How can we keep patients with dementia safe in our acute hospitals? A review of challenges and solutions

Abstract

Maintaining patient safety in acute hospitals is a global health challenge. Traditionally, patient safety measures have been concentrated on critical care and surgical patients. In this review the medical literature was reviewed over the last ten years on aspects of patient safety specifically related to patients with dementia. Patients with dementia do badly in hospital with frequent adverse events resulting in the geriatric syndromes of falls, delirium and loss of function with increased length of stay and increased mortality. Contributory factors include inadequate assessment and treatment, inappropriate intervention, discrimination, low staff levels and lack of staff training. Unfortunately there is no one simple solution to this problem, but what is needed is a multifactorial, multilevel approach at the seven levels of care – patient, task, staff, team, environment, organisation and institution.

Improving safety and quality of care for patients with dementia in acute hospitals will benefit all patients and is an urgent priority for the NHS.

Introduction

Patient safety has become a major focus of global health research over recent decades, and as a result much has been learnt about the causes, prevention and amelioration of healthcare-associated harm (1). At the same time, inpatient hospital populations are becoming older and frailer, with 70% of inpatients over the age of 65 years, and at least 30% of all inpatients in general hospitals suffering from dementia (2).

It is well recognised that people with dementia, often with multiple co-morbidities and reduced functional as well as cognitive reserve, are particularly vulnerable to adverse outcomes of hospitalisation (3). In the U.K. there has been widespread recent acknowledgement that the care of patients with dementia in the general acute hospital needs to improve, following the Francis Inquiry (4) and other high profile reports (5). The National Dementia Strategy in 2009 (6) set out to improve the care of patients with dementia across all settings; in hospital this means not only reducing under-diagnosis and improving management of dementia itself, but also ensuring that patients with dementia admitted for any clinical reason receive high quality and safe care. There is a suggestion that improving acute hospital care for patients with dementia may result in considerable financial cost savings for the NHS (5).

Despite all of this, there has been a tendency for many recent national patient safety initiatives to focus on improving safety in clinical areas from which older patients with dementia are often excluded, such as surgery and critical care (7). The aim of this review, inspired by a recent meeting organised by the Patient Safety and Geriatrics & Gerontology sections of the Royal Society of Medicine (at which the authors were all speakers) is to consolidate what is known about patient safety issues that are particularly relevant to patients with dementia and, importantly, to describe strategies that can be taken to keep this vulnerable group safe in hospital.

Literature review method

This review is based on a search of Medline over the last ten years, up to February, 2012, using keywords and MESH terms generated around the following facets: i) dementia, cognitive impairment, ii) patient safety, adverse events, iatrogenicity iii) acute hospital care. Additional articles were identified from reference lists and the authors’ personal files.

Results

A pathway of harm, with multilevel causes and interventions

The findings of this review led to the conceptualisation of a pathway of harm that can affect patients with dementia that are hospitalised with an acute illness, and an associated multifactorial approach to patient safety intervention for this group. This harmful pathway is outlined in Figure 1.

Figure 1.

The harmful pathway of adverse events in patients with dementia admitted to hospital.

Types of harm, including development of the geriatric syndromes

Adverse events are the units by which hospital associated harm have traditionally been measured. They are classically defined as “an unintended injury caused by medical management rather than by the disease process, which is sufficiently serious to lead to prolongation of hospitalisation or to temporary or permanent impairment or disability to the patient at the time of discharge or death” (1). There is substantial evidence from large retrospective studies that adverse event incidence in acute hospitals increases with age (8). This relationship is related to frailty, comorbidity and reduced functional ability rather than age alone – all of which are characteristics of patients with dementia.

Frail older patients experience not only those types of adverse events also experienced by younger people (e.g. procedure-related complications, hospital acquired infections, adverse drug events), but also can acquire a range of “geriatric syndromes” (namely delirium, falls, incontinence, poor nutrition, immobility, functional decline and pressure sores) during a hospital admission (9). All of these have undesirable consequences, such as reduced quality of life, prolonged length of stay, institutionalisation, or even death, and are to some degree preventable.

Delirium is a key geriatric syndrome commonly affecting patients with dementia. There is a strong argument that if delirium arises newly during a hospital admission, we should consider it to be an adverse event (10). In patients with dementia, hospital-acquired delirium can have particularly devastating effects, as it worsens cognitive function and impairs recovery from acute illness. It is preventable in at least a third of cases with good care (11), yet it is often unrecognised, under-investigated and inappropriately treated (12). Even though delirium is sometimes unavoidable in dementia, early recognition can help with immediate treatment of the underlying cause or causes and prevention of complications.

Some direct causes of this harm

1. Inadequate assessment and treatment

   The rapid pace and technological focus of modern hospital care makes good clinical assessment and treatment more difficult in older people with dementia. As these patients are often unable to communicate their needs, collateral information is essential from relatives and carers, especially regarding previous function and cognition. This is made difficult by the rapid transit through admissions systems not designed with older people in mind. Multiple nursing and medical handovers lead to the loss of information and increase the risk of serious adverse drug events (13).

   Cognitive impairment creates its own barriers to safety; patients with dementia are not able to be as involved in their care as those who are cognitively intact, for example, in questioning staff about handwashing in the prevention of hospital acquired infection.

   The immediate hospital requirements for efficient bed management may take precedence over individualised care. For example, there is often poor insight and appreciation of falls risk in an individual patient when decisions are made regarding ward allocation. For patients with dementia, multiple moves and room changes, sleep deprivation and being in unfamiliar surroundings also increase the risk of delirium (14). Patients with dementia with or without delirium are at particular risk of being restrained or kept in bed unnecessarily, which can lead to functional decline, hospital acquired pneumonia, thromboembolism and pressure ulcers. In patients with dementia it is important to encourage mobility, even though this may increase the risk of falls. Patients with dementia are particularly prone to what is called `cascade iatrogenesis’, where a relatively innocuous first adverse event, which could have been prevented, leads to the serial development of multiple medical complications (15).

2. Inappropriate interventions

   The reduced physiological and cognitive reserve of patients with dementia means that standard clinical interventions need to be used with more care in this vulnerable group of patients. Decisions concerning nursing and medical interventions need to be very carefully considered. Patients with dementia are particularly prone to adverse reactions to medication, especially antipsychotics which can cause oversedation, dysphagia, parkinsonism, stroke and increased mortality (16). Care also needs to be taken with other groups of medications, such as anticholinergics which can worsen cognitive impairment and precipitate urinary retention (17), or opiates which can cause delirium and constipation. Polypharmacy in itself is harmful, particularly because of reduced concordance and accumulation of side effects.

   Inappropriate clinical procedures are also a problem, for example unnecessary urinary catheters are used commonly in patients with dementia, despite their association with urosepsis and delirium (18). Overdiagnosis and unnecessary treatment is also a problem; a common pitfall in patients with dementia, who frequently have asymptomatic bacteria, is overdiagnosis of urinary tract infection and unnecessary prescription of antibiotics which may in turn lead to microbial resistance and clostridium difficile infections (19). Antibiotics should be avoided unless there are definite symptoms of a urinary tract infection or evidence of sepsis or inflammatory markers are raised (19).

Underlying contributory factors

1. Discrimination

Older patients with confusion are often discriminated against in acute hospitals (12). Sometimes they may even be perceived to be `unwelcome’ in acute emergency departments (20). Doctors and nurses spend less time with confused than with non- confused patients (21). Frequently, underlying serious illness in these patients is not recognised (e.g. pneumonia, sepsis, myocardial infarction) and patients are given an inappropriate label, such as social admission or `acopia’, which discourages proper assessment and investigation and treatment of the underlying problem (22). Underlying societal negative attitudes towards older people, particularly those with dementia, are reflected by infantilisation and “elderspeak” amongst hospital staff (23).

Lack of trained staff

Frequently patients with dementia are highly dependent and need one-to-one care, especially for nutrition and fluid intake and even more so when they are also suffering from delirium. However, many acute medical and surgical wards occupied by older people with dementia are currently understaffed, with fewer trained nurses and more non-trained staff (24). There may also be fewer doctors and therapists per patient. This can directly increase the risk of adverse events – particularly falls, drug errors, poor nutrition and dehydration. Nursing staff need to be given training in `risk enablement’ and management backing to take `calculated risks’, for example by encouraging mobility in a patient at risk of falls in the overall best interests of their patients. A lack of personal interaction as a result has a particularly damaging effect on cognition - nearly half of carers report that being in hospital had a negative effect on the general physical health of the person with dementia which was not a direct result of the medical condition (5).

In summary, patients with dementia are prone to a series of circumstances and factors in modern acute hospitals that result in adverse events including the geriatric syndromes of delirium, falls, immobility, incontinence and functional decline. These can lead to a downward cascade of interacting problems, resulting in further dependency, institutionalisation, or death in hospital (Figure 1). Apart from its effect on the patient with dementia, this can lead to serious carer strain and dissatisfaction. Can this disastrous path of adverse events be prevented?

Potential solutions

One of the greatest obstacles to progress in patient safety is paradoxically the attraction of neat simple solutions (1). One such solution is to simply prevent admission of patients with dementia to hospital. Certainly patients with dementia may sometimes be unnecessarily admitted when the care they require can be provided at home. However, in the `real world’ patients with dementia often live alone and it is currently very difficult and very expensive to organise twenty-four hour care in the community at short notice when there is a medical crisis. Also, older people with dementia can present in very non-specific ways with severe treatable illness, especially life threatening infections, as well as with high levels of functional dependency (25) and they should not be disadvantaged by being denied modern high technological investigations and treatment. They may actually benefit more from modern treatments for stroke, myocardial infarction and sepsis than, for example, younger patients because they are particularly vulnerable to the consequences of undertreated illness. Intermediate care does have a role in preventing admission in patients with dementia, but its role is best limited to patients with a defined single diagnosis suitable for home treatment and its cost effectiveness has been questioned (26).

A multifactorial multilevel approach

There are many single interventions which can potentially halt the progress of specific adverse events and therefore reduce harm. However, experience shows that improving the safety and reliability of any clinical system requires action at many different levels (27). Hence, if we are to make real progress in improving the safety of care for patients with dementia in hospital, this is more likely to be as a result of multifactorial packages of interventions rather than single steps (1). A previously described multifaceted conceptual framework (28) can readily be applied to this problem, in which the following seven safety levels are considered: patient factors, task factors, individual (staff) factors, team factors, environmental factors, organisational factors and institutional context factors. These are summarised in Figure 2.

1. Patient factors.

   The gold standard for practice at the patient and staff level is the “person centred” approach advocated by the National Dementia Strategy (6). This means valuing people with dementia, treating them as individuals, viewing systems of care from the patient’s perspective and creating a positive social environment (29). Regard for the individual and creating the right environment, as well as being compassionate, creates safety as well as quality. Frequently the medical aspects of dementia are over-emphasised in hospital in comparison with the emotional and psychological aspects. Many aspects of Inouyes multicomponent intervention for delirium focus on these patient factors (11). Put simply, patients with dementia are much less likely to become delirious `if their care is directed to helping you see what you are doing, hear what’s happening, move around, get some sleep and have enough to drink, in an environment where people talk to you and let you know what is going on’ (1).

2. Task factors.

   Keeping patients safe relies upon sound, clear task design and the availability and utility of protocols and necessary patient-related information. We need to build on a recently increasing tendency to include more older, frailer patients in clinical trials, so that we have a stronger evidence base for their optimal medical management. The increasing use of centralised electronic records will also enhance safety for patients with dementia, in whom relevant information can be held by different healthcare stakeholders, and can be difficult to procure during the early stages of an acute hospital admission.

3. Individual (staff) factors.

   The individual attributes, skills and behaviours that are known to enhance patient safety are particularly important in the care of complex patients with dementia and should be developed and encouraged. Communication skills are particularly important here, and training of the specific ways in which individuals can effectively communicate with cognitively impaired people in order to gain accurate information, identify problems and formulate tailored management plans and decisions should be common practice.

4. Team factors

   The evidence for team factors in improving safety and quality for patients with dementia revolves around effective multidisciplinary assessment and comprehensive geriatric assessment, which improves all outcomes, including mortality. It is a paradox that dementia patients are sometimes regarded as having `no rehabilitation potential’ when they have most to gain from a problem solving multidisciplinary approach. Time and staffing pressures are major barriers to good team working.

5. Environmental factors.

   Improving the hospital environment for patients with dementia, as well as improving quality of care and carer and patient satisfaction may reduce mortality. There is a growing literature on the practical aspects of design of the care environment to enhance the safety of people with dementia and the role of the environment in delirium (18).

6. Organisational factors.

   An often-suggested solution to the problems of looking after patients with dementia in the acute hospital setting is to increase nursing staffing levels on the wards. Although very welcome, this is unlikely to be effective in isolation, without increasing training and education and changing the organisation of acute care services.

   There is evidence that improving the organisation of care using specialised teams can improve outcomes. Examples of innovative ways of changing systems to improve care for frail older people include the development of Acute Care of the Elderly Units focusing on comprehensive geriatric assessment and multicomponent delirium prevention. It is an historical accident that psychiatric hospitals and acute hospitals have developed separately, usually on separate sites and there may be merit in the development of joint geriatric/psychiatric wards to combine expertise and improve management for patients with dementia and delirium (30).

7. Institutional context factors

   There is some evidence that the culture of an organisation leads to improved outcomes in the care of older people. It is likely that for any safety initiatives to be successful then they need to be promoted throughout the organisation with leadership at both board and clinical levels. Certainly there is a growing wider societal appreciation that the care of vulnerable older people needs to be a high priority. The implementation of the National Dementia Strategy to appoint Hospital Clinical Leads is a welcome initiative (6). However, the setting of emergency care national targets can sometimes divert time and resource from concentrating on improving communication and relationships to improve care for dementia patients.

Figure 2.

Interventions at the seven levels of safety for hospitalised patients with dementia.

Conclusions

Patient safety is at the heart of quality in medicine. It is crucially important in our modern, caring society to provide high quality and safe care to the more disadvantaged sections of our population, including patients with dementia. In the past, patient safety efforts have concentrated on younger patients, especially in litigious areas and where problems are perhaps more clear-cut, such as surgery or obstetrics. Now is the time to turn our attention to frail older patients with dementia. Improving the safety of patients with dementia will also have benefits for all patients, especially by improving the hospital environment and encouraging person-centred care.

In order to address this challenge, we need to think more closely about the design of systems of acute care for complex patients with multiple co-morbidities. We need to increase awareness and understanding of the ways in which the manifestations of healthcare related harm are different in these patients, often presenting as the geriatric syndromes such as falls, delirium, incontinence and functional decline. The subtlety of presentation and detection of adverse events in patients with dementia is compounded by a culture of `low expectation’, which runs counter to the required `safety culture’. There should be the same attention to preventing delirium in an older patient with dementia as there is for preventing a wound infection in a young surgical patient.

This review would suggest that simple single condition-specific interventions and guidelines may be helpful, but are unlikely to be sufficient to improve overall safety for these patients. Instead, we require a total safety approach, involving actions at multiple levels: patients, tasks, staff, teams, environment, organisational and even societal.

We require a greater emphasis on interventions to change behaviours, attitudes and awareness, together with ways of facilitating organisational and system change. There is a need for better training in `risk enablement’ and the decision making skills required to balance the danger of inappropriate interventions against the danger of denying older people with dementia equitable access to services and expertise.

Geriatricians are already familiar with multicomponent system approaches to the complex problems of old age and there are many successful service models. We need to expand this approach so that for any frail older person with dementia, admission to hospital with an acute illness is a beneficial intervention at their time of greatest need rather than a critical life event leading to functional decline, institutionalisation and death.

Declarations

Competing interests

None declared

Funding

JG: none. SL and CV: The CPSSQ is funded by the National Institute for Health Research.

Ethical Approval

Not applicable

Guarantor

JG

Contributorship

All authors contributed equally.

Provenance

Submitted; peer reviewed by Rowan Harwood