given swift attention by hospital staff regardless of a lack of referral. The urgency with which a patient is treated by a hospital should not be wholly dependent upon how they arrive there and under whose instructions they were told to attend. To allow a hypothetical inquiry such as this to influence causal calculations is potentially dangerous because how a patient may have been treated is a question fraught with conjecture. Thus, in Fallon v Wilson, Eady J acknowledged the ‘fast track’ argument, but treated it with hesitancy given that it demanded a degree of speculation.141 Caution is definitely sensible, yet, notwithstanding that, it does seem that judges may be receptive to this line of reasoning. If it holds sway in the future, it is something else that could be advantageous to patients in the domain of causation. The judgment of the then House of Lords in Gregg v Scott could, at first blush, be construed as being detrimental to a claimant’s prospects of recovery where there is a slower, yet more sinister, condition that goes undetected by a GP.142 Cancer is one such archetypal condition. While Gregg v Scott effectively precludes a patient from recovering damages for having lost the chance of a cure, it became apparent that this ruling did not act as a complete bar to patients recovering at least something in missed cancer and related claims. Despite judges remaining resolute in denying loss of chance arguments, they would sometimes manoeuvre their way around that position in order to grant a measure of damages that ensured that deserving claimants were not left wholly uncompensated. Judges approved an aspect of Lady Hale’s judgment in Gregg v Scott, in which she endorsed an alternative to the pure loss of chance argument. She suggested that, despite recovery not being possible for loss of chance, where the median life expectancy of a patient may have fallen due to any delay in treatment, there may be a ‘modest claim in respect of the “lost years”’.143 This effectively allows a judge to perform a calculation based on the median life expectancy of a patient whose complaint was identified and dealt with at the correct time.144 Where there is a negligent delay in treatment, a patient’s median life expectancy may be reduced by the amount of time of the delay. Bean J in JD v Mather deduced that the patient’s prospects were worsened not only in the sense of a reduced chance of survival beyond 10 years, but also in the sense of a reduced expectation of life.145 He therefore decided that, measured against the median, the patient’s life expectancy was reduced by 3 years.146 In reaching this conclusion, a crucial factor seemed to be that the negligent delay resulted in a changed staging of the melanoma, and good quality statistical information was available to show the impact of this on median life expectancy.147 However, where that statistical information is not of the same quality, and perhaps not as convincing as to the impact of any negligent delay on median life expectancy,148 it appears claimants’ prospects of success in relation to this line of argument may be diminished.149 For some, the continued failure of the courts to recognise loss of chance may still seem illogical and unsatisfactory.150 It is beyond the scope of this article to explore the merits and pitfalls of allowing such an approach,151 yet, for those who regard the current legal position as unduly restrictive, some comfort can be taken from the fact that there are at least some ways to circumvent the harsher effects of the current bar to loss of chance recovery. VI. Conclusions This study provides an original insight into the judicial reasoning that has underpinned negligence claims concerning GPs over a 20-year period. The research feeds into the broader conceptual debate about the changing nature of the doctor–patient relationship and explores judicial appetites for forensically scrutinising medical decision-making. It also provides useful practical guidance to help avoid complaints and legal redress. The findings confirm that the majority of claims that make it as far as court centre on the issue of misdiagnosis and inadequate referral. While claimants tended to lose in the majority of these, there was still some evidence of judges being prepared to engage with the medical testimony to a greater extent and, on occasion, rely on their own objective assessment of the question of reasonableness where very obvious poor practice was demonstrated, or where there was a disproportionate risk to benefit ratio. In situations where a patient was exposed to a potentially severe risk, which could have been avoided with a relatively straightforward precaution, judges showed at least some willingness to hold GPs responsible. Judges did seem to recognise that the question of whether or not to refer is complex and should not be resolved by recourse to an overcautious approach in every circumstance. What is needed is a rounded balancing of the risk versus benefits of a particular treatment decision in order to reach a defensible conclusion as to its logicality and reasonableness. In some instances, this exercise was conducted by judges in the way that it should have been, but, in others, it was not. Certainly, in the future, judges ought to be more inclined to hold a GP liable in negligence for behaviour which exposes the patient to a risk that could easily have been avoided by taking a simple precaution. The study has also demonstrated that, while causation is indeed a difficult obstacle to overcome in some cases, it is not necessarily an impediment to success where a breach of duty was established. A GP and her defenders should not be lulled into a false sense of security in thinking that, even where breach is proven, the majority of claims will still nonetheless be defeated on causation. Communication and information exchange were also important themes in many of the cases.152 A GP should not approach communication in a blinkered fashion, with an overdependence on standard questions to elicit a speedy and accurate diagnosis. She should enter into a dialogue with her patient in order to determine any wider factors that may be relevant to the diagnosis. In order to achieve this, it is imperative that communication is approached in a subjective manner, with questioning being tailored to the patient as an individual. In the rare cases that do make it as far as court, being able to demonstrate strong communication skills between not only the GP and the patient, but also between GPs themselves, may go some way towards reducing liability. Similarly, taking the time to keep detailed notes will usually work to the advantage of a GP should any complaint become the subject of subsequent legal action. Faced with increasing pressures, both in terms of time and resources, any suggestion to spend more time conversing with patients, and on writing comprehensive notes, may not be welcomed by those in practice. Nonetheless, in order to avoid potential future law suits, these recommendations ought not to be ignored.153 At the very least, it should be acknowledged that allocation of resource questions, while pertinent to the issue of referral, should not be the driving factor mitigating against such a decision. Concerns about a patient’s health should remain the paramount concern for a GP in her decision-making processes and, while the law of negligence should not necessarily treat the over-cautious approach as the gold-standard, she may well be advised to exercise vigilance if there is ever any doubt in her mind as to the need to refer a patient. ACKNOWLEDGEMENTS My sincere thanks are due to Professor Margot Brazier, of the University of Manchester, for her constructive comments and feedback on an earlier draft of this piece. My thanks are also due to both anonymous reviewers for their helpful suggestions. Conflict of interest statement. None declared. Footnotes 1 Beccy Baird and others, Understanding Pressures in General Practice (King’s Fund, London 2016). 2 Ruth Robertson and others, Clinical Commissioning: GPs in Charge (King’s Fund, London 2016); Jonathan Botting and others, ‘Safety of Community-Based Minor Surgery Performed by GPs: An Audit in Different Settings’ (2016) 66 British Journal of General Practice e323. 3 GPs also face a number of mental health and social issues, which account for a significant amount of their time. These issues fall outside the scope of this article. 4 Aneez Esmail and others, Case Studies in Litigation: Claims Reviews in Four Specialties (Manchester Centre for Healthcare Management, University of Manchester 2004). 5 Kevin Williams, ‘Doctors as Good Samaritans: Some Empirical Evidence Concerning Emergency Medical Treatment in Britain’ (2003) 30 Journal of Law and Society 258. 6 A Freedom of Information Request was submitted by the author to the National Health Service Litigation Authority (NHSLA, now known as NHS Resolution) on 16 October 2017, requesting data on claims made against GPs over the last 6 years. The NHSLA’s response, sent on 2 November 2017, provided data on all the claims made against GPs over the last 6 years since its CNST Scheme began in 1995. The data reflected the cumulative number of claims made during that time period. Thus, in 2012, the total number of claims made against GPs since 1995 was 363. In 2013, the figure was 446. In 2014, the figure was 524. In 2015, the figure was 550. In 2016, the figure was 672. Thus, as of 31 March 2017, the total number of claims made against GPs since 1995 is 727. The most significant increase in claims was between 2015 and 2016. It is difficult to discern from the data above why this may have been the case. One possible reason is the changes made to the Clinical Negligence Pre-Action Protocol, which came into force in April 2015. Equally, the continued impact of one-way costs-shifting, introduced in April 2013 for personal injury (including clinical negligence) claims, may have affected the figures. However, it is impossible to tell from the data provided. 7 Baird and others (n 1). 8 Rob Heywood and José Miola, ‘The Changing Face of Pre-Operative Medical Disclosure: Placing the Patient at the Heart of the Matter’ (2017) 133 Law Quarterly Review 296; Harvey Teff, Reasonable Care: Legal Perspectives on the Doctor Patient Relationship (OUP 1996). 9 Charles Foster, ‘Autonomy in the Medico-Legal Courtroom: A Principle Fit for Purpose?’ (2014) 22 Medical Law Review 48; Brooke LJ, ‘Patients, Doctors and the Law (1963–2003): A Few Reflections’ (2004) 72 Medico-Legal Journal 17. 10 Sarah Devaney and Søren Holm, ‘The Transmutation of Deference in Medicine: An Ethico-Legal Perspective’ (2018) 26 Medical Law Review 202; Raymond Tallis, Hippocratic Oaths: Medicine and its Discontents (Atlantic 2004); The Right Honourable the Lord Woolf, ‘Are the Courts Excessively Deferential to the Medical Profession?’ (2001) 9 Medical Law Review 1. 11 Bolam v Friern Hospital Management Committee [1957] 1 WLR 582. See James Louis Montrose, ‘Is Negligence an Ethical or Sociological Concept?’ (1958) 21 Modern Law Review 259. 12 ibid (McNair J) 587. 13 ibid. 14 Maynard v West Midlands Regional Health Authority [1984] 1 WLR 634. 15 Harvey Teff, ‘The Standard of Care in Clinical Negligence—Moving on from Bolam’ (1998) 18 Oxford Journal of Legal Studies 473, 481. 16 Michael Jones, ‘The Bolam Test and the Reasonable Expert’ (1999) 7 Tort Law Review 226. 17 Hucks v Cole (1968) [1993] 4 Med LR 393 at 397. It should be noted that this case was actually decided in 1968, and only reported later in 1993. 18 Bolitho v City & Hackney Health Authority [1998] AC 232. See Andrew Grubb, ‘Negligence: Causation and Bolam’ (1998) 6 Medical Law Review 378. It should be noted that Lord Browne-Wilkinson restricted his reasoning only to matters of diagnosis and treatment. Disclosure of risks fell outside the scope of the judgment. 19 Margaret Brazier and José Miola, ‘Bye Bolam: A Medical Litigation Revolution?’ (2000) 8 Medical Law Review 85; John Keown, ‘Reining in the Bolam Test’ (1998) 57 Cambridge Law Journal 248. 20 Bolitho (n 18) (Lord Browne-Wilkinson) 243. 21 Grubb (n 18). 22 Rob Heywood, ‘Litigating Labour: Condoning Unreasonable Risk Taking in Childbirth?’ (2015) 44 Common Law World Review 28; Rachael Mulheron, ‘Trumping Bolam: A Critical Legal Analysis of Bolitho’s “Gloss”’ (2010) 69 Cambridge Law Journal 609; Rob Heywood, ‘The Logic of Bolitho’ (2006) 22 Journal of Professional Negligence 225. 23 See Marriott v West Midlands Regional Health Authority [1999] Lloyd’s Rep Med 23. 24 [2015] UKSC 11; [2015] AC 1430. 25 ibid [87]–[88]. See Emma Cave, ‘The Ill-informed: Consent to Medical Treatment and the Therapeutic Exception’ (2017) 46 Common Law World Review 140. 26 ibid (Lords Kerr and Reed) [75] and [81]–[83]. 27 Heywood and Miola (n 8). 28 Judy M Laing, ‘Delivering Informed Consent post-Montgomery: Implications for Medical Practice and Professionalism’ (2017) 33 Journal of Professional Negligence 128, 142–52. 29 Tallis (n 10). 30 Kay's Tutor v Ayrshire and Arran Health Board [1987] 2 All ER 417. 31 Maguire v Northwest Strategic Health Authority [2012] EWHC 3272 at [18]–[19]. Similar arguments have also been made in respect of pharmacists, who also should have noticed errors on the part of a GP. See Dwyer v Roderick (1983) 127 SJ 806; Prendergast v Sam and Dee Ltd [1990] 1 Med LR 36. 32 [2011] EWCA Civ 669; [2013] QB 312. 33 ibid [37]. 34 ibid [61]. 35 Nicholas J McBride and Sandy Steel, ‘Suing for the Loss of a Right to Sue: Why Wright is Wrong’ (2012) 28 Journal of Professional Negligence 27; Kumaralingam Amirthalingam, ‘Causation and the Medical Duty to Refer’ (2012) 128 Law Quarterly Review 208. 36 Baxter and Others v McCann [2010] EWHC 1330; [2010] Med LR 473; A v Enskat [2003] EWHC 1630. 37 [2005] UKHL 2; [2005] 2 AC 176 at [5]. See Emmanuel Voyiakis, ‘The Great Illusion: Tort Law and Exposure to Danger of Physical Harm (2009) 72 Modern Law Review 909. 38 ibid. 39 Lords Nicholls and Hope dissented. 40 Wright (n 32) at [84]. 41 National Health Service Litigation Authority, Factsheet 3 (August 2017). Since 1995, the total number of claims made against GPs currently stands at 727. By comparison, in the same time period, the number of surgical claims stands at 50,432, and the number of obstetrics and gynaecology claims stands at 23,415. 42 Margaret Brazier and Emma Cave, Medicine, Patients and the Law (6th edn, Manchester University Press 2016) 220. 43 Denis Campbell and Sarah Boseley, ‘GP Negligence Claims: Alarming Rise Revealed’ The Guardian (London, 29 July 2011) <https://www.theguardian.com/society/2011/jul/29/gps-missing-serious-illnesses> accessed 2 October 2017. 44 See <https://www.gmc-uk.org/concerns/hearings_and_decisions/an_introduction_to_adjudication.asp> accessed 2 October 2017. 45 Campbell and Boseley (n 43). 46 Dwyer and Prendergast (n 31). 47 Nick Goodwin and others, Improving the Quality of Care in General Practice (King’s Fund, London 2011) 46. See also Campbell and Boseley (n 43); and Esmail and others (n 4). 48 See n 44. 49 Heywood and Miola (n 8). 50 Goodwin and others (n 47); Baird and others (n 1). 51 See n 6. 52 Campbell and Boseley (n 43). Medical defence organisations, such as the Medical Defence Union and the Medical Protection Society, provide indemnity for GP members against clinical negligence claims. They are not, however, insurers. 53 Brazier and Cave (n 42) 219. 54 Esmail and others (n 4). 55 Bolitho (n 18). 56 Bolam (n 11). 57 Some information on abandoned claims can be found in the Medical Protection Society and Medical Defence Union annual reports <https://www.medicalprotection.org/uk/resources/case-reports> (accessed 2 October 2017). It should also be noted that the coverage of case law across different legal databases does not always match, and the search in this study did not take account of cases which may have appeared on either LexisNexis or Bailii. Prison doctors, who fulfil the role of GP for prisoners, were included in the scope of the study. 58 Goodwin and others (n 47) 46; Esmail and others (n 4). 59 ibid. 60 [2013] EWCA Civ 547; [2013] Med LR 327. 61 ibid [27]. 62 [2007] EWHC 1602; [2007] LS Law Med 676, [39]. 63 Baird and others (n 1) 30; Health and Social Care Information Centre (HSCIC), 2006/07 UK General Practice Workload Survey (Health & Social Care Information Centre 2007). <https://digital.nhs.uk/catalogue/PUB01028> (accessed 2 October 2017). 64 [2011] EWHC 219. 65 ibid [117]. 66 Teff (n 15); Jones (n 16). 67 See Coakley v Rosie [2014] EWHC 1790, [60]–[61]. 68 Enskat (n 36) [161]–[162]. 69 FB v Rana [2015] EWHC 1536; W v Johnson [2015] EWHC 276; Doy (n 60); J v Hillard and Others, 24 May 2001, Unreported. 70 Teff (n 15). 71 [2010] EWHC 1827. 72 ibid [34]–[43]. 73 ibid [34]. 74 ibid [41]. 75 ibid [90] and [94]–[95]. 76 [2008] EWHC 1937, [112]. 77 ibid [43]. 78 Coakley (n 67) [62]. 79 [2006] EWHC 2880; (2007) 93 BMLR 166. 80 Alan Gardner and others, ‘Cauda Equina Syndrome: A Review of the Current Clinical and Medico-Legal Position (2011) 20 European Spine Journal 690. 81 Zarb (n 79) [15]. 82 ibid [3]. 83 ibid [111]. 84 See Large (n 76); Coakely (n 67); and Ganz v Childs [2011] EWHC 13; [2011] Med LR 113. 85 Mulheron; Heywood (n 22). 86 Marriott (n 23). 87 [2008] EWHC 741, Conclusions, [xiv]. 88 It should be noted that the number of treatment cases in this study was statistically very small. 89 Maguire (n 31). 90 Even though the patient eventually contracted meningitis from the original injury, for the purposes of analysis, this case was not categorised as a pure meningitis case. 91 Maguire (n 31) [30]. The tilt test is a relatively simple and non-invasive diagnostic procedure, the aim of which is to assess whether or not any fistula has healed, or if there is still evidence of a recurrent leak when the patient experiences downward pressure. 92 ibid [118]. 93 Indeed, HHJ Cotter QC was satisfied that a responsible body of clinicians would not have undertaken a tilt test, [116]. 94 [2004] EWHC 2424. For the purposes of analysis, this was classified as a treatment case, which the claimant lost. A subsequent allegation of negligent disclosure was heard before the Court of Appeal, discussed in more detail below. 95 ibid [18]–[20]. 96 ibid [17]. 97 ibid [22]. The delay would not have placed the child in any immediate danger. 98 The 1996 edition of the ‘Green Book’ was identified by both experts as the definitive work that a reasonably competent GP should apply. See Thompson (n 94) [23]. 99 The claimant in Thompson eventually lost, which the Court of Appeal later affirmed. See [2005] EWCA Civ 1439; [2006] Lloyd’s Rep Med 95. For an interesting criminal law example of a GP case in which a judge engaged with the expert opinion in a thorough manner, which effectively led to the Crown being hoist by its own petard, see R v Rudling [2016] EWCA Crim 741. 100 Ganz (n 84). Emphasis in the original judgment, [101]. 101 ibid. 102 [2010] EWCA Civ 694, [23]. 103 ibid [14]–[18]. The defendant relied upon the then NICE 2005 Guidelines on Diagnosis of Breast Cancer and the Kent & Medway Cancer Network Guidelines. 104 See NICE Guideline (NG 12), ‘Suspected Cancer: Recognition and Referral’, June 2015, [1.4] and [1.16]. 105 Denis Campbell and others, ‘GP Practices “Offered Rewards” for Not Referring Patients to Hospitals’ (The Guardian, London, 1 October 2015) <https://www.theguardian.com/society/2015/oct/01/gp-practices-offered-rewards-for-not-referring-patients-to-hospitals> accessed 2 October 2017. Also, Robertson (n 2). 106 For an example, see the judgment of Gross J in Mellor v Sheffield Teaching Hospitals NHS Trust and Others [2004] EWHC 780 at [228] (1) (iii) (c). 107 Communication errors, and the need for good communication, is not a new problem manifested solely in this study. See, for example, Judith Laing and Jean McHale (eds), Principles of Medical Law (4th edn, OUP 2017) at [4.83]; GMC, Good Medical Practice (GMC 2013) 13–18. 108 Debra Roter and Judith A Hall, Doctors Talking with Patients/Patients Talking with Doctors: Improving Communication in Medical Visits (2nd edn, Praeger 2006); LML Ong and others, ‘Doctor-Patient Communication: A Review of the Literature’ (1995) 40 Social Science and Medicine 903. 109 Jimmy Nguyen, ‘Communication-Related Allegations Against Physicians Caring for Premature Infants (2017) 37 Journal of Perinatology 1148; Joseph Domino and others, ‘Lack of Physician-Patient Communication as a Key Factor Associated with Malpractice Litigation in Neonatal Brachial Plexus Palsy’ (2014) 13 Journal of Neurosurgery: Pediatrics 238. 110 Karen L Posnerand others, ‘The Role of Informed Consent in Patient Complaints: Reducing Hidden Health System Costs and Improving Patient Engagement Through Shared Decision Making’ (2015) 35 Journal of Healthcare Risk Management 38; Steven R Eastaugh, ‘Reducing Litigation Costs Through Better Patient Communication’ (2004) 30 Physician Executive 36. 111 [2014] EWHC 720. 112 Large (n 76) [104]. 113 Maguire (n 31). 114 ibid [92]. 115 Jonathan Herring and others, ‘Elbow Room for Best Practice? Montgomery, Patients’ Values, and Balanced Decision-Making in Person-Centred Clinical Care’, (2017) 25 Medical Law Review 582. 116 Maguire (n 31) (HHJ Cotter QC) [78]. 117 A classic example of the potential consequences of failing to initiate an appropriate dialogue with a patient is found in the adult GP referral case of Holt v Edge [2007] EWCA Civ 602; (2007) 97 BMLR 74, a Court of Appeal decision included in this study. Here, it was recognised that the GP did not engage is sufficient ‘direct’ questioning of the patient in order to illicit an accurate diagnosis. The claim eventually failed because it was determined that, even if the GP had asked the appropriate direct question, it would not have prompted a response that would have necessitated an immediate hospital referral. 118 McCabe v Moore and Others [2015] EWHC 260; Donald and Others v Ayrshire & Arran Health Board and Others [2013] CSOH 23; 2013 GWD 7–166; Ganz (n 84); Large (n 76); Langdon (n 87). 119 ibid. 120 The need for accurate contemporaneous note taking by doctors is already contained in the GMC’s Good Medical Practice. See GMC (n 107) 9. The guidance stipulates that the documents doctors should make (including clinical records) to formally record work must be clear, accurate, and legible. Records should be made at the same time as the events being recorded or as soon as possible afterwards ([19]). Clinical records should include: relevant clinical findings; the decisions made and actions agreed and who is making the decisions and agreeing the actions; the information given to patients; any drugs prescribed or other investigation or treatment; and who is making the record and when ([21]). 121 Doy (n 60) [51]. 122 Montgomery (n 24). 123 Laing (n 28) 143, citing Phil Whitaker, ‘How One Tragic Case Changed the Law About Medical Consent for Us All’ (New Statesman, London, 28 May 2015) <https://www.newstatesman.com/politics/2015/05/how-one-tragic-case-changed-laws-about-medical-consent-all-us> accessed 2 October 2017. 124 ibid. 125 Thompson (n 94). 126 ibid [27]. 127 [2005] EWCA Civ 1439; [2006] Lloyd’s Rep Med 95. 128 The reasoning of the Court of Appeal was based on the fact that, as the specific risk of contracting Polio from the vaccine was unforeseeable to the GP, there was no ‘relevant’ duty to disclose it. 129 Montgomery (n 24). 130 Botting and others (n 2). 131 Baxter (n 36) [113] and [123]; Bulled v McClure [2006] EWHC 2444, [81]–[85]. 132 Toth v Jarman [2006] EWCA Civ 1028; [2006] 4 All ER 1276. 133 Large (n 76) [116]; Coakley (n 67) [111]. 134 Rana (n 69). 135 ibid [173]–[175]. 136 Coakley (n 67) [118]. See Bailey v Ministry of Defence [2008] EWCA Civ 883; [2009] 1 WLR 1052. 137 Bailey, ibid, (Waller LJ) [46]. 138 Jane Stapleton and Sandy Steel, ‘Causes and Contributions’ (2016) 132 Law Quarterly Review 363. 139 Fallon v Wilson [2010] EWHC 2978, [76]. 140 Rana (n 69) [174]. 141 Fallon (n 139) [76]. 142 Gregg (n 37). 143 Gregg (n 37) [206]–[207]. 144 Recovery predicated on any proof of lost years alone may yield very little in damages. Thus, alongside proof of lost years, damages may also be recovered for pain and suffering and financial losses associated with a delayed correct diagnosis. See, for discussion, Rachael Mulheron, Principles of Tort Law (CUP 2016) 474–75. 145 [2012] EWHC 3063; [2013] Med LR 291 at [46]. See also Patterson J in Cutting, n 111 at [105]. 146 JD, ibid [48]. 147 ibid [46]–[47]. 148 On the relevance of statistical evidence, see Gordon C Wishart and Andrew Axon, ‘Proof of Causation: A New Approach in Cancer Cases’ (2013) 19 Clinical Risk 130. 149 Oliver v Williams [2013] EWHC 600; [2013] Med LR 344 at [42]. 150 See Michael Jones, Medical Negligence (5th edn, Sweet and Maxwell 2018) 588–92; Michael A Jones, ‘Another Lost Opportunity’ (2003) 19 Journal of Professional Negligence 542. 151 Jane Stapleton, ‘Loss of the Chance of Cure from Cancer’ (2005) 68 Modern Law Review 996. 152 See GMC (n 107). 153 ibid 9. © The Author(s) 2018. Published by Oxford University Press; All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - “If the Problem Persists, Come Back To See Me … .”—An Empirical Study of Clinical Negligence Cases Against General Practitioners
JF - Medical Law Review
DO - 10.1093/medlaw/fwy030
DA - 2019-08-01
UR - https://www.deepdyve.com/lp/oxford-university-press/if-the-problem-persists-come-back-to-see-me-an-empirical-study-of-ftv9tya0QK
SP - 406
VL - 27
IS - 3
DP - DeepDyve
ER -