12. In reaching our findings, we have assessed Mrs O’s complaint in two parts. In part one, we will look at whether the Trust followed the post-discharge care plan (complaint: a). In part two, we will look at the Trust’s management of Miss O’s potassium levels (complaints: b, c, d & e).
Part 1 – the complaint about the Trust’s failure to follow the post-discharge community care plan 13. Mrs O complains the Trust failed to follow the post-discharge community care plan that was put in place at the MDT meeting at Newbridge House on 13 June 2017.
14. The Code says nurses should accurately identify, observe and assess signs of normal or worsening physical and mental health in the person receiving care. NICE transition guidance says that a follow-up appointment should happen within seven days of discharge from inpatient mental health care for non-suicidal patients. NICE transition guidance also says patients are considered high risk after discharge so engaging with a community treatment package should be prioritised to reduce the risk.
15. Our advice is that good clinical care would mean that a post-discharge community care plan for Miss O would have needed to be informed by the evidence base for her illness, any assessed risks to her, her preference and willingness to engage with services and the resources available to the community team.
16. It would have needed to address both her mental and physical health needs. This would have included blood test and ECG monitoring (at least monthly) along with weight measurement and sitting and standing blood pressure monitoring. Moreover, time lost with family and friends to professional interventions should be considered and balanced with any assessed risks.
17. Records show the discharge plan agreed by Newbury House and the Trust for Miss O was:
• the Reaching Out Service to see Miss O twice weekly until enhanced cognitive behavioural therapy (CBT-E, a psychological treatment for people with eating disorders) two times per week started. The first appointment was arranged for 21 July 2017 • a specialist eating disorder nurse to see Miss O weekly
• systemic family therapy to be provided weekly • a first appointment with a dietitian on 18 August 2017 (frequency of appointments are not stated).
• review by a consultant psychiatrist (frequency of reviews are not stated).
In addition to these interventions, there was a referral to social care and an agreement there would be input from social care early help twice a week.
18. NICE guidance says family therapy for bulimia nervosa is the first line treatment for children and young people, with CBT-E recommended if family therapy is unacceptable, contraindicated or ineffective. Both interventions were on Miss O’s discharge plan.
19. We have seen evidence the Reaching Out service’s senior mental health practitioner (the Nurse) saw Miss O on 21 July 2017 as planned. This fulfilled the requirement for a follow-up appointment within seven days of discharge from inpatient mental health care for non-suicidal patients in line with NICE transition guidance. Miss O’s weight and sitting and standing blood pressure were checked at this appointment.
20. Miss O travelled abroad to Brussels with her father from 23 to 27 July and on her return to the UK went camping with him for four days on the south coast. She then went abroad with her mother to Barcelona from 9 to 16 August. Miss O was due to see her psychiatrist on 11 August, but this was postponed as she was out of the country. Her care plan was reviewed and updated on 15 August by the Nurse who noted that the agreed care plan had not been put in place due to Miss O being abroad. The plan was for Miss O to engage with the care plan on her return from Barcelona.
21. Miss O was seen by a specialist eating disorder nurse on 17 August when her sitting and standing blood pressures were checked. There was also a plan for blood tests to be done the same day, and Miss O’s potassium levels were assessed as normal. The next day she was seen by a dietitian as planned. The Reaching Out Service saw Miss O and her mother on 21 August and considered inpatient hospital care might be potentially needed. Notes from the meeting said Miss O and her family had disengaged from family therapy although Mrs O told us this was incorrect. She said she had arranged voluntary family-therapy input through an external charity, and a session took place on 24 August.
22. Miss O was reviewed by her psychiatrist, specialist eating disorder nurse, family therapist and the Nurse on 22 August at the MDT meeting. On 24 August, the Reaching Out Service saw Miss O again. Her sitting and standing blood pressure was measured at this appointment and her weight was checked.
23. We have seen evidence to show Miss O’s care plan was appropriate in the circumstances as it included most of what our adviser explained we would expect to see. What it did not formally include was monthly blood testing and ECG monitoring along with weight measurement and sitting and standing blood pressure monitoring. This was a crucial omission we will consider later.
24. We can also see the Trust followed the care plan to some extent. However, it would have been impossible for the Trust to follow the care plan in every respect as Miss O was on holiday abroad and in the UK for a lot of the post discharge period, which the Trust acknowledged on 15 August.
25. NICE transition guidance says patients are considered high risk after discharge so engaging with a community treatment package should be prioritised to reduce the risk. Our adviser told us that in practice this means there should have been discussion between the CAMHS team and Miss O and her parents about the risks posed to her recovery if she was out of the country and unable to engage in her community treatment. There should also have been contingency planning for any anticipated difficulties Miss O might encounter if it was not possible to cancel the trips abroad.
26. We have seen no documented evidence of such a discussion happening or contingency planning being put in place. Mrs O told us she did raise her concerns to CAMHS about the effect of the trips at the MDT meeting on 13 June 2017. She said at no point did CAMHS raise any concerns. We can see evidence the Trust knew about the upcoming trips at the MDT on 13 June. However, there is nothing in the notes about Mrs O’s concerns about this.
27. We have found no reason to disbelieve Mrs O’s account of events. On the contrary, it seems likely that any concerned parent would raise this issue given the circumstances. The fact the Trust did not raise an objection to Miss O’s travel plans or put in place contingency planning shows it was not taking sufficient steps to prioritise her community care. This was not in line with NICE transition guidance.
Summary Part 1
28. Based on the evidence, we find the Trust did follow the care plan in many respects. However, it was hindered by Miss O’s trips abroad (and in the UK) from conducting the care plan in its entirety. We are not critical of the Trust for this, but we would have expected it to communicate any concerns to Miss O and her mother about the impact going abroad would have on its ability to deliver the care plan in full. We have seen no evidence of such a conversation.
29. In fact, Mrs O told us that the Trust suggested the trips abroad would not cause any issues. That it did this in the absence of any contingency planning in the event of Miss O having difficulties abroad is extremely concerning. In this aspect of care, the Trust did not act in line with the relevant standards. The failings we have identified are so serious to amount to service failure.
Part 2 - The management of Miss O’s potassium levels
30. Mrs O complains about the way the Trust managed Miss O’s potassium levels in the community. Her concerns are the Trust failed:
• to review Miss O on her return to the UK following an admission to hospital abroad for low potassium on 24 July 2017
• to coordinate with Miss O’s GP to ensure her potassium levels were appropriately monitored
• to obtain or consider crucial blood test results at the MDT meeting of 22 August 2017 (and at a home visit on 24 August)
• to produce a management plan to minimise the risks of low potassium at the MDT meeting on 22 August or adequately consider an inpatient admission including getting a second opinion about this.
• to assess Miss O’s mental capacity to make decisions about the risks and management of her potassium levels.
31. As we described above, there was a crucial omission in the post-discharge care plan in that it did not formally include monthly blood testing (including potassium levels) and ECG monitoring. With this in mind, we consider the first of Mrs O’s complaints about her daughter’s potassium management – the failure to review Miss O following her hospitalisation in Brussels for low potassium levels.
32. The Code says nurses should accurately identify, observe and assess signs of normal or worsening physical and mental health in the person receiving care. It also says nurses should share information to identify and reduce risk.
33. During Miss O’s trip to Brussels, she was hospitalised and tests showed she had low potassium levels. Miss O’s mother and father separately emailed the Nurse on 26 July to tell her of their daughter’s admission to hospital for low potassium levels.
34. We have been advised that Miss O would have been rated red (the highest level of risk) according to the RCP guidance due to her low potassium levels in Brussels (and later in Barcelona where in addition to low potassium levels an ECG also showed she had an abnormal heart rate) and uncontrolled vomiting. Our advice is had these episodes happened in the UK, admission to a paediatric ward for assessment should have been actioned.
35. Miss O’s hospitalisation in Brussels should have alerted the Nurse to her increased risk and this information should have been shared with the wider MDT to consider formalising blood testing and ECG monitoring and their frequency in the care plan. A discussion about the advisability of going on further trips abroad and within the UK with Miss O should also have been discussed with her parents.
36. The records show the Nurse did talk to Miss O’s father about her hospitalisation in Brussels and advised him to limit her physical activity levels and to adhere to her meal plans. However, there does not seem to have been a discussion about the potential detrimental impact of going away on holiday could have on Miss O’s engagement with the care plan.
37. There also appears to have been a lack of recognition of the seriousness of Miss O’s low potassium levels by the Nurse because there is no evidence she shared this information with the MDT to consider. The result of this was the care plan was not amended to address this increased health risk, either immediately after the trip to Brussels, or following Miss O’s hospitalisation in Barcelona when the care plan was reviewed on 15 August. We find this was not in line with the Code.
38. The NICE eating disorder guidance and RCP guidance do not specify how frequently abnormal potassium levels and ECGs should be assessed, but we have seen evidence that blood tests for potassium levels did take place approximately once a week after Miss O’s return from Brussels. Our advice is this was an appropriate frequency, although the records show this was done in an ad-hoc way due to lack of a formalised plan that should have had a designated person responsible for checking the results and for referring for treatment if necessary.
39. We now turn to Mrs O’s complaint that the Trust failed to coordinate with Miss O’s GP to ensure her potassium levels were appropriately monitored. Good Medical Practice says doctors must refer a patient to another practitioner when this serves the patient’s needs. The Code says nurses must share information to identify and reduce risk.
40. In 2017, most mental health services were relying on GPs to arrange blood testing. We have been advised that arrangements for blood testing and other tests, such as ECGs, their frequency and the service responsible for doing them should have been documented in the care plan. The care plan, or a letter summarising the care plan, should then have been sent to Miss O’s GP.
41. Any changes to the care plan, should also have been communicated to the GP. Usually, the health care team ordering the test is responsible for ordering the test and acting on any abnormal results.
42. We have not seen any evidence of communication between the Trust and Miss O’s GP about blood (or ECG) testing. Nor have we seen any evidence that a care plan was shared with the GP denying the GP important oversight of Miss O’s care. Even if it had been, the lack of a formalised plan for blood and ECG testing would not have given the GP vital information about this crucial aspect of Miss O’s care. Because of this, we find the Trust did not act in line with Good Medical Practice and the Code.
43. We have carefully looked at Mrs O’s complaints about the latter period of Miss O’s care. In relation to her complaint that the Trust failed to obtain or consider crucial blood test results at the MDT meeting of 22 August 2017 (and on a home visit 24 August) and failed to produce a plan to minimise the risks of low potassium, we have taken account of the Code and Good Medical Practice.
44. The Code says nurses should take measures to reduce as far as possible, the likelihood of mistakes, near misses, harm and the effect of harm if it takes place. Good Medical Practice says doctors must adequately assess the patient’s condition. We have seen evidence Miss O’s blood was tested on 17 August after she was seen by a specialist eating disorder nurse. The results showed her potassium level was normal, but her serum bicarbonate was raised.
45. Our advice is the Trust should have obtained the blood test results of 17 August in advance of the MDT meeting of 22 August to be fully informed of the status of Miss O’s health. If this had been done, the MDT would have seen that although Miss O’s potassium level was normal, her raised serum bicarbonate level indicated continued purging. This should have led the Trust to put in place a revised care plan outlining the frequency of blood test monitoring and who would be responsible for it.
46. The evidence shows the Trust received Miss O’s blood test results of 17 August and these were checked the following day. It is unclear why the MDT on 22 August did not have the results or who was responsible for sharing them as there is only an incomplete record of the MDT meeting in the clinical notes. What is clear is the Trust did not have crucial information about Miss O’s health that allowed it to accurately assess her risks and amend her care plan.
47. This meant that although a plan was made to check Miss O’s potassium levels on 22 August, there is no evidence of a formal schedule of blood testing to manage Miss O’s potassium levels after that, or previously, as outlined above. Moreover, we do not know if the blood test requested by the MDT was done. This is because there is no record of it in the notes or evidence that Miss O was asked at her review with the Nurse on 24 August if the blood test had been done. In fact, Miss O’s last documented potassium level was on the 17 August – nine days before she died. We find this was not in line with the Code and Good Medical Practice.
48. In terms of whether the Trust should have considered inpatient admission for Miss O or obtained a second opinion about this, we have taken account of Good Medical Practice, which says doctors must refer a patient to another practitioner when this serves the patient’s needs. Our adviser told us the Trust should have considered the risk of Miss O’s continued purging and whether she was consenting to follow the care plan in assessing whether she needed inpatient care.
49. If she had agreed to follow the care plan, but then did not and refused inpatient care, Miss O could have been assessed under the Mental Health Act 1983 and detained for inpatient care. If the risks to Miss O had been assessed as too high for community treatment and she agreed to inpatient treatment, she could have been admitted to hospital voluntarily.
50. We have seen evidence the Trust did consider making a referral for inpatient care for Miss O (at Mrs O’s request) when she was reviewed on 21 August. The plan was to discuss the referral at the MDT meeting the following day. We know from the Serious Incident investigation undertaken after Miss O’s death that the MDT decided not to refer Miss O for inpatient care but to continue to implement the care plan. Our advice is this would have been a reasonable decision if Miss O consented to the care plan because it was the least restrictive way to care for her.
51. We have not seen anything to suggest Miss O did not consent to the care plan. However, it is concerning we do not have the complete notes of the MDT meeting on 22 August. We consequently do not know the rationale for continuing with community care or whether getting a second opinion about this was discussed.
52. The MDT was also not aware of the blood test results of 17 August, which indicated Miss O was continuing to purge so any decision on admission would not have been made with reference to all the facts. We are critical of the Trust for not being able to provide the complete record of this meeting. because it means we are unable to say whether doctors acted in line with Good Medical Practice or not.
53. Finally, we assess Mrs O’s complaint that the Trust failed to assess Miss O’s capacity to make decisions about the risks and management of her potassium levels. The Act is a law designed to protect and empower people over the age of 16 who may lack the mental capacity to make decisions on their care. There are five key principles:
• presumption of capacity • support to make decisions • the ability to make unwise decisions • the best interests of the patient • use of the least restrictive option (of care).
54. A person is considered is considered to lack capacity for a specific decision if they are unable to do at least one of the following four tests:
• understand the information relevant to the decision, including the consequences • retain the information long enough to make the decision • use or weigh the information as part of the decision-making process • communicate their decision in any recognisable way.
55. Our advice is that capacity assessments are decision specific and are made at a point in time. A person can therefore lack capacity about an aspect of their care at one specific point in time and have capacity about another aspect of care at the same time, but this can change hour to hour or day to day and issue by issue.
56. Our advice is the Trust should have documented whether Miss O was consenting to treatment, and if she was not, assessed and recorded her mental capacity to make decisions about her care and treatment. We have seen no evidence Miss O was refusing her consent for treatment. In fact, the evidence shows she was engaging with the care plan.
57. For example, the dietitian notes of 18 August show Miss O consented to have a blood test and was engaged in the process. Additionally, there is some evidence Miss O agreed with the plan not to admit her to hospital at the MDT meeting on 22 August. We have not found any evidence Miss O refused her consent for care. Because of this there was no reason to think she was unable to meet at least one of the four tests. As such a mental capacity assessment was not required. This was in line with the Act although we note Mrs O’s disagreement with this.
Summary – Issue 2
58. In summary, we find there was a lack of recognition of the seriousness of Miss O’s low potassium levels following her hospitalisations in Brussels and later in Barcelona. Information about the increase in risk to Miss O’s health following these trips should have been shared with the MDT with a view to amending the care plan to formalise blood testing, and following the trip to Barcelona, ECG testing as well. This did not happen even when the care plan was reviewed on 15 August. There was also a further missed opportunity to discuss the potential negative impact of going to Barcelona with Miss O and her parents when she returned from Brussels.
59. There is no evidence of any communication or coordination between the Trust and Miss O’s GP about blood (or ECG) testing, or that her care plan (and its amendments) were shared. There is also no evidence any ECGs were done in the post-discharge period. Regular blood testing was done up until 17 August. However, the results of this last blood test (which showed Miss O was still purging) were not made available to the MDT on 22 August.
60. This meant the MDT did not have access to all relevant information when it made the decision not to admit Miss O to hospital or get a second opinion about this. A decision which we cannot say was reasonable or not as we have not seen the rationale for it due to incomplete record keeping. Moreover, incomplete record keeping means we do not know if a formalised plan was put in place for blood testing and ECGs at the MDT on 22 August.
61. We have seen nothing to suggest that Miss O’s capacity should have been assessed by the Trust. We have considered very carefully if the failings we have identified in Miss O’s potassium management amount to service failure. We find they do as they fall significantly below the relevant standards we have set out in our consideration.
Injustice
62. We have found service failure because the Trust did not robustly address the impact of Miss O’s trips abroad on its ability to deliver the care plan in full. It also failed to make contingency plans for any difficulties Miss O might have encountered out of the country. We also found service failure in the Trust’s management of Miss O’s potassium levels.
63. We will now assess the impact of those service failures on Mrs O. Mrs O told us she believed the Trust’s failings meant her daughter did not receive appropriate care and this led to her death. She said this continues to cause her profound distress.
64. Our advice is that when the body loses gastric acid through vomiting, it rebalances its acid and alkaline levels by producing additional potassium through the kidneys. Potassium is crucial for muscle and heart function, and low levels can lead to dangerously dysfunctional heart rhythms and to the heart stopping. Conversely high potassium levels can also be dangerous.
65. Blood testing and heart function monitoring by ECG consequently form an important part of treatment for people who self-induce vomiting as potassium levels can be corrected with treatment reducing the risk of harm or death. This is why it was crucial for Miss O’s care plan to set out a formalised testing regime, so that any abnormalities in potassium levels and heart rhythms could be spotted early and corrected.
66. If the Trust had put in place a formal testing regime at the appropriate frequency (at least weekly), it would have recognised Miss O was at high risk as the indications were that she was continuing to purge. Putting in place weekly ECG and blood testing would not have been possible in the immediate period post-discharge as Miss O was out of the country for a lot of the time. It is therefore concerning that the Trust did not address this matter with Miss O or her parents especially as she was hospitalised in both Brussels and Barcelona.
67. In the event, Miss O did have weekly blood tests for potassium levels when she returned from Brussels. However, these seem to have been done by accident rather than by design as Miss O’s care plan was never amended to included formalised testing. Nonetheless, the fact Miss O’s potassium levels were being monitored appropriately at that point after her discharge suggests there was minor impact to her in the absence of a formal testing regime.
68. This changed when Miss O returned from a trip to Barcelona on 16 August. She had been hospitalised for low potassium while abroad and an ECG showed an abnormal heart rhythm. This was another cue to instigate a formal blood and ECG testing schedule, but this information was not passed on to the MDT.
69. Miss O had one further blood test on 17 August which showed her potassium levels were normal but her bicarbonate level was high. These test results were never shared with the MDT meeting on 22 August. We asked the Adviser if the decision not to admit Miss O for inpatient care might have been different if the MDT had access to these blood test results.
70. The Adviser told us the high bicarbonate level indicated Miss O was highly likely still purging, but we do not know if she confirmed or denied this at the MDT meeting or if it was discussed. The Adviser said clinicians could have countered any denial from Miss O if the test results had been available to them and this may have led to further consideration of hospital admission. Alternatively, if Miss O admitted to continued purging, the normal potassium levels may have reassured her clinicians that continuing with the care plan was appropriate in the circumstances.
71. However, the Adviser told us that, in the absence of the complete notes from the MDT meeting, it is impossible to know the rationale for not admitting Miss O to hospital or whether this decision would have changed with the blood test results. We also do not know if the test results would finally have prompted a formalised testing plan. A blood test was arranged at the MDT meeting, but there is no evidence it was ever done.
72. In the nine days before Miss O died, she had no blood testing (and no ECGs were done throughout the whole post-discharge period). Had these tests gone ahead and showed Miss O’s potassium levels and heart rhythm to be abnormal, this could have been identified and corrected – potentially saving Miss O’s life. Unfortunately, we do not know what those test results would have shown had they been done and so cannot be certain Miss O would have survived.
73. Even if the tests had been normal, they could have become abnormal at any point in the future due to the frequency of Miss O’s purging. Because of this, we cannot say it was more likely than not that Miss O would have survived had it not been for the failings we have identified. What we can say is the Trust missed opportunities to have potentially averted the sad outcome in the days immediately prior to Miss O’s death.
74. We recognise Mrs O continues to be profoundly distressed about her daughter’s death and the failings she believes led up to it. She told us in response to our provisional views that the impact of failings in her daughter’s care on her were profound, enduring and life altering and she suffers from ongoing post-traumatic stress. Our findings will undoubtedly compound her trauma and distress now she knows the extent of those failings. Mrs O had a right to expect her daughter’s community care would protect her as far as possible from any avoidable harms.
75. Formalised regular blood and ECG testing should have functioned as a safety net in the event of Miss O continuing to purge. That safety net was ad-hoc at best during the beginning of Miss O’s community care and completely absent during the last nine days of her life. Mrs O will have to live with the knowledge that she and her daughter were failed by the Trust in the last weeks of Miss O’s life.
76. Mrs O will also have to live with the uncertainty of whether her daughter’s life could have been potentially saved had the Trust provided appropriate care. That uncertainty has been exacerbated by missing clinical records. We have no doubt that these are serious and ongoing injustices to Mrs O in consequence of the service failure we have identified. We therefore partly uphold her complaint.