Appendices A & B to Chapter 3 - Investigations
Jump to
Appendix A: Report by the External Professional Advisers to The Health Service Ombudsman on Complaints from Mr J
First external professional assessor:Mrs P - MA RGN HV Cert
Relevant experience: Diabetes Nurse Specialist
Second external professional assessor: Dr U-MB BS
Relevant experience: Consultant PhysicianThird external professional assessor: Professor E
Relevant experience: Professor of Clinical Psychology
Fourth external professional assessor: Dr S - PLD, FRCP
Relevant experience: Emeritus Consultant Physician
Fifth external professional assessor: Ms J - RGN MSc
Relevant experience: Clinical Risk Manager
Introduction
1. We have been asked to consider whether the care and treatment provided to Miss J was adequate in terms of what might reasonably be expected of staff at a District General Hospital.2. Our advice in relation to specific issues is set out below; one or more external professional assessors have contributed to each section. We have examined the nursing and medical records in detail; the papers provided by Mr J, the Trust and the investigator's interview notes.
Complaint (a) the care and treatment provided to Miss J by Halton General Hospital NHS Trust was inadequate
Miss J's care at the Diabetic Clinic
Clinical care
Initial diagnosis of maturity-onset diabetes of the young (MODY)
1 Miss J saw the first consultant in the general diabetic clinic in July 1996. She was not overweight and had not lost weight. The plasma glucose concentration and the haemoglobin Alc (HbAlc) percentage were both in the normal range. The consultant understood that two family members were diabetic: Miss J's mother and her maternal grandfather. The consultant diagnosed maturity-onset diabetes of the young (MODY). Miss J was told that the diagnosis was MODY; her clinic card had written on it "I have MODY diet controlled diabetes". The general practitioner was told that the diagnosis was MODY.
2. Miss J understood that a diagnosis of MODY meant that she would never require insulin. The diabetes nurse specialist taught her to test her blood glucose using reagent strips. She was advised to test her blood glucose about three times each week. She did not take any medications.3. The diabetes nurse specialist and the first consultant recollect that Miss J was advised to test her blood glucose frequently - at least daily - if she was in any way unwell. Miss J's brother states that she did not receive this advice. Miss J was not advised to test her urine for ketones if she was unwell.
4. The first consultant saw Miss J at intervals in the young persons diabetic clinic. When she was seen at the clinic the haemoglobin Alc (HbAlc) percentage was measured.
5. Miss J measured her blood glucose reasonably regularly up to the 14th of July 1997. Her results do not show significant worsening of her diabetes apart from the final result, which was 11 mmol/L. The diabetic clinic haemoglobin Alc results show slight worsening of her diabetes in 1997 compared with 1996.
Discussion
6. When Miss J, a 19 year-old, presented with diabetes mellitus it was not necessary to make a diagnostic choice as to the type of diabetes as her diabetes could have been managed quite satisfactorily as 'diabetes mellitus, type unspecified'. In time the true diagnosis would have become apparent.
7. There were two possible further diagnoses: Maturityonset diabetes of the young (MODY) and type 1 diabetes. Type 1 diabetes is also known as insulin-dependent diabetes mellitus (IDDM). These diagnoses are mutually exclusive. Maturity onset diabetes or non-insulin-dependent diabetes mellitus (NIDDM) is now referred to as type 2 diabetes.
8. MODY is an inherited type of diabetes with onset in adolescence to young adulthood. It is genetically and phenotypically heterogeneous. Some phenotypes are not benign as the diabetes may progress with age, requiring treatment with insulin and causing microvascular complications. Ketoacidosis does not occur in MODY.
9. Type 1 diabetes is more common than MODY in adolescents and young adults. The underlying pathology is autoimmune destruction of the insulin secreting beta-cells in the pancreas. If the diagnosis of diabetes is made when a considerable number of beta-cells are intact the diabetes may be well controlled without treatment until such time as a significant decline in beta-cell numbers occurs. Progressive loss of beta-cells is associated with deterioration in metabolic control, which, when severe, results in ketoacidosis. In most type 1 diabetics dependence on insulin develops within one year of diagnosis; in some it may take years to develop.
10. These two types of diabetes cannot be distinguished with certainty on clinical grounds at the time of diagnosis. If diabetes can be managed for a prolonged period - several years - without insulin MODY becomes the more likely diagnosis.
11. Laboratory tests can aid diagnosis: in some, but not all forms of MODY genetic defects can be identified, and in type 1 diabetes circulating antibodies to insulin and pancreatic islet cell components can be detected. These tests are not routinely available.
12. A diagnosis of MODY could only have been made when Miss J presented with diabetes if type 1 diabetes could have been excluded. She was, however, given a diagnosis of MODY without type 1 diabetes having been excluded.
13. By being given the diagnosis of MODY Miss J was not made aware of the risk of metabolic deterioration to ketoacidosis which can occur in type 1 diabetes. She was not told about testing for ketones in urine if the blood glucose is high. She was given advice appropriate for a type 2 diabetic. It is not known if the symptoms of uncontrolled diabetes were explained to her.
14. In fact, regular monitoring of the blood glucose with regular checks of the haemoglobin Alc was an appropriate way to monitor Miss J's diabetes provided she was not unwell. In both type 1 and type 2 diabetes metabolic deterioration occurs during intercurrent illness so that all types of diabetes require frequent blood glucose monitoring during illness of any sort. In Miss J's case blood glucose monitoring alone would have been sufficient to reveal the serious metabolic deterioration after the 14th of July 1997.
Conclusions
15. It was an error for the consultant to give Miss J a diagnosis of MODY in July 1996. The consultant should not have excluded the possibility of type 1 diabetes developing slowly. If he had included this possibility Miss J would have been made aware of the risk of ketoacidosis. This might have altered the way she responded when she became unwell in July 1997. It might also have altered the way the general practitioners, whom she saw when she was unwell, responded. The HbAlc results revealed that there was a slight deterioration in diabetic control which would not require anything other than a continuation of a watch and wait policy and monitoring.
Recommendations
16. The diagnosis of maturity-onset diabetes of the young should be made only when type 1 diabetes has, essentially, been excluded.
17. Both general practitioners and physicians should be vigilant as a small number of diabetics considered to have MODY or type 2 diabetes may in time turn out to have type l diabetes.
18. Diabetics and general practitioners must be aware of the risk of metabolic deterioration in association with intercurrent illness. Appropriate monitoring of the diabetes is mandatory.
Compliance
19. Compliance in medical usage means acting in accordance with rules, plans or advice.
20. Miss J was advised to measure her blood glucose approximately three times each week. The exact advice was not recorded. Miss J did measure her blood glucose frequently. The most common interval between tests was four days, followed by three, and then five days. There were occasional intervals of several weeks when she did not record any results. These appear to have occurred when she was on holiday - and feeling well. As far as we can ascertain Miss J was very careful with her diet. Her compliance with the advice she had received was good until she became unwell in July 1997.
21. After recording a blood glucose result of 11 mmol/L on the 14th of July Miss J attempted to test again on 17 July but could not read the result and then did not test her blood glucose again even though she became progressively unwell.
22. Why Miss J did not test her blood glucose between the 17th and the 22nd of July is not known.
23. In general Miss J followed the advice that she was given. What happened after the 17th of July 1997 is a mystery. We do not know for sure what she was advised to do if she became unwell; the right advice was to check the blood glucose at least once daily. We do not know what she was advised to do if the blood glucose results were high; the right advice was to contact the practice nurse or her general practitioner. We do not know if she knew the symptoms of uncontrolled diabetes: thirst, frequent urination, tiredness and weight loss.
24. We can come to no clear conclusion about compliance during the week before Miss J presented with severe diabetic ketoacidosis. Before then her compliance was good.
Nursing care
25. We have been asked to consider the role of the diabetes nurse specialist in the initial diagnosis of MODY, the subsequent educational programme and monitoring of diabetic control.
26. Miss J was diadnosed as a type 2 diabetic (MODY).
Discussion
27. The diabetes nurse specialist, with less than two years experience in this post, would have had little exposure to the management and possible progression of MODY. Although more information is now being published about MODY, in 1996 we would consider that the diabetes nurse specialist would not have had easy access to this little reported yet complex subject. A more experienced nurse may have asked for further clarification but we do not find it unreasonable that the diabetes nurse did not challenge the consultant's diagnosis.
Conclusion
28. We do not consider that the diabetes nurse specialist can be held accountable for the diagnosis.
Recommendation
29. t is useful for members of the diabetes team to have regular meetings to discuss unusual or complex patients. This acts as an educational process for the health professionals and ensures that there is agreement about the advice given to the patients, the treatment plan and parameters of control before further intervention.
The education programme and monitoring
30. Diabetes education began at the GP surgery and by the time Miss J saw the diabetes nurse at the hospital, she had been subject to several consultations concerning diabetes. The education programme initiated by the diabetes nurse specialist was that related to type 2 diabetes. Although not present during the first consultation, she saw Miss J at the clinic and once at home.
Discussion
31. The diabetes nurse specialist's records were very brief and it is not possible to determine precisely what information was given to Miss J. She said that she gave both verbal and written information about diabetes: Mr J disputes the content of both. Over the years there has been a wealth of diabetes literature published by many of the pharmaceutical and product companies concerned in diabetes care. Leaflets and booklets covering a wide range of diabetes topics have been renamed and redesigned and have readily been used by diabetes nurse specialists for patient education. If Miss J received "Diabetes, Diet and Tablets" (although no longer published, it was in print throughout the early nineties), then there is a section on illness and deterioration of diabetes control.
32. The diabetes nurse specialist did not record the level at which Miss J was supposed to contact them. The evidence given, at various times during the different stages of investigation of Mr J's complaint, has varied. The consultant and the nurse have said that Miss J would have been told to report a reading of 9, 10 or 11 and that she was told to test 2-3, 3-4, or 4-5 times a week. In the event she tested approximately every 3-5 days which roughly accords with the advice stated to have been given.
33. Miss J contacted GPs when unwell and told them that she had MODY. She does not appear to have told them that her last recorded reading was llmmols.
34. The diabetes nurse specialist told the investigator that when teaching someone, like Miss J, to test their blood she would tell them what the normal range was and that they would expect their fasting blood sugar to be 4-7, their pre-meal reading to be 4-7 and her random blood sugar to be not above 9. If their readings were regularly above that (i.e. more than 23) or if there seemed to be a different pattern to the results they should let the clinic know. Miss J was told to test 2-3 times a week. If they got a high reading, for example 11, they would be advised to test again the next day if they felt well or later that day if they felt unwell. They would always be asked to consider how they were feeling. She would explain that the reason they were testing was to avoid hyperglycaemia; they would be encouraged to understand their normal levels so that they had a pattern against which to compare.
Conclusions
35. It is possible that by April 1997, more frequent monitoring could have been advised because of the rising HbAlc. This could have produced an opportunity to further discuss any appropriate treatment changes but the change was relatively small and the nurse was not in error to continue with the regime already implemented.
36. We do not find the diabetes nurse specialist remiss for not teaching ketone testing and its relationship to diabetic ketoacidosis in this situation of education based on type 2 principles. It may be argued that should this information have been imparted it is unlikely that it would have been remembered. Therefore the awareness of change if unwell or not is a more important message. We can not tell from the evidence whether or not this information was given to Miss J.
37. What clinicians are looking for, is an altered pattern of results, not a single deviation from the norm. The diabetes nurse specialist did not remark upon an earlier instance of a reading of 11 in the notes, but she did record 'one odd one of 9mmols' which indicates that the reading had significance for her. We cannot determine, from the evidence, precisely what Miss J was told. All we can be sure of is that the diabetes nurse specialist did not fully document her discussions with her patient.'Info. given' and 'general chat' are not sufficiently specific.
38. If one were to speculate, however, it would seem unlikely that Miss J would have been told to report a single high reading but would rather have been advised to report more than one. She recorded a single high reading on 14 July and when she tried to test again, she thought, according to her niece's evidence, that the reading was between 7-9. By the time she might next be considered due to test, she was feeling unwell and had consulted GPs on four occasions who did not themselves test her blood sugar or advise her to do so. The fact that Miss J did not report a single high reading of llmmols is, therefore, not evidence that she received inadequate education.
39. The diabetes nurse specialist has accepted that her documentation was insufficient and she has taken steps to improve it. The standard of her note keeping is not in fact any less than that of many nurses in practice and falls within the norm. However, this case has demonstrated that that standard, in diabetic care, is not itself adequate. On balance, and being careful not to judge her actions with hindsight, we do not think that the diabetes nurse specialist provided inadequate care to Miss J. Clearly, if the diagnosis had been different, so too would the level of information provided to Miss J.
Recommendations
40. As there is a diverse range in the quality of printed diabetes educational materials as provided by companies, it may be more prudent for a diabetes team to agree to standardise and print their own. This has financial implications but would prevent inappropriate or out dated material being used.
41. We recommend that nurses should record that they have taught normal/abnormal parameters of control; advised the patient to contact them if they record x number of results which deviate from the norm at a level of x, and told them to test x times when feeling ill. We understand that nurses find it difficult to find time to document their care as well as deliver it, but records are a vital nursing tool - they act as an effective check-list against which a nurse can evaluate whether they have provided all the care/information they intended. They are the means by which a nurse can act as a fully accountable practitioner and demonstrate the fact.
Admission and initial management of diabetic ketoacidosis
Clinical care
Discussion
42. Miss J was seen by the SHO within 15 minutes of admission on the morning of 22 July. The SHO made a rapid, detailed assessment of her patient, diagnosed severe diabetic ketoacidosis, requested all appropriate investigations and instituted initial resuscitation measures. Within the hour she had discussed the situation with her immediate senior colleague, the first registrar, and met the parents to explain that their daughter was seriously ill and, on the registrar's advice, would be transferred to ITU. The SHO had no further involvement with Miss J's care until late afternoon.
43. The SHO was employed as a locum and was new to the hospital. The SHO told the investigator that she had not received any formal induction on taking up her locum post nor was she made aware of any protocols for the management of common medical emergencies until 11.00pm that night. Her management was based on her previous experience of caring for people with diabetic ketoacidosis.
Conclusion
44. The SHO's initial assessment and management of Miss J cannot be criticised. She gave exemplary care to Miss J and her immediate family.
Recommendation
45. All new and locum personnel should receive formal induction into the working practices of the medical department which should be centrally co-ordinated. This should include protocols for the management of common medical emergencies which should be available in all areas where such emergencies are likely to present for care.
Admission to ITU and further management of diabetic ketoacidosis
46. Following admission to ITU, the first registrar attempted to insert a UP line to monitor Miss J's fluid balance but she was unsuccessful. She decided that a UP line was not essential partly to avoid further distress to Miss J and because her pulse, blood pressure and urine output were considered satisfactory. Miss J declined a urinary catheter and the first registrar complied as her urine output was copious. The consultants were aware of her decision and did not reverse it. The first registrar also asked an anaesthetic SHO to insert an arterial line but this too was unsuccessful.
47. The first consultant told the investigator that he viewed his role as advisory. He saw Miss J on three occasions during the day and provided telephone advice during the evening. The second consultant said that he understood the first consultant to have taken over clinical responsibility for Miss J and as a result did not assess her himself until after her collapse at 3.00am. He had indirect involvement in the patient's care.
48. The first consultant's notes were of an advisory nature. There was no clear documented management plan. The SHO was not aware that there was a diabetic ketoacidosis management protocol until 11.00pm. The first consultant was not aware that the SHO did not have access to this protocol according to his written statement.
49. At 6.00pm the SHO resumed responsibility for Miss J's care. There was a locum registrar (the second registrar) on duty but he had no direct involvement before Miss J's collapse apart, from one telephone discussion with the SHO.
50. At 8.00pm the SHO was called to ITU because Miss J had developed a severe headache. The SHO noted that her metabolic state had not improved and was unsure which advice to follow. She called the second consultant who advised her to call the first consultant at home. She successfully contacted him at 9.40pm and he advised a change in treatment (see paragraph 18 main report). By 11.30 Miss J appeared settled. At 3.30am she was called urgently to ITU because Miss J had collapsed.
Discussion
51. When a patient presents with severe diabetic ketoacidosis in a state of mental confusion, significant dehydration and a high urine flow most diabetologists and critical care physicians would consider a UP line a prerequisite for the proper assessment of intravascular fluid status and the rate of fluid replacement needed. Likewise, the necessary, frequent measurement of electrolytes, pH and blood pressure to monitor the effects of therapy often necessitates the use of an arterial line. The first registrar initially, correctly concluded that both were desirable. She was deflected from achieving her objective by the practical difficulties of inserting them, which may have been the result of inexperience in practical procedures, and a desire not to cause her patient further distress. Her consultant colleagues concurred with her decision and should have realised that this made their own active role in the supervision of Miss J's monitoring and management all the more important.
52. By late afternoon the patient's acidosis was not significantly better, nor had she received adequate fluid replacement. The consultant was clearly concerned about her potassium level and pH. His notes, however, are of an advisory nature. There was no clear, detailed plan of management entered in the notes with set parameters, targets and related instruction. As a consequence, the SHO who had received conflicting advice, was unclear what action to pursue and was unaware that there was a treatment protocol.
Conclusion
53. Overall, the ITU management of Miss J's diabetic ketoacidosis was satisfactory from the point of view of her electrolyte status but her fluid balance and acid-base derangement were never adequately assessed or fully and rigorously attended to. In a case of diabetic ketoacidosis there may be a deficit to the equivalent of five or more litres of normal saline. In Miss J's case, according to the available fluid records, and (allowing for inaccuracies in the nursing fluid charts), there was a negative balance (if insensible loss is taken into account) during the first nine hours of treatment and even by the twelfth hour of care there was less than a litre positive balance. Thus the use of a CVP line would have been invaluable in assessing both the status of her intravascular volume and fluid requirements. The pH never improved significantly during the first twelve hours of therapy. The use of 8.4% sodium bicarbonate is debatable. Intensivists working in an ITU would probably be confident in giving boluses of such bicarbonate strength, but, in general, most diabetologists would err on the side of caution and give only 1.26% or 2.8% at the most, and tentatively. The administration of bicarbonate runs the risk of lowering the potassium, producing a fall in the pH of the cerebrospinal fluid (increasing brain acidosis) and impairing the level of consciousness. The SHO was unsure as to what to do with regards to Miss J's acidosis. In fact, the 'front line' doctor care for Miss J on the ITU was being supplied by three junior doctors; a locum SHO, a locum Registrar and a locum anaesthetic SHO, all of whom acquitted themselves well for their level of expected competence.
54. Neither of the senior doctors considered that he had responsibility for overseeing Miss J's total care, indeed each thought that the other was responsible. The first consultant, in his written statement, seemed under the impression that the SHO was following the hospital protocol for the management of diabetic ketoacidosis: this was not the case. As a consequence, the SHO was unsure how to proceed with Miss J's care at a time when she had been receiving therapy for over ten hours and yet her acidotic state had not improved and, despite the SHO's best efforts it was almost two hours after she was called to the ITU at 8.OOpm to review Miss J that the SHO finally received further instructions concerning a change in therapy for tackling the continuing acidosis. Only after Miss J's condition had deteriorated did the consultant physician on call (the second consultant) review her himself. Whether regular, at the bed, assessments by a senior doctor would have altered the outcome of Miss J's admission is open to speculation.
Recommendations
55. Any person admitted to hospital with a serious, lifethreatening condition such as diabetic ketoacidosis should have their case reviewed by a doctor of registrar status or above. Where a case of diabetic ketoacidosis does not respond to therapy in the expected manner there should be available a consultant with an over arching responsibility for coordinating all aspects of medical care. The consultant on call should always adopt a 'need to know' policy as to who has specific responsibility for this over arching role. In this particular instance this does not appear to have been the case. There must be a clear, consultant led, management plan.
56. Where a person in a state of diabetic ketoacidosis is not, within a few hours of treatment starting, positively to progress with a steady rise in their pH, a senior doctor should be available, in person, to reassess the situation and look for causes of continuing acidosis; for example lactate build up, hyperchloraemia or inadequate fluid replacement. It should never be the case that an on call SHO has responsibility for the ITU management of such a serious condition as diabetic ketoacidosis.
Hyponatraemia
57. When Miss J was admitted to the hospital on the 22 July 1997 the initial tests in the minor injuries unit revealed hyponatraemia in addition to acidosis (arterial pH 7.09). The serum concentration of sodium was 109 mmol/L, potassium 6.7 mmol/L, creatinine 186/imo1/L and glucose 38.8 mmol/L. Treatment with insulin and with 0.9% saline solution, given intravenously, was started at about 11.OOam.
58. By 3.00pm the volume of saline infused was between 2fi and 3fi L. The record of infused fluids over this period was inadequate. By 1.30pm urine flow was good, suggesting that the extracellular compartment volume had been restored. By 3.OOpm the serum concentration of sodium was 122 mmol/L with potassium 4.9 mmol/L, creatinine 108/1mo1/L and glucose 14.2 mmol/L. The central venous pressure was not monitored; an attempt to insert a cannula had failed. No explicit treatment plan was formulated.
Discussion
59. Hyponatraemia is defined as a serum sodium of less than 136 mmol/L. Hyperglycaemia causes translocational hyponatraemia; the presence of a high concentration of glucose in the extracellular compartment decreases the serum sodium concentration by shifting water from the intracellular compartment to the extracellular compartment. [This reduction is 3.0 mmol/L of sodium for each 10 mmol/L increase of glucose. This leads to a net increase in osmolality of approximately 3.6 mOsm/Kg for each 10 mmol/L increase, of glucose.]
60. Miss J was clearly much more hyponatraemic than expected purely on account of the translocational effect of hyperglycaemia. Her serum osmolality was below normal. The probable cause of this component of the hyponatraemia was that she drank a lot of water when she was dehydrated and sodium deficient. Hypotonic hyponatraemia does cause brain swelling. It is of considerable interest that hyponatraemia has been reported in a high percentage of patients with diabetic ketoacidosis who develop cerebral oedema.
61. Treatment of ketoacidosis with insulin and the restoration of the extracellular compartment volume with 0.9% saline solution take precedence to concerns about hyponatraemia per se in diabetic ketoacidosis. However, it is prudent to avoid rapid administration of 0.9% saline after the extracellular fluid volume has been restored to normal in order to avoid a rapid increase in the serum sodium concentration which can occur in the presence of high urine output. Neurological sequelae following rapid correction of hyponatraemia include pontine demyelination which is associated with shrinkage of the brain as the serum sodium concentration rises. The infusion of 0.9% saline does not cause an increase in the intracellular compartment volume; infusion of excess saline does not cause brain swelling.
62. In this situation the measurement of the central venous pressure would have been most helpful in guiding the rate of fluid replacement.
Conclusion
63. Miss J had significant hyponatraemia; more than could be explained by hyperglycaemia. This was an additional factor reflecting the severity of her metabolic disorder. It increased the risks of neurological problems occurring during corrective treatment.
Recommendations
64. A physician with experience of diabetic ketoacidosis and hyponatraemia should manage a case of this nature. A careful treatment plan is needed with estimates of the deficits of water, sodium and potassium. Objectives should be set and progress monitored frequently.
65. Regular neurological monitoring is required in severe diabetic ketoacidosis, particularly when hyponatraemia is present.
Cerebral oedema
66. At 8.00pm on 22 July Miss J developed a severe headache. She was agitated. Her conscious level was said to be normal; she had been somewhat confused when she was first admitted to the hospital. Treatment with insulin and intravenous fluids had started about 9 hours before the headache developed. Miss J was given some Paracetamol. The headache settled after about one half hour.
67. The headache recurred at about 2.OOam the next morning. Within 10 minutes Miss J's conscious level had deteriorated and after 30 minutes she was unable to sustain adequate respiration. At this point the neurological catastrophe was irreversible.
Discussion
68. Children, occasionally, and young adults, rarely, die suddenly from the rapid onset of neurological coma during treatment for diabetic ketoacidosis. This neurological catastrophe - cerebral oedema - tends to occur between 8 and 24 hours after the initiation of treatment. Treatment often appears to be going well. The outcome is usually death, occasionally recovery with severe neurological damage.
69. Brain swelling has been shown to occur generally in diabetic ketoacidosis. During treatment to reverse the dehydration, acidosis and hyperglycaemia the brain swelling increases. The catastrophe occurs when the cranium can no longer accommodate the expanding brain. Why brain swelling occurs in all cases of ketoacidosis but the syndrome of cerebral oedema in very few is not understood.
70. There is no consensus as to the cause of cerebral oedema. It occurs in severe ketoacidosis not in hyperglycaemia per se, even with extremely high glucose levels. Those who present with hyponatraemia and ketoacidosis have an increased risk of developing cerebral oedema. There is a suggestion that excessive fluid replacement may be a causative factor, (it should be noted that 0.9% saline and glucose solutions have quite different effects). Rapid correction of hyperglycaemia has been suggested as a factor leading to brain swelling; many advise slow correction.
71. If diabetics with severe ketoacidosis and those with associated hyponatraemia have regular neurological observations close monitoring may pick up early signs of cerebral oedema. Treatment to reduce brain swelling can then be considered in suspect cases. Treatment could include intravenous Mannitol, a reduced rate of fluid administration and mechanical hyperventilation. There are, however, no studies to confirm or refute the value of these interventions.
72. When Miss J developed severe headache at 8.OOpm she probably had early cerebral oedema. She certainly had cerebral oedema by 3.OOam the next morning when the headache recurred.
73. The medical staff, who were involved in Miss J's care were generally unaware of cerebral oedema as a potentially life-threatening problem and appeared to have little idea of the action which could be considered if cerebral oedema was suspected. Eventually dexamethasone was given both for cerebral oedema and possible pituitary apoplexy.
Recommendations
74. Young adults as well as children, with severe diabetic ketoacidosis, should have regular neurological monitoring with the aim of detecting the early signs of cerebral oedema.
75. Advice about cerebral oedema should be included in the protocols used for the management of diabetic ketoacidosis by all intensive therapy units that might look after young adult diabetics or children with diabetes.
ITU nursing care
Discussion
76. It is of concern that the fluid charts were not maintained accurately. It is particularly important to keep an accurate record of fluid balance in relation to any disorder where large volumes are infused and where the output (urine) materially alters the plans for input.
77. There is no evidence that a care plan was drawn up for Miss J, the care provided by nursing staff was commensurate with the directions they received from medical staff. It is a matter of concern that the coordination and monitoring of Miss J's care was left solely to the SHO, even though she was a locum and unsure as to how the care should be fully managed. It is reasonable to expect ITU staff, particularly at senior level and in charge of a unit, to be aware of any patient where there is clear evidence that treatment regimes are failing to have an impact on the underlying clinical presentation. In this particular situation where there was a significantly deteriorating metabolic acidosis which was not responding to treatment, it is reasonable to expect that the nursing staff should equally take steps to address the issue which might necessitate the facilitation of a more senior medical review. Such action should have been linked to an agreed formal plan of treatment and nursing care plan along with frequent reviews of progress against outcome objectives.
78. Under the Code of Professional Conduct, nurses have a responsibility to raise issues of concern regarding patients to more senior clinical staff if they see fit. Miss J's fluid replacement/balance and continual metabolic derangement should have prompted such action.
79. The standard of nursing notes is reasonable. We note however, that the accounts of Miss J's headaches as recorded in the notes do not match the staff's or family's descriptions given in evidence. It would have been helpful if they had recorded the location of the headache (occipital region) and the sensations of doom reported by the patient. Miss J reported that she felt as if she was having a brain haemorrhage and later that she felt that she was going to die. With hindsight, it is clear that Miss J was suffering a neurological catastrophe. The impression given by the nursing notes is that she suffered a relatively minor headache which settled with Calpol and that was the impression gained by the SHO. The charge nurse said that the patient suffered a headache more or less continuously. The headache at 2.00am was not reported to medical staff. The first consultant said that he was not aware of the headache and that he would probably have suggested a CT scan, although he would not have thought of cerebral oedema.
80. Of particular concern is the evidence which suggests that staff had formed a less than favourable view of their patient and therefore to some extent appear to have imposed value judgments upon the information she provided. She was viewed as a 'difficult' patient obstructing her care by refusing interventions. It appears that this coloured the staff's response to her reports of pain; pain is a subjective experience and it is essential that the patients experience is both believed and acted upon, not interpreted by staff. Miss J was extremely ill and the metabolic imbalances caused by her acidotic state were bound to affect her mental and emotional functioning. It appears from the evidence, that staff did not understand how ill Miss J was or how this was likely to make her feel. Although this was unlikely to have affected the outcome in this case it would be useful for staff to reflect upon this episode of care as part of developing professional practice.
81. We note that Miss J's care was allocated to an agency staff nurse with no formal training or qualifications in ITU. We accept the reasons for allocating her care to a female member of staff in view of Miss J's preference and note that the charge nurse, in his evidence, has stated that he maintained a watching brief over her care. It was not however, entirely appropriate for such an ill patient to be allocated to a less experienced member of staff. There is no evidence in the notes that her care was being supervised or reviewed. It must be incumbent on the nurse in charge of the unit to be aware particularly of any patient whose condition is giving rise to concern.
Conclusion
82. We are satisfied that the care provided was adequate, save for the proviso in paragraph 79 above. Nursing staff must be proactive in supervising the care of sick/deteriorating patients under their care.
Recommendation
83. We recommend that the nurse manager take steps to address the shortfalls which we have identified, including through staff training, to improve the standard of care.
Final conclusion
84. We were asked to consider whether the care and treatment provided by the Trust to Miss J was inadequate. In providing our advice, we have had two considerations in mind: first, whether the standard of care was that which might reasonably be expected at a District General Hospital; and second, whether there were any useful lessons to be learnt in terms of 'best' practice for the future benefit of patients. We have concluded that the consultant was in error in making a diagnosis of MODY without first excluding slow onset type l. Miss J's management in the diabetic clinic was adequate, and would not have been different even had the correct diagnosis been made, but the education given to her would have varied. This case highlights the extreme importance of ensuring that everyone involved in diabetic care, patients, doctors, GPs and nurses are aware that it is essential to test blood glucose more frequently during an intercurrent illness, regardless of the type or severity of diabetes.
85. Following admission to hospital we have concluded that the care and treatment provided was flawed. Cerebral oedema is a complex condition and the direct causes are not understood. Cerebral oedema in association with diabetic ketoacidosis in adults is a comparatively rare but recognised condition. Once it had developed the Trust did not mismanage her care to any significant extent. We do think, however, that the presence of hyponatraemia in association with severe diabetic ketoacidosis should have ensured that Miss J's care was closely controlled by a consultant because of the increased neurological risk. We therefore conclude that her care should have been better managed although we cannot say whether this would have altered the outcome.
(b) the report of the independent review panel inappropriately included a statement that Miss J was in denial of her diabetes.
Denial
Context
86. Miss J was advised on the management of her diabetes by the first consultant and diabetes nurse specialist. In the week or so before her death, Miss J was unwell and became increasingly ill, eventually requiring emergency hospitalisation and intensive care. Near the beginning of that period, she saw her own and other doctors several times before eventually being admitted to hospital. Miss J's care at the Trust was considered by an independent review (IR) panel under the provisions of the national NHS complaints procedure.
87. The first consultant in his evidence to the IR panel noted that Miss J's "diabetic control had always been extremely good", but said also that "he realises that he did not understand [Miss J's] psychological makeup and that she was probably frightened of becoming insulin dependent, and perhaps went into some kind of denial when she realised her blood sugar was rising".
88.The IR panel's clinical assessors said in their report that "denial" by Miss J could be implicated in what happened to her. In their view, "Denial of diabetes and the need for insulin therapy is a common feature .... particularly in those young people of the patient's age.".
89. It is the assessors' view that in the special atmosphere created by the diagnosis of MODY, a "too relaxed view" was engendered in many people and "the understandable fear of insulin on behalf of the patient must also have contributed to the delay which occurred in seeking and achieving appropriate management."
90. In his letter of 17 March one of the IR panel members said that both he and his colleague were "in agreement that there was a strong likelihood that such denial took place" and "possibly" led Miss J to fail to act in a way that would protect her from the adverse outcome of her illness.
91. These accounts are the context for the IR panel's presumption that "denial" played a role in the events that led to Miss J's death.
92. It also needs to be noted that in the period just prior to her hospitalisation and death, Miss J was looking forward to a special event, going away for a week- end, the first time with her fiance, to join his cousins at a holiday park to celebrate a birthday.
Self monitoring regime
93. We have two accounts of the self monitoring regime commended to Miss J. The diabetes nurse specialist stated that her advice to her would have been that for "values of blood sugar between 4 -7, she would be advised to undertake fasting blood sugars, whilst [for values from] 7 to 9 she should do random blood sugars. Generally, she should test her blood 4-5 times a week, more often if she went on holiday or was ill, or if anything else arose which disrupted her usual routine."
94. Miss J's brother reports that the diabetes nurse specialist at a Trust interview on 30 March 1999, confirmed that blood glucose monitoring "would be perfectly safe and acceptable every 4-5 days"; that if going on holiday Miss J "would only need to carry out a blood glucose test before she went away and after she returned", "to monitor more frequently when ill"an instruction refuted by the family; and that the diabetes nurse specialist confirmed that she had no "worries or concerns with [Miss J's] ability to monitor blood glucose levels indeed [Miss J's] diary entries can be confirmed as being accurate by the HbAlc readings recorded at the [hospital]". The diabetes nurse specialist "did not have any concerns about Miss J's condition psychologically or physiologically at any of their meetings."
95.These two accounts attributed to the diabetes nurse specialist lead to substantially different management regimes: one account requires monitoring every second day or more frequently, virtual daily monitoring when on holiday or when ill. The other account leads to a regime of monitoring about twice a week, no monitoring on holiday but more frequent monitoring when ill.
Miss J's Self-monitoring Records
96. From our examination of Miss J's written personal records (kept from 12.7.96 to 14.7.97), it is clear that at least up until about the time just before her admission to hospital, she was taking readings mainly at 3 to 4 day intervals.
97. A further notable feature is several extended intervals between some of the recordings, for instance, a 61 day interval between 8 January 1997 and 11 March 1997 when, apparently, no readings were taken. There were several other occasions when intervals of over 10 days occurred between readings.
98. In the 3 months or so before Miss J's admission to hospital, most of the readings were taken in intervals of 5 days or less, in line with the apparent advice. According to the record, the time of readings was varied, ranging between 8.OOam and 11.OOpm In examining Miss J's records over all the occasions of testing, it is noted that 50 of the 63 recordings (79%) of blood sugar level are at 7, with the remaining 13 (21%) either above or below that value. This degree of consistency is likely to be an artefact of the stepwise system used to grade blood glucose levels using this method. The fact that about 20% of her readings varied from 7 is evidence that she was discriminating the levels where these appeared unusual. There were some occasions where she also recorded a range.
Adherence to the Advice on Self Monitoring
99. From the evidence of Miss J's own records, it would appear that she was adhering fairly closely to the advice described in the submission provided by her brother, not that claimed by the diabetes nurse specialist during the Trust interview.
100. If Miss J was acting incorrectly - and her records are clear enough - it is of concern that no one seems to have called her attention, or that of her family, to this, or sought to educate and motivate her to adopt the correct practice.
101. There appears to be no account of any action taken to modify Miss J's recording procedures (for instance the extended periods in which there were no recordings) or to address the high reading in September 1996 which was followed by an extended interval when no reading was recorded. Presumably such action would have been called for if the regime claimed by the diabetes nurse specialist, in her evidence, was required for proper management.
102. As no such corrective action has been drawn to our attention, then it would be appropriate to assume that Miss J adhered fairly closely to the advice given, or what she had understood to have been advised, and had no reason to believe that she was behaving incorrectly.
103. The account given by Miss J's brother of the diabetes nurse specialist not having concerns about Miss J psychologically or physiologically is also consistent with a view that Miss J was regarded as managing her condition acceptably.
104. The degree of consistency in the levels recorded also appears to have been un-remarked.
105. The clinic must have had some awareness of her record because of the comment that her reading levels coincided with some of the more precise laboratory tests.
106. In sum, the impression given is that the unusual features in Miss J's personal records of her blood sugar monitoring went unchallenged.
Blood Testing on 17 July
107. Miss J's niece, with whom she had a special relationship, gave an account of events on the 17 July that has an important bearing on the contention by the IR panel that she was in "denial".
108. Although Miss J's written record ends on the 14 July, according to the statement of her 17-year-old niece, Miss J tested on two occasions on the morning of 17 July, with the niece having been asked to act as a time-keeper and to give her opinion of the level as Miss J, according to her niece, was unable to make it out precisely.
109 We cannot be certain of the levels Miss J obtained on both occasions of testing on the 17 July. Miss J's niece states that she observed both instances and, when asked to look at the second sample, she - the niece - could not determine the level and in her evidence reports Miss J's statement that it was between 7 and 9. There were 4 occasions before this when she entered a somewhat similar range rather than a single figure in her records. This is the level ("RBS around 7") recorded by her GP on the 21 July and is allegedly based on Miss J's report when he asked her on that occasion.
110 It is perhaps unusual that Miss J could not "read" the level despite two occasions of testing, especially as by this time she had had extensive experience of making readings, including at least two at 11 and others well below. It is possible that it was her physical state - she had been ill for 3 or more days by then and had been to see her GP the day before - impaired her functioning.
111. That the GP at the consultation on 16 July appears not to have initiated questions about her diabetes, may also have led her to believe that her diabetes and hence her blood sugar level were not of concern in relation to her then present condition. This in turn could be either because she did not know that she should report such findings or because doing so might have led to a disappointing change of plans. This latter possibility is dealt with shortly.
112. This behaviour - taking two readings and involving her niece in the process, of both timing to ensure accuracy and getting her to try and read the value, seems inconsistent with "denial". Further it is difficult to understand why, had she been in "denial" when visiting the GP the day before, she should then be involved in what appears to be a serious attempt the very next day to get a public and accurate reading.
113. However, the possibility of missing the important week-end celebration could be a reason for her not disclosing her high reading. The possibility that she might miss the week-end celebration did not, however, stop her going to the GP on the 16 July nor did it stop her visiting a doctor while at the holiday site.
114. It is also possible that, having been to her own doctor 3 days before with no apparent diabetes-related action, she did not attach any further significance to the reading, whatever its level and therefore did not raise it with the doctor at the holiday park who also appears to not have checked her blood sugar level.
115. In sum, it is noteworthy that Miss J did take action to communicate that she was ill in the period, that she continued testing in a semi-public manner, and we do not have strong evidence that she tried to hide the earlier reading of 11. Her records showed a level of 11 previously without any consequences and there are some grounds for suggesting that for reasons other than "denial", the level of 11 may not have had the significance for her that others might attach to such a reading.
Further developments
116. According to her fiance, Miss J did inform the doctor at the holiday park that she was a controlled MODY. Again, this is not suggestive of someone in "denial" unless one is trying to assert that she was deliberately setting out to throw him off track.
117. Nevertheless, the earlier high reading was apparently not brought to his attention. Again, one could argue that the lower reading as reported by her niece might have been taken as evidence by Miss J that this was not the problem. She had also seen her own doctor several days earlier and as noted above, he had not pursued anything in relation to the diabetes.
118. Miss J attended the celebration but by early the following morning (20 July) was so ill that her fiance brought her home.
119. She was seen again by a doctor on the 20 and 21 July. It was on the latter occasion that the note of "RBS about 7" is noted, consistent with the niece's report of the unrecorded reading 4 days earlier. Miss J was hospitalised on the following day.
Impact of Miss J's Deteriorating State
120. It is important to emphasise that Miss J was becoming progressively more ill over the last weeks of her life. It is possible too that the progress of her illness and the discomfort made her increasingly less able to focus on anything other than her on-going state. We know that by the night before her admission, she was in great distress and probably at least confused for periods, if not delirious.
121. Whether or not her mental state was sufficient for her to recollect a reading from a week previously, and associate it with her present state, especially if it had been a "one-off" like the previous "one-off", can be questioned.
122. It would, however, be surprising if her physical deterioration were not accompanied by some deterioration in her ability to function normally in terms of her mental capacities and emotions: she was not just physically ill, it was likely that she was also in great discomfort and mental distress.
The case made by the first consultant: introduction
123. It is not uncommon for diabetes services to have to deal with a number of individuals who do not adhere to the self-management regime prescribed by the service. This failure of adherence is sometimes attributed to "denial" that they are ill, or that they need to conform to the prescriptions for their healthcare. Most, if not all, practitioners in diabetes services, will be likely to have encountered individuals in whom "denial" is regarded as a major reason for poor adherence.
124. At the independent panel review, the first consultant introduced the possibility that Miss J had been in "denial", a view subsequently picked up by and endorsed by the experts assisting the panel. Such an allegation, if supportable, would of course affect the extent to which the first consultant was responsible for what transpired.
125. The first consultant, in his interview with the investigating officer, commented again about Miss J not wanting insulin and her relief at the diagnosis of MODY as part of the evidence for his claim that she was "in denial" about the transition in her diabetes to a state requiring insulin.
The case made by the first consultant: fear of insulin
126. A fear of acquiring insulin-dependent diabetes was introduced by the first consultant to explain why Miss J did not report the blood sugar level of 11 even though it was logged in her record card. We have no evidence that Miss J expressed such a fear.
127. The comments regarding her reaction to the MODY diagnosis were taken to be evidence consistent with later "denial". However, her relief reactions were both understandable and normal. We can safely assume that most people faced with the prospect of a diagnosis that would require life-long insulin injections would be very pleased at being told that they would not need to do so. Her 'relief at the diagnosis' is most likely a normal, understandable and presumably common reaction. Hence, this is unlikely to constitute a strong ground for implicating "denial".
The case made by the first consultant: transition denial
128. It appears accepted that Miss J was, over most of the period subsequent to the diagnosis, a "compliant" individual about whom there were no concerns regarding her diabetes management and monitoring.
129. The thrust of the first consultant's argument is that she in effect denied her transition from a form of diabetes that does not require insulin to a form that would. He expresses a concern that she did not contact him or his team. The first consultant's view was supported by the IR panel.
130 If we accept that Miss J was a compliant individual, and we also accept, as we must, that by the 16 July she was quite ill, it is not surprising that she did not go to the diabetes service. Her GP did not refer her and did not implicate her diabetes in the presentation; she was ill and likely to have been distressed; and from other accounts of her behaviour, it is likely that she would have attended clinic had she been instructed to do so.
131. Underlying all this, Miss J had not been warned that there was a risk of transition from MODY to insulin dependent diabetes and no questions had been raised as to her previous adherence to the advice she had been given. The evidence is that she did adhere to the self monitoring regime set for her.
Denial as an explanation
132 Miss J's case raises important specific as well as general issues in relation to the use of psychological concepts such as "denial" in attempts by clinicians to understand, account for and manage diabetes and other conditions requiring acceptance of the diagnosis and adherence to a prescribed management regime. The terms "compliance" and "denial" involve assumptions drawn from behaviour, a better term is "adherence" which describes the behaviour only.
133. We accept that it is common in diabetes services for clinicians to see what they regard as "denial". "Denial" is in fact an assumption or inference from the patient's action or inaction, that is, from behaviour, in relation to the diagnosis or some other aspect of care. It is however proffered as an explanation for observed behaviour, why the individual does not appear to take the diagnosis seriously or otherwise adhere to the care prescriptions advanced by the health care professionals who advise on management.
134. Nevertheless, the introduction of psychological explanations to account for what is believed to be a failure to accept the diagnosis or inadequate self management of illness, is a very serious matter. To begin with, it enables those in positions of power to attribute failure and blame to the patient, thereby exonerating themselves of responsibility for proper care. It colours attitudes towards the patient and, consequently, can adversely influence the motivation of healthcare staff and the adequacy and quality of care provided thereafter.
135. In particular, if "denial" is to be invoked to account for poor adherence, it necessitates consideration of factors in the patient, the illness, the clinicians involved in the patient's treatment and the patient's social environment. In sum, it is not something that should be invoked lightly and then, only on the basis of good evidence gleaned from systematic investigation and the due consideration of alternative explanations.
136. The more neutral term "adherence" is current in the literature on responsiveness to health care advice and is used to refer to the behaviour of individuals who follow the advice given to them by healthcare professionals. Adherence, as a term describing behaviour is different from other terms such as "denial" that are inferences from behaviour. It is recognised that "adherence" allows for variations in the extent of adherence.
137. There are well-established protocols to deal with some of the more common reasons for poor adherence, such as information booklets, special education programmes and the like. It is also wellrecognised that despite such standard good practices, there will be individuals who continue to cause concern, those who, while knowing what they should do, nevertheless fail to act as they should.
138. Under such circumstances, it is important to consider alternative explanations, among which might be something akin to some form of denial. However, given the complexity of human abilities, beliefs, motivations and life circumstances, and the complexity of the concept, "denial" should never be the sole initial conjecture.
139. In dealing with problems of adherence, healthcare staff must attempt to glean reliable evidence on which to base a judgment about the degree of adherence and then reach a decision as to whether or not, in cases of poor adherence, further action needs to be taken.
140. The nature of this action will depend on gaining some understanding of why the individual is behaving in the way they do. In many instances, sensitive, individual centred enquiry may lead to an identification of the reasons. Even when it is clear that the individual does not believe they have the illness, it is important to try and understand why.
141. On occasions, it takes some time for the mental realisation to become fully integrated into the individual's thinking about themselves and their daily lives. They have to create a new view of themselves and make the corresponding behavioural changes, something that takes some time for most people.
142 To sum up, evidence of clinically unacceptable levels of adherence should be a signal for further investigation rather than what can sometimes be a reflexive labelling of an individual as being in "denial".
Conclusion
143. In Miss J's case I have found no evidence that she was in denial of her illness. It was, in my view, unwise for the IR panel to have introduced this issue. Denial, as I have explained, is a complex concept that does not lend itself to reflexive labelling and should not be advanced lightly as an explanation of behaviour.
Appendix B - Glossary of Terms
| Technical term | Meaning |
|---|---|
| acidosis | an abnormally high concentration of acid in the blood |
| catheter (urinary) | a flexible tube inserted into the bladder to allow the removal of urine |
| cerebral oedema | swelling of the brain within the skull |
| CVP - central venous pressure | the pressure within the central veins returning blood to the heart |
| dehydration | loss or deficiency of water in body, tissues and blood |
| fluid balance chart | a chart which records and sums a patient's input and output of liquid on a daily basis |
| gastritis | inflammation of the lining of the stomach |
| glycosuria | glucose in the urine |
| HbAlc - haemoglobin | an indirect measure of the effectiveness of blood glucose ('sugar') control in past two weeks |
| hyperglycaemia | a high concentration of glucose in the blood |
| hypoglycaemia | a low concentration of glucose in the blood |
| hyponatraemia | a low concentration of sodium in blood, usually as a result of a relative excess of water but sometimes a result of salt loss |
| hypotension | low blood pressure |
| hypothyroid | reduced thyroid function |
| ketones | substances arising in metabolism when there is insufficiency of insulin |
| mass lesion | a solid abnormal appearance on scanning |
| pituitary apoplexy | collapse as a result of severe pituitary gland failure |
| subarachnoid haemorrhage | haemorrhage on the surface of the brain |
| subclavian | under the collar bone |
| tachycardia | rapid heart beat |
| type 1 diabetes | insulin dependent diabetes - requires treatment with insulin |
| type 2 diabetes | non insulin dependent diabetes - controlled with diet and/or tablets |
