GPs are regularly cited as defendants in clinical negligence litigation. In this article, Justin Valentine, head of our clinical negligence team, identifies the most common areas of GP negligence based on cases with which he has dealt. These are bowel cancer, cauda equina syndrome, complications of diabetes and administrative errors. He discusses why these clinical areas prompt litigation and offers practical guidance for clinical negligence practitioners in relation to the identification of cases worthy of attention.
According to a 2018 article in the BMJ, GPs can expect to be sued once every 10 years. As a barrister practising in clinical negligence I see a range of claims against medical professionals. I undertook an audit of cases I have dealt with involving GPs in the last ten years to identify specific areas which prompt litigation.
There are three common features identifiable in many of the cases discussed below. Firstly, the clinical area involves a time-sensitive condition where if the correct treatment is not provided swiftly irreversible harm may result. Secondly, that harm is serious resulting in significant morbidity (or death), ie they are life-changing conditions which may prevent a patient from working or require extensive care needs. A third feature is that the symptoms complained of may be frequently encountered and can be relatively innocuous but in certain cases are indicative of far more serious pathology.
In many of the cases with which I have dealt, GP involvement within the treatment pathway has been exemplary. I have, for example, dealt with a number of squamous cell carcinoma cases which were initially assumed to be basal cell carcinoma. In all of those cases, the errors have been made during secondary care, eg dermatology, after the GP has correctly identified a suspicious lesion and made a two week wait suspected cancer referral.
Bowel Cancer
There are other areas of cancer however where delays in referral are identified on a more regular basis and foremost amongst those is colorectal cancer. Bowel cancer is an area where avoidable harm through late diagnosis has been recognised as a national issue and where NICE guidelines have correspondingly been tightened up. The introduction of faecal immunochemical testing has also assisted.
Of course, in the main GPs demonstrate good awareness of recent guidance either from NICE or from online resources (such as patient.info). A common area, however, is that of whether rectal bleeding is “unexplained” (NICE guidance suggests urgent referral for patients over 50 with “unexplained” rectal bleeding). This, for example, is a GP entry for a 62 year old [1]:
Several days intermittent PR bleeding. Not mixed with stool. Had been loose, noticed blood in pan and passed couple of clots (showed me a photo). Has had a bit of bleeding in past on toilet tissue.
There was nothing on examination, no external haemorrhoids, no masses and no blood on the glove. The patient made no complaint of pain or itching on passing stool and no complaint of constipation. Despite this the GP diagnosed haemorrhoids and advised cream (with nothing to put it on). The patient was asked to re-attend if the symptoms did not settle but did not do so as the patient was reassured they had haemorrhoids, did not wish to waste their GP’s time and was embarrassed. In the event, they developed bowel cancer. A settlement was reached though damages were limited by short life expectancy.
Bowel cancer is a tricky area for two reasons. Firstly, many patients present with rectal bleeding and in the vast majority of those cases the reasons for that bleeding will not be ominous. Not all patients can be referred for a sigmoidoscopy or under the two week suspected cancer referral for rectal bleeding; hospitals would be over-flowing. Secondly, patients are understandably reticent about re-attending and if they have received a reassuring diagnosis (usually of haemorrhoids or anal fissure) may not do so.
It is noteworthy that the oncological and colorectal experts commissioned to address causation in bowel cancer cases are often more critical of the GP’s failure to refer than the GP expert. The GP expert has greater awareness of the challenges facing GPs but it is the oncological and colorectal specialists who have to pick up the pieces where there is delay in referral. This suggests a misunderstanding, or lack of clarity, as to the threshold for referral either on the part of the GP or within local referral pathways.
Cauda Equina Syndrome (“CES”)
CES is an area which generates litigation for three reasons. Firstly, CES can cause permanent bladder, sexual and bowel dysfunction as well as mobility problems; patients who have suffered such life-changing injuries will scrutinise their treatment pathways if they feel that their symptoms have been missed. Secondly, the symptoms the patient complains of (at least initially) are frequently encountered; back pain is very common. Thirdly, CES is a time sensitive condition where delays in undertaking decompression surgery measured in hours can alter the outcome.
Not every patient with sciatica (even bilateral sciatica) can be referred for an MRI scan or to the local spinal or neurosurgical service [2] but, according to GIRFT [3], sudden onset bilateral sciatica without CES symptoms should lead to referral within two weeks and where there are CES symptoms referral should be made on an emergency basis.
CES cases generally involve failure to recognise impending CES where there is bilateral sciatica (leading to no or non-urgent referral), failure to provide adequate “red flag” warnings or failure to recognise “red flag” symptoms. The latter is the most marked. Consider these GP entries where breach of duty was identified:
Telephone encounter.
Low back pain 1 week and radiating down right thigh with sciatica radiating down to the toes.
Numbness buttock and into groin.
Referral to musculoskeletal clinic.
There is no benign explanation for numbness in the groin (which most patients understand as meaning the private area). The patient should have been referred immediately to the local spinal or neurosurgical service or simply told to attend the Emergency Department. Or this GP entry:
Numbness and burning sensations going down leg and over left buttock and left side of vagina … No faecal incontinence – but can’t feel bottom …
Impression: Sciatica, borderline for hospital review. Try Diazepam.
Again, this falls far below the standard to be expected of a reasonable GP; an immediate referral should have been made.
In relation to red flag warnings, litigation can arise where difficulty with urinating or with bowels has been mentioned by the GP but not numbness in the perineal area with the result that patients do not seek further treatment with developing numbness, especially if they are already waiting for a musculoskeletal or neurosurgical appointment, before finally seeking urgent treatment when they go into urinary retention. There appears on occasion to be a knowledge gap in relation to perineal numbness which is generally a precursor to urinary and then bowel problems. Once there are urinary symptoms then full recovery is far less likely.
The Musculoskeletal Association of Chartered Physiotherapists (“MACP”) have developed an excellent warning card for patients[4]. The first three warning signs on the back of that card are directed at numbness or altered sensation to the perineal area before then moving onto urinary dysfunction and then bowel dysfunction, ie mirroring the usual development of CES symptoms. If a patient has been warned about saddle numbness (and other CES symptoms), given a leaflet or the MACP’s card about CES and this has been noted in the medical records then a GP is unlikely to be in breach of duty in relation to giving red flag warnings.
CES cases often involve multiple treatment providers and confirmatory bias can be evident. If, for example, a patient has been to the Emergency Department (“ED”) and been reassured (wrongly) that there is no impending CES this may then be accepted by the GP when the patient returns to the practice with even more ominous symptoms. One of the commonly identified features in a human factors approach to avoiding clinical accidents is hierarchy and it cannot be assumed that practitioners within a secondary care setting do not make diagnostic errors.
Again, it is worthy of note that some GPs demonstrate exemplary care. In a case I have recently dealt with the patient, suffering with perineal numbness was referred back to their GP from the ED. The GP wrote a letter for the patient to take to the ED. He noted her attendance the previous day but that in his opinion she needed an urgent MRI scan and that CES could not be excluded.
Patients with Diabetes
People with Type 1 and Type 2 diabetes have a greater risk of infection. Many of the cases that I encounter arise out of failure fully to appreciate this increased risk or that oral antibiotics are unlikely to be effective. This can lead to significant morbidity; lower limb amputation, sepsis or, in rare cases, necrotising fasciitis.
The following patient (55 at the time) presented to the local out of hours primary care service where they saw a GP:
History: painful R bum cheek 24 hours, sweating and shivering last night …
Examination: temp 38.1 BP 101/70 pulse 98, very tender R buttock localised just outside the anal verge – no cellulitis.
Diagnosis: buttock abscess ??brewing.
Treatment: should respond to oral abc
In a diabetic patient with a raised temperature, raised pulse, systemic symptoms of infection and obvious signs of an abscess, oral antibiotics will have no effect. The only treatment is surgical and they should have been urgently referred. Unfortunately, this patient went on to develop necrotising fasciitis requiring extensive surgical debridement, a split skin graft and a defunctioning colostomy.
The following diabetic patient had long-term foot problems and had been referred to diabetic podiatry (on an urgent not emergency basis). They then saw the practice nurse who noted:
History: Blister on right foot broken down over weekend.
Examination: Foot warm to touch .. blister on underside of heel of foot broken down. Looks like black area underneath blister, wound open approx 0.5cm in diameter with thick bloodstained exudate. Surrounding area inflamed. Discussed with [GP]. Podiatry referral made – advised patient to call if not heard anything by tomorrow. Abx prescribed.
At this stage, the patient was a diabetic foot care emergency with evidence of rapidly developing infection. Oral antibiotics were unlikely to be effective and referral to podiatry inadequate. NICE guidance mandated referral to a multidisciplinary diabetic foot care team within 24 hours. There were subsequent failures in both primary and secondary care and the patient suffered a below-knee amputation. In the event, the trust compromised the claim.
Administrative Failures
Administrative and system failures are sufficiently common to be identified as a class of cases in and of themselves. In many of these cases, no one individual can be identified as at fault or it may be that a member of the administrative staff has made an error. However, under the law of joint and several liability, the partners of the practice will be liable if fault is identified.
Administrative and system failures do not generally require expert evidence on breach of duty since it is accepted that absent that error the appropriate treatment would have been provided or referral made. Rather, there may be issues of fact as to what a patient was told, by whom and when. This is usually pieced together from the medical records, from phone records and from the patient’s own recollection.
Examples of administrative failures are varied. I have, for example, encountered the following:
- A patient was aware of a strong family history of osteoporosis. She was sufficiently concerned that she presented to her GP and a DEXA scan was arranged. When she rang for the result she was told that no action was required even though the surgery had not received the results at that time only confirmation that the scan had taken place. The medical records, consistent with patient’s account, noted “no action”. The hard copy report was apparently never received by the surgery. She subsequently suffered several wedge fractures as a result of osteoporosis which went untreated.
- A patient with bilateral sciatica was prescribed Tramadol. She began to feel loss of feeling when wiping after going the toilet. She rang her practice asking for an appointment. The receptionist put her on hold and spoke to somebody (presumably a nurse or GP). The receptionist then informed the patient that numbness could be caused by Tramadol (which the leaflet confirmed). She called the practice again when she was in urinary retention and was advised to attend the ED. CES was diagnosed and surgery undertaken but the patient did not make a full recovery.
- An MRI received in the practice identified Charcot foot which was not actioned by the GP at all and the patient was subsequently informed she merely had osteoarthritic changes. The patient’s Charcot foot was left untreated for many months resulting in avoidable deterioration and permanent disability.
The cases above suggest that there must be a clear protocol as to how to deal with incoming test results understood by all members of the practice (including locums). Relying on patients to request the outcome of an investigation is unsafe. There needs to be clarity in relation to interactions by reception staff with patients.
Practice Points
If a condition, such as CES or bowel cancer, can lead to significant disability, then it is the duty of the GP to be particularly alert to any warning signs and to ensure the patient understands the potential seriousness of the condition. As well as reviewing the entirety of the patient’s (unredacted) records, any recorded calls made by the patient (to the GP or to NHS 111) or by the GP to a hospital should be obtained. This can provide strong evidence of the contemporaneous symptoms complained of and the patient’s understanding of those symptoms.
Cases based on system failure can be investigated by scrutiny of relevant protocols and GP systems in relation to the receipt and review of tests results and to all contacts between the practice and the patient.
In general terms, it is hoped that the above analysis will assist practitioners to identify areas of potential negligence early in the life of the case (even at a screening stage). Attention should always be given to cases where there is a time-sensitive condition, which may lead to significant morbidity, and where the symptoms complained of can be relatively innocuous but possibly indicative of serious pathology. Risk analysis promotes identification not only of the likelihood of risk but also the potential impact should that risk materialise and would suggest that clinical areas which meet these criteria are worthy of particular attention.
Justin Valentine is a barrister and head of clinical negligence at St John’s Chambers, Bristol.
[1] The cases cited are actual cases albeit that identifying features have been removed or altered where possible. The key clinical information has not been modified.
[2] According to patient.info “CES is reported in approximately 0.04% of all patients presenting with low back pain”.
[3] Pathway supports clinicians to diagnose and treat Cauda Equina Syndrome without delay. Read here.