A medical practice in the Surrey area

5. Miss A complains about the care provided to her father, Mr A, by a GP at the Practice. She complains the GP:

· failed to restart Mr A’s diabetic medication (metformin) and a blood thinner (edoxaban), after his discharge from Hospital A in July 2019 · incorrectly stopped his gliclazide medication on 15 November.

6. As a result of the lack of diabetic medication, Miss A said her father nearly died and spent two months in hospital. She says it took the hospital 17 days to declare him stable on insulin. During his hospital stay, she says Mr A contracted methicillin-resistant staphylococcus aureus (MRSA), suffered pressure sores, and lost many skills and flexibility in his joints which caused him pain.

7. Miss A says her father missed the comforts and activities from his care home during his hospital admission. She explains he lost his will to live, and his admission advanced the decline in his health. Miss A says he suffered distress in having to move into a different care home and given his dementia he found it difficult to make new friendships and became isolated. She explains her family felt stressed because of the failings by the GP and the effect this had on Mr A.

8. As an outcome, Miss A would like answers to her questions, acknowledgement of failings, and service improvements to minimise the risk of this happening to anyone else.

This is a summary of events to put the complaint into context.

9. On 14 December 2018, Mr A’s HbA1c was recorded as 45mmol/mol.

10. The Practice carried out a diabetic review in February 2019. It reduced Mr A’s dose of gliclazide from twice to once daily. Gliclazide medication treats Type 2 diabetes by increasing the amount of insulin the pancreas makes.

11. On 17 June, Mr A’s HbA1c was recorded as 44mmol/mol.

12. On 29 June, Hospital A (not part of this complaint) admitted Mr A because of repeated blood in his urine. During his admission, the Hospital A stopped his metformin and edoxaban medication. Mr A’s last dose of metformin medication was on 12 June.

13. Hospital A discharged Mr A on 2 July.

14. On 15 November, the GP reviewed Mr A’s medication with the Practice’s pharmacist. They agreed to stop gliclazide. The GP emailed Mr A’s care home to stop his gliclazide. The care home stopped this medication on the same day.

15. On 1 December, Hospital B (not part of this complaint) admitted Mr A with community-acquired pneumonia (CAP), hyperosmolar hyperglycaemic state (HHS), and acute kidney injury (AKI).

Failed to restart medication

19. Miss A complains the GP at the Practice failed to restart her father’s diabetic medication (metformin) and a blood thinner called edoxaban, after his discharge from Hospital A in July 2019. She says Hospital A had temporarily stopped these medications while Mr A was an inpatient in June and July 2019.

20. The medical evidence shows that Hospital A stopped metformin and edoxaban in June 2019, during Mr A’s admission, due to repeated blood in his urine. The discharge letter dated 2 July 2019, shows Hospital A had referred Mr A to the outpatient urology haematuria clinic for follow up. Hospital A said it would arrange an appointment within three to four weeks of his discharge. Hospital A asked the Practice to chase this referral up. The Practice explained that, as Mr A had a history of prostate cancer, and was on ongoing hormone management, it was concerned about restarting edoxaban medication until the urology review happened. However, it said the urology review did not happen.

21. The Practice said the blood test in December 2018 showed excellent diabetic control. Mr A had a diabetic review in February 2019 and the GP reduced the dose of gliclazide to once daily. It said Mr A had further blood tests in June which showed even better control of his diabetes, and he was scheduled for repeat blood tests in December.

22. To help with our consideration of this complaint, we sought advice from a GP adviser (our adviser).

23. The relevant guidance is the General Medical Council (GMC): Good medical practice, paragraph 15. This says doctors should: ‘Adequately assess the patient’s conditions, taking account of their history.’

24. The medical evidence shows Hospital A stopped the metformin and edoxaban medication in June/July 2019, during Mr A’s admission. It stopped the edoxaban due to blood found in his urine, and the metformin was stopped due to raised lactose levels in his blood.

25. Hospital A’s discharge report, dated 2 July, asked the GP to review Mr A’s metformin and edoxaban medication. Our GP adviser said that although Hospital A did not suggest a set time frame to review the medication, there is no evidence in the records the GP reviewed Mr A’s medication.

26. The Practice has provided evidence and an explanation for why it did not restart Mr A’s edoxaban medication following his discharge from Hospital A in July. Before restarting this type of medication, our GP adviser said the GP has to weigh up the risks of further blood in the urine. It would therefore be good clinical practice for the GP to wait for the advice from the urologist before reviewing and restarting edoxaban. Therefore, the explanation provided by the GP is in line with GMC: Good medical practice, which says: ‘Provide effective treatments based on the best available evidence.’

27. However, the decision by the GP to wait for the urologist’s advice before reviewing the edoxaban medication should have specifically been recorded in Mr A’s medical notes, in line with GMC: Good medical practice, which says: ‘Clinical records should include…b) the decisions made and actions agreed.’

28. The GP should have reviewed Mr A’s medication to decide whether he should have restarted the metformin and edoxaban after his discharge from hospital, and documented these decisions in Mr A’s records, in line with GMC guidance. There is no evidence the GP did this.

29. We consider the Practice should have reviewed Mr A’s medication and documented these decisions in Mr A’s records, to be in line with GMC guidance. This is a failing on the part of the Practice. We will consider the impact this had on Mr A and his family later within this report.

Stopped gliclazide medication

30. Miss A complains the GP incorrectly stopped her father’s gliclazide medication on 15 November 2019, without carrying out any tests before or after, to monitor the effect of its withdrawal. She says it did not offer, or prescribe, any replacement medication.

31. The Practice said the stopping of gliclazide was the correct course of action and in line with both NICE guidelines and Diabetes UK guidelines. It said its clinical pharmacist reviewed Mr A’s medication and recommended stopping the gliclazide due to the risk of hypoglycaemia. Hypoglycaemia is where the level of sugar in the blood drops too low. It said the risk of hypoglycaemia outweighed the risk of worsening diabetic control.

32. The Practice said, in February 2019, it had already reduced Mr A’s gliclazide dose from twice daily since his ‘control had been excellent.’ It said Mr A’s HbA1c was 45 in December 2018, and after reducing this dose to once daily, his diabetic control seemed to have improved to 44. It said it felt confident making this change as it was in line with NICE guidelines. It stopped Mr A’s gliclazide medication on 15 November and emailed his care home to inform them to make that change.

33. Miss A also raised a complaint with NHS England (NHSE) concerning the actions of the Practice. As well NHSE asking the Practice to look into her complaint, NHSE also investigated her concerns. After looking at Mr A’s medical records, NHSE said the GP followed NICE guidance: Type 2 diabetes management in Adults [NG28], December 2015 and NICE guidelines on Type 2 diabetes, September 2019. On reviewing the NICE guidance, it said the advice from the diabetic prescribing nurse and Practice’s pharmacist, makes it clear why the decision to stop gliclazide was made by the GP and documented in Mr As’ medical records. The GP’s intention was to offer and provide Mr A the best care to manage his Type 2 diabetes and reduce the risk of prescribing multiple medications that could harm patients.

34. Since Miss A’s complaint, the Practice explained the GP has discussed this case with the Head of Endocrinology Service at Hospital C (this organisation is not part of this complaint). The Practice said the Head of Endocrinology felt the stopping of gliclazide was the correct course of action. It said metformin could have caused additional problems when Mr A developed a chest infection, such as causing a high lactate again, and subsequent acidosis. Acidosis is a condition where there is too much acid in the body fluids. It said the advice was that Mr A had a proven pneumonia requiring intravenous (IV) antibiotics and this would have been the source of the problem, rather than his medication being stopped.

35. To help us consider this complaint, as well as seeking advice from a GP we also sought the advice from an endocrinologist.

36. NICE: Type 2 diabetes in adults: management NG 28 says:

‘Measurement

1.6.1 In adults with type 2 diabetes, measure HbA1c levels at: · 3 to 6-monthly intervals (tailored to individual needs), until the HbA1c is stable on unchanging therapy · 6-monthly intervals once the HbA1c level and blood glucose lowering therapy are stable. [2015]

1.6.2 Use methods to measure HbA1c that have been calibrated according to International Federation of Clinical Chemistry (IFCC) standardisation. [2015]

1.6.3 If HbA1c monitoring is invalid because of disturbed erythrocyte turnover or abnormal haemoglobin type, estimate trends in blood glucose control using one of the following: · quality-controlled plasma glucose profiles · total glycated haemoglobin estimation (if abnormal haemoglobins) · fructosamine estimation. [2015]’

37. Based on NICE guidelines: Type 2 diabetes in adults, our endocrinologist adviser said there was a change in therapy when the hospital stopped metformin in June 2019. The June HbA1c test result, five days after the metformin had stopped, would not have reflected this recent change. This is because the HbA1c provides an estimate of average glucose levels over approximately the preceding 120 days (assuming normal red blood cell lifespan) so is not informative about acute changes. Between the stopping of metformin in June 2019, and the stopping of gliclazide in November, Mr A’s glycaemic control could have changed due to discontinuing the metformin and/or other factors. These could include changes in oral intake, or changes in renal function for example.

38. Our GP adviser said the medical records show the pharmacist noted Mr A’s last HbA1c blood test in June 2019 was on the low side and advised the GP to stop gliclazide. However, there is no evidence to show the pharmacist gave information to the GP about the metformin that had stopped in June/July 2019 and not been restarted.

39. While the GP acted on the information they had been given by the pharmacist, our GP adviser said the GP had a responsibility to check Mr A’s medical records before stopping the gliclazide medication, in addition to assessing the impact of stopping the metformin, in line with NICE guidelines [NG28], Type 2 diabetes in adults: management. Had they checked the medical records, the GP would have realised metformin had been stopped in June 2019 and had not been restarted.

40. It is understandable how the events took place, as the GP was not given all the relevant information. However, our GP adviser said the GP does carry clinical responsibility for stopping the medication and should have checked the records more thoroughly before they stopped gliclazide.

41. As Mr A was on a high dose of metformin prior to Hospital A stopping this in June 2019, our advisers agreed the Practice should have taken this into account and carried out a further HbA1c blood test before stopping the gliclazide, in line with NICE guidelines. Or, at least put in a place a plan (with the care home) to monitor the capillary blood glucose levels short-term to assess the impact of the change in medication, in line with good clinical practice. This would have shown the effects of the withdrawal of metformin and informed the Practice whether stopping gliclazide was appropriate. This is a failing.

Impact

42. Where there are failings, we consider the impact claimed to have been suffered as a result. We also consider the actions taken by the organisation in response.

43. We sought advice from our endocrinologist adviser to consider the clinical impact of stopping Mr A’s gliclazide and not restarting metformin, and whether this led to Mr A being admitted to hospital in December 2019. We also asked our GP adviser to consider the clinical impact of not restarting edoxaban.

44. The medical records show that Mr A was admitted to hospital on 1 December 2019 with CAP, HHS, and AKI.

45. Our endocrinologist adviser said Mr A’s advanced age, frailty, and comorbidities (diabetes, prostate cancer and dementia) would increase his susceptibility to infections. Comorbidity means more than one disease or condition is present in the same person at the same time. These same risk factors would increase the likelihood of him needing hospital admission to treat the pneumonia, independent of his glycaemic control (blood glucose levels). An acute severe illness such as an infection, for example CAP, will increase the risk of hyperglycaemia and AKI in frail older people with diabetes.

46. Joint British Diabetes Societies (JBDS): Inpatient Care of the Frail Older Adult with Diabetes, October 2019, says: ‘Hyperglycaemia in hospital settings: Hyperglycaemia is a common problem in hospital settings in people with diabetes. In frail older people with diabetes factors such as infections, acute illness and/or the physiological stress associated with surgery, can place older people at increased risk of hyperglycaemia.’

47. Taking into consideration JBDS guidelines, Mr A’s comorbidities (and risk of infections) and clinical experience, our endocrinologist adviser said a CAP can trigger HHS and AKI in frail elderly people with Type 2 diabetes, even if their diabetic control is usually good. On balance, it is more than likely Mr A would have needed to be admitted to hospital because of CAP, even if he did not have HHS or AKI.

48. To help address the question of the relative contribution stopping gliclazide and metformin, and the CAP had to the HHS, a HbA1c checked on admission to hospital in December 2019 would have been helpful, but not conclusive for the following reasons:

· The gliclazide had only been discontinued two weeks prior to admission, so the HbA1c would not reflect that change, but would give an idea of the preceding glycaemic control for a period of two to three months.

· Mr A had haematuria (blood in the urine) and anaemia (low red blood cell count), so it is possible the HbA1c might not accurately reflect his blood glucose control.

49. Our endocrinologist adviser has reviewed the blood test results during Mr A’s hospital admission in December 2019. A plan to check the HbA1c is mentioned in the hospital notes, but there is no evidence the hospital carried out a HbA1c.

50. Without a HbA1c result, carried out before the gliclazide was stopped (and/or on admission to hospital), or results of capillary blood glucose monitoring following the discontinuation of the gliclazide prior to hospital admission, it is not possible to say whether stopping gliclazide and metformin contributed to Mr A’s risk of HHS.

51. Based on the evidence, it is not possible to say whether stopping gliclazide (and not restarting metformin) contributed to Mr A’s risk of CAP, with HHS and AKI, which resulted in Mr A being admitted to hospital in December 2019. Based on JBDS guidelines and clinical advice, Mr A was at risk of infections and CAP could have triggered his HHS and AKI, independent of his diabetic management. He may have acquired CAP irrespective of the stopping of gliclazide and metformin. It is therefore more than likely Mr A would have needed to be admitted to hospital in December 2019 for pneumonia, even if he did not suffer from HHS and AKI.

52. With regards to not restarting edoxaban, our GP adviser said edoxaban medication does not have any other benefit, for example it does not treat other symptoms or make someone feel better. Mr A would not have felt any different by not being on edoxaban, but he did have a higher risk of having a stroke. As Mr A did not suffer from a stroke, there is no clinical impact to not restarting this medication.

53. We understand Mr A and his family would have been upset at finding out the Practice had not reviewed and/or restarted edoxaban and metformin medication, as well as stopping gliclazide without carrying out a HbA1c blood test first. However, it is not possible to directly link the clinical effects Miss A says her father suffered to the actions of the Practice.

54. We now consider what actions the Practice has taken as a result of this complaint.

Outcome

55. Based on our Principles for Remedy, where there have been failings leading to an injustice, the public organisation should try to offer a remedy that returns the complainant to the position they would have been in, if the failings had not happened.

56. An appropriate range of remedies will include:

· an apology, explanation, and acknowledgement of responsibility · remedial action, i.e., service improvements to minimise the risk of this happening again · financial redress.

57. As an outcome, Miss A would like answers to her questions, acknowledgement of failings, and service improvements to minimise the risk of this happening to anyone else.

58. We consider our investigation provides answers to Miss A’s complaint about the Practice and resolves part of her desired outcomes.

59. We are pleased to see the Practice has apologised for the distress caused in the lead up to Mr A’s admittance to hospital. However, it has not acknowledged to Miss A that it did not review Mr A’s medication and did not document these decisions in his records. It has also not acknowledged it stopped Mr A’s gliclazide medication without first carrying out a HbA1c test, or at least putting in a place a plan to monitor Mr A’s capillary blood glucose levels short-term to assess the impact of the change in medication.

60. The Practice has not apologised for the upset caused to Mr A and his family when they found out.

61. As a result of Miss A’s complaint to the Practice, it has:

· discussed the case at the Practice’s clinician meeting and with Mr A’s care home · fed back to the Practice’s pharmacist, Practice nurse and Tier 3 Specialist Diabetic nurse · the GP has reflected on their management of patients with diabetes to ensure they are up to date with current guidelines · discussed in the GP’s appraisal.

62. We are pleased to see the Practice has reflected on this complaint. However, as it has not acknowledged the failings identified, we cannot see evidence the Practice has put in place service improvements to minimise the risk of these failings happening again. Therefore, we do not consider the Practice’s actions go far enough. We will consider this within our recommendations.

1. We acknowledge this was, and continues to be, an upsetting time for Miss A and her family. Our final decision is not made without recognition of the distressing circumstances around these events.

2. We partly uphold Miss A’s complaint about a medical practice in the Surrey area (the Practice). This is because we have found failings in the following areas:

· the Practice did not review Mr A’s medication to decide whether he should have restarted the metformin and edoxaban after his discharge from hospital or document these decisions in his records. Metformin is a diabetic medication and edoxaban is a blood thinning medication.

· the Practice did not carry out a haemoglobin A1c (HbA1c) test before stopping Mr A’s gliclazide medication in November 2019. HbA1c is a person’s average blood glucose (sugar) level.

3. We acknowledge Mr A and his family were distressed in finding out the Practice had not reviewed his edoxaban and metformin medication, as well as stopping his gliclazide without carrying out a HbA1c blood test first. However, we have not been able to link other areas of the claimed impact to the indicated failings. Based on the evidence, it is not possible to say whether the stopping of gliclazide (and not restarting metformin) contributed to Mr A’s risk of community-acquired pneumonia (CAP), hyperosmolar hyperglycaemic state (HHS) and acute kidney injury (AKI), which resulted in Mr A being admitted to hospital in December 2019. HHS occurs in people with Type 2 diabetes, who experience very high blood glucose levels. We cannot see evidence of any clinical impact on Mr A not restarting edoxaban.

4. We recommend the Practice:

· write to Miss A to acknowledge the failings we have identified and apologise for the impact these had on Mr A and his family.

· it should explain the service changes it has made, or will make, to prevent similar failings happening again.

63. We acknowledge the difficult circumstances around this complaint, and the impact this has had on Mr A and his family. We hope our considerations provide some assurances of the care provided by the Practice and that, where appropriate, learning and improvement will be achieved as a result of this complaint. We recognise the Practice has reflected on this complaint and has already taken some action to put matters right. However, we consider more action should be taken.

64. In considering our recommendations, we have referred to our Principles for Remedy. These state that where poor service or maladministration has led to injustice or hardship, the organisation responsible should take steps to put things right.

65. Within four weeks of the date of our final report, the Practice should write to Miss A to acknowledge the following failings:

· failure to review Mr A’s medication to decide whether he should have restarted the metformin and edoxaban after his discharge from hospital, and failure to document these decisions in his records · failure to carrying out a HbA1c test before stopping Mr A’s gliclazide medication.

66. The Practice should apologise for the distress caused to Mr A and his family. A copy of this letter should also be sent to us.

67. Our Principles say that public organisations should seek continuous improvement and should use the lessons learnt from complaints to ensure they do not repeat maladministration or poor service. In line with this, we recommend the Practice should explain what action it will take (or has taken) to minimise the risk of these failings happening again.

68. The Practice should send a copy of this to our office and to Miss A within two months of the date of our final report.