Haloperidol prescribing to Scotland’s elders

In a previous post the FOI returns on Haloperidol prescribing in NHS Scotland were shared.  This followed on from my consideration of a BMJ report regarding the scale and potential harms of  such “off-label” prescribing to our elderly in hospital.

Since that time I have had a response from Professor Jason Leitch, National Clinical Director, Healthcare Quality, Scottish Government:

Today I have sent this reply to Professor Leitch:

To: Professor J. Leitch,
National Clinical Director, Healthcare Quality,
Healthcare Quality and Strategy Directorate
Planning and Quality Division
St Andrew’s House,
Regent Road,
Edinburgh EH1 3DG

8th June 2015

Dear Professor Leitch,
I was most grateful to receive your letter of reply dated 2nd June 2015.

I thought it best to reply to you to clarify the focus of my concerns. I wish to try and keep my reply short and focussed on the points you raise.

Point ONE:
You state that the Scottish Clinical Advisor for Dementia informed you that the “off-label use of Haloperidol for dementia is not especially unusual”. This would seem to diverge from  this BMJ change page made by NHS England’s National Clinical Director for Dementia, Professor Alastair Burns (I attach the full paper)

Dont use

You cite SIGN 86 guidelines on Dementia. These guidelines were issued 9 years ago when it was stated that “they will be considered for review in three years.” SIGN 86 is specifically for dementia and not delirium. The SIGN website indicates that there is no current plan to update SIGN 86 nor to introduce a Guideline on Delirium:

SIGN 86 was criticised in this research: Knűppel H, Mertz M, Schmidhuber M, Neitzke G, Strech D (2013) Inclusion of Ethical Issues in Dementia Guidelines: A Thematic Text Analysis. PLoS Med 10(8): e1001498. doi:10.1371/journal.pmed.1001498. I find it disappointing that an outdated and flawed guideline is still the basis for prescribing in dementia.

Ethical issues

Point TWO:
Haloperidol prescribing is part of the “Comprehensive Delirium pathway” introduced across NHS Scotland by the Scottish Delirium Association (SDA) and Healthcare Improvement Scotland (OPAC). You will be aware of this as I note that you are giving the key-note talk this week at the conference: Transforming delirium care in the real world”. Over a year ago the Secretary of the Scottish Delirium Association asked me to summarise my views on delirium improvements happening in Scotland. I did so and shared these with the SDA and with OPAC. I am disappointed to note that no reply has been forthcoming. I attach this summary for you with this letter.

It is welcome to hear that the Scottish Government are taking actions here. It is the case, by Scottish Government figures, that antipsychotic prescribing is increasing year-on-year in NHS Scotland. I seek improved care for individuals with delirium and dementia. I am concerned that current approaches, along with staff shortages and increased demands on staff time, are making it more rather than less likely that our elders may receive antipsychotic medication that can result in significant harms.

Yours sincerely,
Dr Peter J. Gordon

Included with letter:

The following is a summary on Delirium Screening written by myself in March 2014. I wrote this at the request of Professor Alasdair MacLullich who was one of those who led improvement work in delirium. I shared this with Healthcare Improvement Scotland, the Scottish Delirium Association and OPAC (Older People in Acute Care Improvement programme). I had no replies.

I have been asked to try and summarise my considerations on delirium assessment. In what follows I will also try and outline an alternative approach to the one currently being recommended across Scotland following Healthcare Improvement Scotland (HIS) Inspection visits.

1. DETECTION: I have been asked the entirely understandable question as to what approach I would advocate for detection of delirium if we were to depart from the Healthcare Improvement Scotland (HIS) mandate to screen all those 65 or over for “cognitive impairment”. Here, I shall try to make clear that my principle concern is with screening rather than with the brief rating scales themselves. These scales have a place, even if not yet fully validated. However, in my view, and that of NICE, they should be used for clarification, and for on-going assessment, of those who are determined – by professional nursing and medical assessment (including routine history and examination and collateral history) – of being at risk. It was in this sense, the patient-centred sense, that I used the word “holistic”. I apologise that I was not clearer about this in my lengthy piece, “the faltering, unfaltering steps”.

I use the word “holistic” in the general sense: time-honoured professionalism of person-centred nursing and medical care. This is the “professional excellence” that I am certain that we all support and wish to improve. I would like nurses and doctors to have training in delirium, and so provide a reasonable level of such care, ahead of deciding if a rating scale for triage is necessary. I am not aware that any RCT study has been done to compare: (1) current standard approach (which evidence demonstrates fails to detect 50-70% of delirium), with (2) brief screening tests, with (3) improved staffing levels, improved staff time and education (education, that the likes of HIS are undertaking) on delirium and cognitive disturbance?

2. THE 4AT SCREENING TEST: I was recently provided with two papers on the 4AT screening test which I now have had the chance to read closely. I had not intended commenting specifically on any one screening test as there are many that are currently being studied in the acute hospital setting. However, I thought it would be helpful to look at this study published in the journal of Age and Ageing by Bellelli et al: “Validation of the 4AT, a new instrument for rapid delirium screening: a study in 234 hospitalised older people” as it highlights some of the considerations I raised in my blog. I would suggest that the other paper, set specifically in a Stroke Unit is less generalisable in terms of HIS Improvement work, so I will make no further comment on it.

My view, in terms of being an evidence basis for Improvement Work in Scotland, that the title of this paper may be misleading. The Bellelli study only considered rapid delirium screening” whereas the 4AT test also “screens” for “cognitive impairment”. As the 4AT screening test is used for both we can only, at best, state that it has been validated for the former.

One of the four main features, following brevity of the test, is that “no special training is required”. In the Bellelli study the 4AT assessments were “performed by experienced physicians, though no specific training in the 4AT was given. Further research is needed to assess the ease of use of 4AT among other professional groups of varying levels of seniority.” Furthermore the Bellelli study “did not assess inter-rater reliability for the 4AT or the reference standard assessment” or the “clinical outcomes in relation to ‘possible delirium’ as assessed by the 4AT.” The Bellelli study was not set in A& E or hospital front-door settings. Additionally, as stated in the paper “because of insufficient power, we were not able to analyse the characteristics of misclassified (false negative and false positive) patients.” Thus we do need to be cautious in generalising the findings of this “validation” study.

I retain the view that we must also measure metrics such as patient/clinician acceptability before we can be confident for general application of any test.

The Bellelli paper concludes “future studies in larger populations and other centres should further assess its performance, including the determination of whether detection of delirium using the 4ATmay improve the clinical outcomes of patients.”

3. DEMENTIA and DELIRIUM: Screening for cognitive impairment. This has yet to be validated for the 4AT. As you may know I have concern about the delayed diagnosis of dementia. However I am also concerned about the potential for wrong diagnosis. Here, in the acute setting we have an almost intractable problem that I wonder if the 4AT can address. We will need to wait for further studies. However the incorporated AMT4 test for cognitive screening is one of the most basic (some would say “reductionist”) tests available to clinicians. How it then performs when one has delirium has not yet, to my satisfaction, been established. The Scottish Government have confirmed in a letter of response to Anne Begg, MP, that “there are no plans at present to introduce such a national screening programme for Scotland.”

I have noted that the use of screening tests (AMT4, CAM or 4AT) is being measured in many Scottish health Boards in terms of each Board’s “compliance”.

I must confirm that I have no issue with rating scales. Some are better than others. I use them all the time, to the best of my ability, in an evidenced-based way. However if used for screening or “case-finding” – though I note that Scottish Delirium Association (SDA), Older People in Acute Care (OPAC) and Healthcare Improvement Scotland (HIS) generally refer to “screening” – most public health experts would recommend that they need to be very carefully considered (as per the ten Wilson & Jungner World Health Organisation criteria). Please note the clear definitions of screening by the World Health Organisation and that this does not need to be whole (entire) population.

4. TIME: Taking delirium seriously should involve taking appropriate time. I am certain that we all agree on this. We also know the reality of staffing resources. Should we not be spelling out this shortfall to Scottish Government as part of improvement recommendations?

5. INVOLVEMENT: The point I was making in my blog was that, for improvement work like this, we need to engage beyond clinicians and improvement staff. I know that HIS have fully engaged with those who have experienced delirium. I suggest that we also need to give all those over 65 an opportunity to give their considerations. There are of course many others, in professional disciplines across medicine and beyond acute care, who may wish to have an input. I also wonder if the Mental Welfare Commission may have a view of starting out with a screening test without any perceived need (as per HIS recommendations) for individual consent?

6. ETHICS. There are unforeseen consequences of tests that are about the person, being, sentience rather than a unitary bodily process (I have reasoned many times why taking a pulse, for example, has quite different implications to that of cognitive testing.)

One unforeseen consequence of wider cognitive screening is the heightening of fear, as expressed clearly by Dr Iona Heath (Past President RCGP) and John Sawkins (VOX). It is possible our elders may become even more fearful of hospital if they understand that they will be cognitively tested whether they agree or not. Just today it was reported that a woman who was turned down by ATOS for benefits – as based on a screening questionnaire – then took her own life. The Mental Welfare Commission have investigated this and in the report express concerns about a short-cut approach to overall and proper understanding of this woman’s medical history and life situation. This is a most extreme example but we must consider the wider effects of screening however well intentioned and reasoned by the clinical argument. For those who want to learn more I would thoroughly recommend that you might read The patient paradox by Margaret McCartney.

7. INVERSE CARE: With any screening programme there is a potential risk of unintentionally precipitating inverse care. Since mandatory screening was introduced following HIS inspection, referrals to the older adults’ liaison service in my Health Board area have risen to 230 per month and the majority of these individuals are referred on for further community assessment. The community services are already stretched. The risks are that community services may struggle to meet the need of those who need it most and that those with mild cognitive impairment are wrongly medicalsised as suffering from dementia.

8. PEER REVIEW: A recent correspondent asked that I submit my considerations for peer review in formal academic journal. It is my view that those behind improvement work on delirium in Scotland need to do this first and I am of the view that this might be an important element of improvement work before policy recommendations are embarked upon. The Clinical Standards, on which screening recommendations for delirium have been based, are 12 years old. The involvement of Healthcare Improvement Scotland in pilot work, as guided by the Scottish Delirium Association is most welcome. The limitations of other avenues of communication about this improvement work have been pointed out to me. Yet we must acknowledge that OPAC, HIS and SDA are using other avenues extensively (tweets, blogs and videos), thereby inviting responses.

To conclude, I have no major issue with the use of validated scales as a form of assessment. However I agree with NICE that evidence, and wider consideration is currently not there to support use for screening or case-finding. Yet this has become the basis of Healthcare Improvement Scotland recommendations, recommendations that have been made with the guiding support of the Scottish Delirium Association.

The following behind-the-scene communications were recently released as a result of a Data protection request. The communications indicate a tone of disdain for those who may write to the Scottish Government's Department of Health with concerns about patient wellbeing and safety.


Professor Leitch chose not to answer my question about registration with the General Medical Council however he did kindly supply a one-paragraph “CV” which would indicate that he is neither medically trained nor qualified.

Professor Leitch is registered with the General Dental Council .