When an individual experiences a stroke a series of clinical decisions are made (either implicit or explicit) about the most appropriate setting for their care. These decisions can be considered in the form of four main issues, recognising that each individual stroke patient presents a unique set of problems and potential solutions. Efficient and effective management of patients depends on a well-organised expert service that can respond to the particular needs of each individual patient. To achieve this, the organisation of stroke services must be considered at the level of the Health Board, Primary and Acute Trusts and in the patient’s own home or care home.

The main issues in planning services for stroke patients are:

• Organisation of hospital care
• Hospital or home-based care
• Discharge and post-discharge services
• Ongoing rehabilitation and follow-up.

An important part of the assessment process should include identifying whether there were any pre-stroke problems or co-morbidities.

2.1 Organisation of hospital care

A Cochrane review of the benefits of stroke rehabilitation in an organised hospital stroke unit found that in comparison with a general medical hospital ward:¹⁰

• 18% reduction in death (95% confidence interval (CI) 6-29%)
• 20% reduction in death or institutional care (95% CI 10-29%)
• 22% reduction in death or dependency (95% CI 11-32%).

These benefits were seen for those under and over 75 years of age, male or female and those with mild, moderate or severe stroke. Evidence level 1++

Length of hospital stay appears to be reduced by between two to ten days but this result is inconsistent between trials. Evidence level 1++

The benefits of a stroke unit were seen in units that admitted patients directly from the community or took over their care within two weeks of admission to hospital. The evidence of benefit is most clear for units which can provide several weeks of rehabilitation if required. Evidence level 1++

The numbers needed to treat for stroke unit care are:

• For every 33 patients treated in the stroke unit there is one extra survivor (95% CI 20-100)
• For every 20 patients treated in the stroke unit one extra patient is discharged back to their own home (95% CI 12-50)
• For every 20 patients treated in the stroke unit there is one extra independent survivor (95% CI 12-50)

The confidence intervals are wide reflecting modest to substantial benefits. Evidence level 1++

Patients admitted to hospital because of acute stroke should be treated in a multidisciplinary stroke unit.

The stroke unit trials did not directly address the management of younger stroke patients, but subgroup analysis indicates that stroke unit care is of equal benefit to those aged below and above 75 years. Younger stroke patients with specific needs (e.g. vocational rehabilitation, caring for young family) may benefit from referral to rehabilitation services for younger adults.¹⁰ Evidence level 1+

Although admission to an organised stroke unit is the treatment of choice, it may not always be feasible. Small hospitals in rural areas with small numbers of stroke patients may have generic rehabilitation services. The systematic review of stroke units included trials of mixed rehabilitation wards (i.e. where multidisciplinary care is provided to a range of disabled patients including those with stroke).¹⁰ Six trials compared a mixed rehabilitation ward with care in the general medical ward and found that patients in the mixed rehabilitation ward were less likely to die or require long term institutional care or remain dependent. Direct comparisons of mixed rehabilitation wards with stroke rehabilitation wards favour the stroke-specific ward,¹⁰ with fewer patients dying or requiring institutional care or remaining independent. Evidence level 1+

Where rehabilitation in stroke rehabilitation units is not possible, rehabilitation should be provided in a generic rehabilitation ward.

2.1.1 INTEGRATED CARE PATHWAY

An integrated care pathway (ICP) aims to provide organised and efficient multidisciplinary patient care. It should be based on the best available evidence and guidelines. A Cochrane review¹¹ identified three randomised controlled trials (RCTs) and seven non-randomised studies where the use of an ICP was associated with positive and negative outcomes. Although there was evidence of more appropriate use of investigations and fewer urinary tract infections, patient satisfaction was lower in the ICP groups. ICPs may not provide any additional benefit in a well established organised stroke unit if usual care is already excellent. Evidence level 2+

2.1.2 STROKE LIAISON NURSE/CO-ORDINATOR

The ‘Scottish Stroke Services Audit’ reviewed the structure of stroke services in Scotland at April 1998.¹² The audit involved both NHS Trust providers of care and Health Board commissioning bodies. The audit showed that structures of care for stroke patients were varied and complex, although patterns could be identified. The role of the stroke liaison nurse/co-ordinator was valued where introduced. The main responsibilities varied, but all job descriptions included the co-ordination of hospital care.¹³ Further work is required to define the optimum role and service characteristics of such posts, and to demonstrate their effectiveness. Given the audit evidence of significant gaps in stroke service provision in Scotland, it is reasonable to expect nurse co-ordinators to help improve care. The role of the stroke liaison nurse/co-ordinator is different and complementary to that of the stroke family support worker (see section 3.8). Evidence level 3,4

Trusts should consider appointment of a stroke liaison nurse/co-ordinator.

2.2 Admit to hospital or remain at home?

For patients who have had a mild stroke, the healthcare team may wish to consider whether to admit them to hospital or to arrange care in the patient’s own home. Care at home is an attractive idea for patients with acute stroke, but studies evaluating potential alternatives to hospital have been inconclusive. A systematic review¹⁴ found no benefit for care at home against unorganised hospital care. One study compared domiciliary care against two types of hospital care (general wards with a stroke team giving advice and an organised stroke unit).¹⁵ Stroke outcome was significantly better when patients were treated in the organised hospital stroke unit compared to organised domiciliary care or general ward hospital care (with stroke team advice). If hospital stroke unit care is not available, organised multidisciplinary domiciliary care has similar outcomes to unorganised (general medical ward) hospital care.¹⁵ Evidence level 1++

Stroke patients who are dependent in activities of daily living should receive hospital- based care in organised stroke units.

It is worth noting that even if patients are thought to have had a mild stroke, they still need to be investigated.

Patients who have a non-disabling stroke need to be urgently investigated and this may be most efficiently done by immediate admission to hospital or by early access to a neurovascular clinic. Computed tomography (CT) scanning should be performed within 48 hours. If investigation is delayed, CT scanning may miss a small primary intracerebral haemorrhage and rare but devastating causes of stroke such as bacterial endocarditis may be overlooked.

2.3 Early supported discharge and post-discharge support

Stroke unit care typically involves an early assessment of discharge needs and the development of a discharge plan involving the patient and carers. A Cochrane review¹⁶ of seven completed trials indicates that early supported discharge (ESD) services can reduce the length of hospital admission in selected stroke patients. ESD services were provided by a co-ordinated multidisciplinary team who assessed individuals during hospital admission, co-ordinated their discharge and provided post-discharge rehabilitation. Most services excluded those with very mild or very severe stroke and were available for approximately 30% of all hospitalised stroke patients. These services appear to provide an outcome at least as good as hospital care. The impact on service costs is likely to be modest. Evidence level 1+

Early supported discharge services provided by a well resourced, co-ordinated specialist multidisciplinary team are an acceptable alternative to more prolonged hospital stroke unit care and can reduce the length of hospital stay for selected patients.

2.4 Rehabilitation for patients in the home

Many of the principles of good stroke rehabilitation are relevant to people who are not admitted to hospital. Where applicable the evidence to guide practice for patients at home will be discussed in the guideline.