Ward based management of behavioural and psychological symptoms of dementia

Dear Editor,

Many thanks to the authors for this comprehensive and relevant article. My response centres on Table 1, summarising routine investigations suggested for patients with delirium or new BPSD.

As an Infectious Diseases trainee, the listing of urinalysis concerns me. This patient group are generally (but not exclusively) more elderly and we know asymptomatic bacteriuria is a common finding in such patients, estimated by NICE at 10% in men and 20% in women over the age of 65. For this reason, our national guidance in Scotland specifically recommends against the use of urinalysis in the diagnosis of infection in this population or in those with indwelling urinary catheters. Routinely performing urinalysis in this group may lead to administration of antimicrobials for asymptomatic bacteriuria, potentially with adverse effects such as antimicrobial resistance, side effects and development of Clostridium difficile-associated diarrhoea. Another significant risk is delaying the diagnosis of other factors contributing to delirium if UTI is accepted as the trigger for the episode.

For these reasons I would exercise caution recommending it as a routine investigation in this context. A more comprehensive assessment including signs and symptoms of UTI alongside biochemical/haematological evidence of an inflammatory response is more helpful in confirming or excluding UTI, alongside the other routine investigations suggested in the table.

Dear Editor

Nguyen et al. describe the familiar tale of an 81-year-old woman with dementia being admitted to hospital following a fall at home.1 They outline the standard ‘textbook’ management of behavioural and psychological symptoms of dementia (BPSD), which has changed little over the last few decades. Older drugs, like haloperidol, may have been replaced with newer drugs, like quetiapine, but this may simply have reduced anti-dopaminergic adverse effects at the expense of greater anti-cholinergic effects. It is estimated that around 40% of people in acute medical beds have dementia.2

Let’s explore the tale of the 81-year-old woman. Those who have worked in emergency departments (EDs) or medical admission units (MAUs) will recognise the sequence of events. It is likely that she was found lying on the floor by a neighbour or carer around lunchtime. It’s into the afternoon by the time the stretched ambulance service team have handed her over to the ED team. It’s early evening when her assessment in ED is complete. There are no fractures on her X-rays, no acute bleed on her head CT (if performed) and her bloods tests are all within normal limits. By this stage it is judged too late to be able to safely send her home. A bed is found for her on the MAU despite her having no acute illness and no treatment having been commenced.

People with advanced dementia can struggle to express their needs. It may now be many hours since she ate or drank. She probably needs the toilet. She lacks reassuring voices and familiar surroundings. She is sore from lying on trolleys. She doesn’t know why she is not going home. She gets up but is told to sit down because she may fall again. The MAU staff are extremely busy. They started the evening short staffed, half of the patients on their ward are confused. The management team are telling them of the long waits in ED. They don’t have the resource to sit by her bedside. The best they can do is escort her back to her bed each time she reaches the door. They have told her she is too unsafe to go home but she doesn’t seem to understand. On the tenth time of asking her to sit down the tone of their voice has become sharper. She feels threatened and hits out. This is the final straw, something must be ‘done’.

It is now into the night shift. The junior doctor is called to the ward. Crisis point has been reached, the drug cupboard is already open. A tablet is offered to ‘calm her down’ but she is frightened. Why can’t she go home? Why should she trust the group of people huddled around her, solely focussed on her swallowing their dodgy pill? She will not take it. We have reached the final option. The security guards are summoned. Our 81-year-old lady is held down by several people and a painful intramuscular injection is administered. Is that it, problem solved?

Few medications are licenced for BPSD. Any mild benefits are offset by the risk of moderate to severe harm.3 Evidence suggests that either antipsychotic or benzodiazepine intramuscular injections are marginally more effective than a placebo injection. After two hours, with either class of drug, behaviour scores are improved by around 15% compared to placebo.4 This may be no better than not receiving any painful injection. So, our lady will need to be restrained for a prolonged period. She is made more distressed by this process. She may now experience adverse effects of the administered drug, e.g. worsened confusion, an increased risk of falling or aspiration pneumonia.5 She has become trapped in a cycle of increased length of stay, elevated risk of adverse outcomes and plummeting probability of ever returning to her own home. This is not person-centred care, it is not enabling, but actively disabling. Medications, short of deep sedation, will not stop her wandering, will not improve her cognition and will not make her less likely to fall over.

How did we arrive here? She was admitted following a fall in order to ‘keep her safe’, BPSD is a consequence of subsequent events. This is a system failure. Non-pharmacological interventions can be effective if adequately resourced. These are based around person-centred care (i.e. seeing problems as unmet needs, retaining personhood and actively enabling), environment optimisation, adapted communication methods, involving family and de-escalation techniques. The place least likely to precipitate BPSD is her own home. Hospital environments promote BPSD. Yet we have developed a system that sees hospital admission as the best option. Acute medical wards, like many parts of the NHS, are stretched to the limit. Providing optimal person-centred care in these environments is, currently, very difficult.

We cannot continue to accept the nightmare scenario of pinning down frail older people to administer actively harmful medications so that they can be ‘managed’ in resource-starved settings. It’s time to genuinely provide high-quality person-centred care and forget the past dependence on drugs whose adverse effects far exceed any small benefit. We need to develop new ways of working, which will need to be properly resourced, that engage in positive risk-taking to assess and keep more people in their own home and, when hospital admission is essential, develop services that are genuinely capable of providing optimal person-centred care. Effective healthcare systems are designed around the patients they serve. Put yourself in her shoes, what would you choose?

References
1. Nguyen J, Ko I, Martinez-Sosa S, et al. Ward based management of behavioural and psychological symptoms of dementia. BMJ 2021; 374: n1779.
2. Sampson EL, White N, Leurent B, et al. Behavioural and psychiatric symptoms in people with dementia admitted to the acute hospital: prospective cohort study. Br J Psychiatry 2014; 205: 189–196.
3. Ballard CG, Waite J, Birks J. Atypical antipsychotics for aggression and psychosis in Alzheimer's disease. Cochrane Database Syst Rev 2006, Issue 1. Art. No.: CD003476.
4. Meehan KM, Wang H, David SR, et al. Comparison of rapidly acting intramuscular olanzapine, lorazepam, and placebo: a double-blind, randomized study in acutely agitated patients with dementia. Neuropsychopharmacol 2002; 26: 494-504.
5. Herzig SJ, LaSalvia MT, Naidus E, et al Antipsychotics and the risk for aspiration pneumonia in non-psychiatric hospitalized patients: a cohort study. J Am Geriatr Soc 2017; 65: 2580–2586.