Dear Caroline Topping

We write in response to your letter to , Chief Executive Officer of the Royal Pharmaceutical Society, dated 26th July 2024 (Regulation 28 Report - Action to Prevent Future Deaths - Jessica Maizy Anne De Souza). Your letter was forwarded to us as the Publisher of the British National Formulary (BNF). We are sorry to hear the sad circumstances surrounding this case.

We have considered the matters of concern highlighted in your report that directly reference BNF (point ii), and additional concerns that do not specify the publication (points v and vi).

The matter of concern that directly references the BNF states that the clinicians relied on the BNF (and the BMJ Best Practice Bipolar Disorder in Adults) to support their decision to prescribe aripiprazole as prophylaxis for both polarities as a monotherapy. In Summer 2022, when aripiprazole was first prescribed to Ms De Souza, the aripiprazole drug monograph in BNF included one indication that was relevant to the management of bipolar disorder – that is the treatment and recurrence prevention of mania. The indication reflected, and continues to reflect, the licensed use of oral aripiprazole in licensed product information and is unchanged in current BNF content. The BNF does not include information on the use of aripiprazole as prophylaxis for bipolar depression.

With regards to the matters of concern that do not specify the publication, it may be helpful to provide some context around the scope of the BNF. The BNF is designed as a digest for rapid reference for non-specialist health professionals, and it may not always include all the information necessary for prescribing and dispensing. The BNF should be interpreted in the light of professional knowledge and supplemented as necessary by specialised publications and by reference to the product literature. In Summer 2022, the BNF provided, and continues to provide, a general overview of the management of mania and hypomania in the Mania and hypomania treatment summary, with the recommendation that patients with suspected bipolar disorder should be referred to a specialist mental health service and treatment initiated on specialist advice. The treatment summary also advised, and continues to advise, that an antidepressant drug may also be required for the treatment of co-existing bipolar depression, along with brief information on when to avoid the use of an antidepressant. As noted above, the BNF does not include information on the use of aripiprazole as prophylaxis in relation to the depressive polarity in bipolar disorder, and as such, we do not consider that the guidance in the BNF may have been misleading for the clinicians.

As we do for all BNF content, we will continue to monitor for additional sources of information around the management of bipolar disorder, and consider whether we need to include further

information, at a level of detail that is appropriate to the scope of the BNF, for a future update of the BNF.

We trust that this addresses this important issue

Dear Ms Topping Re: Regulation 28 Report - Action to Prevent Future Deaths - Jessica Maizy Anne De Souza As CEO of BMJ Publishing Group Limited (BMJ Group), I write to respond to the above report sent by email to me on 26 July 2024. We were very sorry to learn about the death of Miss de Souza and the circumstances surrounding her illness and would like to express our deepest sympathies to her family. We have considered the matters of concern listed in the report and set out BMJ Group’s response below. First of all, we would like to make clear that BMJ Best Practice is an online clinical decision support tool (https://bestpractice.bmj.com/) to be used as a source of reference material by medical professionals. The clinical content on BMJ Best Practice is live and can be updated at any time in response to changes in the evidence landscape - and in line with our processes. We have reviewed the content that is (according to our internal records) likely to have been online at the time of Miss de Souza’s hospital admission in Summer 2022. The first matter of concern that references BMJ Group (ii) states: The clinicians relied on the BNF and BMJ Best Practice Bipolar Disorder in Adults and the BNF to support their decision to prescribe aripiprazole as prophylaxis for both polarities as a monotherapy The second matter of concern that references BMJ Group (iv) states: The BMJ refers to aripiprazole being used as a monotherapy to treat bipolar disorder, though does advise that it is more effective in preventing mania than depression. BMA House BMJ Publishing Group Ltd Registered in England. No. 3102371 Tavistock Square Registered office: VAT registered no. GB 674 7384 91 London WC1H 9JR W : bmjgroup.com BMA House United Kingdom Tavistock Square London WC1H 9JR

The BMJ Best Practice ‘Bipolar Disorder in Adults’ topic, at the time (and still does) discusses the complex nature of bipolar disorder and the many factors to consider and monitor during treatment. It outlines the different phases in terms of acute mania, acute depression and patients with mixed features, as well as maintenance treatment, and the differing treatment options and considerations during each phase. It is the duty of the treating doctor to take all of this into account when deciding on treatments. We note that aripiprazole was used during the admission for an acute episode of mania, and continued following stabilisation and discharge. In the ‘Bipolar Disorder in Adults’ topic management approach section narrative, options are discussed for maintenance treatment. General comments in the content at the time said that “Monitoring and enhancing adherence is a routine part of long-term bipolar disease management” and that “Effective therapies that maximise adherence share common characteristics: education, self-monitoring, recurrence prevention, managing adverse effects, identifying and managing stressors, and addressing belief systems and attitudes to illness”. More specific information, under the heading ‘Bipolar I’ stated that “Lithium has the strongest evidence for prevention of recurrence in bipolar disorder compared with other agents, and remains the treatment of choice for long-term maintenance therapy. It is effective against relapse of both manic and depressive symptoms and appears to have an antisuicidal effect. Other first-line options include quetiapine (as an adjunct to lithium or divalproex sodium, or as monotherapy), divalproex sodium, lamotrigine, asenapine, and aripiprazole (as an adjunct to lithium or divalproex sodium, or as monotherapy as an oral or monthly injectable preparation). Quetiapine is effective in preventing manic, depressive, and mixed episodes and so may be particularly useful as maintenance treatment for patients with mixed features. Monotherapy with asenapine or aripiprazole is more effective in preventing mania than depression. Lamotrigine is more effective in preventing depression than mania, but is also indicated for prevention of recurrence for any mood disorder.” Aripiprazole is discussed as one of the treatment options, and as noted in the Regulation 28 notice, the BMJ Best Practice topic stated that monotherapy with aripiprazole is more effective in preventing mania than depression. Lithium is stated in the topic to be first line, and quetiapine is also highlighted as a useful option for patients with mixed features. Given that the content points to other agents possibly being preferable, the prescribing clinicians would have made the selection for aripiprazole knowing full details of the clinical context which might have included any previous medications used for treatment, potential adverse effects, or other medical conditions. The BMJ Best Practice website states that: “As a medical professional you retain full responsibility for the care and treatment of your patients and you should use your own clinical judgement and expertise when using this product. This content is not intended to cover all possible diagnosis methods, treatments, follow up, drugs and any contraindications or side effects. In addition, since such standards and practices in medicine change as new data become available, you should consult a variety

of sources. We strongly recommend that you independently verify specified diagnosis, treatments and follow-up and ensure it is appropriate for your patient within your region. In addition, with respect to prescription medication, you are advised to check the product information sheet accompanying each drug to verify conditions of use and identify any changes in dosage schedule or contraindications, particularly if the drug to be administered is new, infrequently used, or has a narrow therapeutic range. You must always check that drugs referenced are licensed for the specified use and at the specified doses in your region” The content in BMJ Best Practice discussed the multiple factors to consider in the context of the treatment options available to the clinicians: aripiprazole is stated to be an option, and the decision that this was the best option for an individual patient remains that of the prescribing clinician. The BMJ Best Practice topic discusses that treatment should be reviewed regularly and it may have been that on clinical review after a period of time, the treatment would have been adjusted or changed which might have been clinically appropriate. However the report describes how the planned follow-up did not take place. In updates since 2022, the topic discusses in more detail the differences between guidelines, and incorporates the NICE guideline in greater detail but there has been no substantive change to the content that is relevant to the report. In keeping with our processes, we will continue to review and update this topic with subject matter experts and will incorporate new evidence into the content as appropriate. We hope this response adequately addresses the matters of concern that reference BMJ Best Practice but please let me know if you require any further information.

Dear Ms Topping Re: Regulation 28 Prevention of Future Deaths Report in respect of Jessica Maizy Anne de Souza I write in response to your regulation 28 report regarding the sad death of Jessica de Souza. I would like to express my sincere condolences to Jessica’s family. We have reflected on the circumstances surrounding Jessica’s death and senior clinical advisers within our patient safety team have reviewed the concerns raised in your report. British National Formulary (BNF) and the BMJ Best Practice Bipolar Disorder in Adults. Your report indicates that the clinicians treating Jessica relied on the British National Formulary (BNF) and the BMJ Best Practice Bipolar Disorder in Adults. The BNF is a joint publication of the British Medical Association and the Royal Pharmaceutical Society. NICE hold the licence to make this resource available on the NICE website to health professionals working in the UK, but we are not responsible for the content. BMJ Best Practice is produced by the BMJ Publishing group and NICE cannot comment on its content. I note that you have sent your report to those parties, and they are best placed to respond to your comments on their content.

Page | 2 NICE guideline on bipolar disorder: assessment and management [CG185] NICE’s guideline on bipolar disorder: assessment and management [CG185] does not recommend aripiprazole to treat an acute manic episode. Our recommendation 1.5.3 says: ‘If a person develops mania or hypomania and is not taking an antipsychotic or mood stabiliser, offer haloperidol, olanzapine, quetiapine or risperidone, taking into account any advance statements, the person's preference and clinical context (including physical comorbidity, previous response to treatment and side effects).’ For longer term management of bipolar disorder (see section 1.7 of the guideline), NICE recommends discussion with the person, and their carers if appropriate, to help people understand that bipolar disorder is commonly a long-term relapsing and remitting condition that needs self-management and engagement with primary and secondary care professionals and involvement of carers. We suggest that the potential benefits and risks of long-term medication and psychological interventions, and the need to monitor mood and medication should be discussed with the patient and their carers if appropriate. We recommend that lithium should be offered as the first-line, long-term pharmacological treatment for bipolar disorder (recommendation 1.7.7). It is not clear from your report if lithium was prescribed in Jessica’s case. The guideline advises that when planning long-term pharmacological treatment to prevent relapse, take into account drugs that have been effective during episodes of mania or bipolar depression. Discuss with the person whether they prefer to continue this treatment or switch to lithium, and explain that lithium is the most effective long-term treatment for bipolar disorder. Our guideline goes on to say that if lithium is ineffective, poorly tolerated, or is not suitable (for example, because the person does not agree to routine blood monitoring), consider an antipsychotic (for example asenapine, aripiprazole, olanzapine, quetiapine or risperidone). Sections 1.6 and 1.7 of the guideline make recommendations on the management options for depression in bipolar disorder, including on the need for review. Our senior clinical advisers acknowledge that the guideline does not explicitly consider the 2 polarities of bipolar disorder in long-term treatment. We will discuss this area with our topic experts and review any new evidence that could impact on our recommendations, updating them if necessary.

Page | 3 I hope this response is helpful in confirming what NICE recommends and the actions that we will take because of your report relating to Jessica. I would like to reiterate my condolences to her family.