In an audit of patient records the authors found inadequate recording of the type of dementia diagnosed, inadequate pain assessment and administration of pain relief and minimal recording of communication with families. Post audit, ward nurses collaborated to design a dementia toolkit that addressed the issues identified in the review. The toolkit included advice on communicating with people with dementia and family members, use of a cognitive impairment pain assessment tool and information on assessment and management of delirium. On introduction, it became evident that ward staff required more education and training to effectively use the toolkit. This was provided and a further audit, six months later, indicated significant improvement in awareness and care of patients with dementia. For example, post implementation; a cognitive impairment pain assessment tool had been used with 95% of patients. The authors suggest the toolkit is discreet and inexpensive while supporting staff to meet the needs of people with dementia.

The authors discuss the discrepancy evidenced in this study and others between prevalence and recognition of dementia among clinicians. This is particularly significant since poorer outcomes are evidenced for people with dementia who had longer hospital stays and were more likely to be readmitted in the next 12months. The authors conclude that significant investment is required in education and resources and this must take place within a wider cultural change in both institutions and society.

The study then compared the neuropsychiatric profile of these groups of patients using three well-validated instruments for delirium and dementia severity, the DRS-R98 (Revised Dementia Rating Scale), CTD (Cognitive Test for Delirium) and NPI-Q (Neuropsychiatric Inventory). It found that delirium, both with and without comorbid dementia, could be distinguished from dementia alone using the former two instruments. It also found that, although both delirium and dementia are characterised by a generalised disturbance of cognitive function, delirium can be distinguished from dementia because of distinct differences in impairment of attention and vigilance. The study argues that the cognitive functions of attention and vigilance should be emphasised in efforts to identify delirium, including in populations where there are high rates of dementia.

While ICPs for dementia care were identified, none that focusd on dementia care in an acute hospital setting were found. The authors also found limited research on the effectiveness and economic evaluation of ICPs generally.

Overall hospital activity related to people with dementia was equal to 10% of total bed days and 5% of total hospital case mix budget, a cost almost 3 times higher than for those without dementia. On average patients with dementia had a 25.6 day length of stay compared to 11.2 days among those without dementia. The authors highlight the likelihood that the data under represents the true impact of dementia as many people presenting to hospital do not have a diagnosis and cognitive impairment is not recognised. They point to the need for better community services, better cognitive screening and perhaps a dementia specific service for this cohort.

The authors set out to describe use of psychotropic, anti-cholinergic and deliriogenic medication among older hospitalised patients, both with and without dementia and to identify the prevalence of polypharmacy (5 or more medications) and psychotropic polypharmacy (2 or more psychotropic agents). They found that over two thirds of older people in hospital experience polypharmacy with a quarter experiencing major polypharmacy. In line with other studies they found high levels of psychotropic medication use with significantly higher levels of these being prescribed to people with dementia. Patients admitted from nursing homes were 5 times more likely to be prescribed an anti-psychotic medication. No significant differences were found between the two groups when it came to anti-cholinergic, deliriogenic or cardiovascular agents. The authors conclude that dementia is under diagnosed among this population and there is a high prevalence of polypharmacy and psychotropic drug use.

Levels of physical assessment ranged from high (80% of patients) in mobility, continence and pressure scores to low (less than 40%) in functioning and BMI, while no pain assessment was carried out for almost 25% of the sample. When it came to mental assessment, less than half of patients were assessed for cognition, delirium, behavioural, psychological symptoms of dementia, and mood. Dementia awareness training was found to be low, with a lack of mentorship and supervision. Most wards were running with less than a full complement of staff,. Discharge planning for people with dementia was found to be poor and high numbers of people admitted from home were discharged to long term care.

Less than one fifth of patients admitted were cared for in a specialist GM ward. Overall both types of wards (GM and non GM) performed poorly in the recording of relevant information, discharge planning and delirium assessment, although delirium assessment was slightly better in the GM ward. There were no statistically significant differences between wards in assessment of nutrition, continence, pressure care, eating and drinking. The Irish National Clinical Programme for Older People has identified access to specialist wards as key to the acute care of frail older people and the authors support this but highlight the importance of remaining self-critical and striving for improvement.

Results indicate sub optimal assessment of dementia symptoms with this being poorer among those at end of life or referred for specialist palliative care (SPC). Examples among this sub group (SPC) include lack of assessment of mobility (21%), ADLs (68%), pain (27%) and cognition (34%). No record of delirium assessment was found in almost 70% of all cases and 93% of those with dementia at end of life were not assessed for behavioural and psychological symptoms of dementia. Associated with the levels of sub optimal assessment is the likely over prescription of anti-psychotic medication (also documented in this review): https://academic.oup.com/qjmed/article/109/9/589/1752790). The authors conclude that a more systematic approach to delivery of palliative care for people with dementia must be developed and this will require the development of a multi-dimensional education programme and a model of integrated dementia palliative care coupled with research that demonstrates the benefits of applying a palliative care approach to dementia.

The paper describes a culture of anti-psychotic prescription; 41% of patients were prescribed an anti-psychotic during admission, a figure that is substantially higher than in the UK. Less than half of these received assessment of mental status and almost one quarter had no assessment of the presence of pain. The authors suggest that staffing issues and lack of dementia specific training are likely contributing factors. They argue that if a reduction in anti-psychotic prescription similar to that which has taken place in the UK is to happen, then a combination of research evidence, political will and strategic planning is required.

The authors employed the VIPS framework, using Values, Individualised, Perspective and Social and psychological concepts to guide analysis. They found barriers to good care included, ineffective pathways, unsuitable environments, inadequate resources and a lack of emphasis on education and training. They argue for capacity building in ethos, organisation and structure, and point to evidence that education must be set within a culture of positive organisational change. The need for more research providing a stronger evidence base is highlighted with particular reference to the need to measure the effectiveness of recent initiatives.

Prevalence is higher among public hospital admissions, reflecting more elective admissions to the private hospital and lower admissions of the “oldest old”. Higher prevalence overall is among acute medical admissions with 23.6% of these related to pneumonia. The authors reflect on the numbers of people being admitted to hospitals with an undiagnosed dementia and the implications of this for their care including assumptions about capacity to understand risks/benefits of treatment, inappropriate prescribing of medication and lost opportunities for planning care.

Over one third of these met the criteria for a diagnosis of dementia with only one third of this group having a prior diagnosis. The study reports a higher prevalence of delirium, polypharmacy, immobility and dependence among this group and a notable overlap between dementia and frailty. The authors were surprised to find no statistically significant differences between the two groups in terms of length of stay and in hospital mortality. However people with dementia were found to be more than two times more likely to die or be readmitted within 12 months. Overall this older population of patients had high mortality and institutionalisation rates and the authors point to the need for structured, age attuned care and comprehensive geriatric assessment.

Where there was a recorded diagnosis of dementia, there was an associated increase in mean length of stay. In recognition of the fact that many older adults entering acute hospital do not have a diagnosis, European age specific prevalence rates of dementia were applied to discharges in specific age groups. The estimated cost of extended stay for people with dementia was calculated at almost €20 million per annum. The authors acknowledge that analysis of this dataset cannot offer clear reasons for the extended stay but cite some likely reasons from existing literature such as co-morbidities, higher fall rates and a lack of training and knowledge of dementia in the acute sector.