See article on page 141

About 4.5 million people worldwide die from stroke each year, and 9 million are estimated to be living with its consequences. One in every three deaths in the UK results from stroke, and it is the single greatest cause of disability in the adult population. Given that the incidence of stroke is estimated to rise by as much as 30% over the next 20 years,¹ it represents a major and ongoing challenge for society. For the people who survive the initial insult, there can be residual difficulties with self-care and mobility, communication, cognition, or emotional wellbeing. Such limitations can bring about major changes in “life”, impacting on roles and relationships within the family, social circle, and work. Stroke can clearly be a devastating condition for the individual and his or her family, but dedicated stroke services have been shown to reduce the impact both in terms of mortality and morbidity.

Quite why stroke units work is yet to be clearly established. They rely on a complex combination of skilled staff in a range of professions, acute treatment with thrombolytic agents, early mobilisation, patient motivation, and a host of other factors. Discerning and monitoring the most active ingredients is therefore a difficult task. However, a number of aspects of management²³ have been identified as key components of improved care for people with stroke, and the paper by Rudd and colleagues in this issue of Quality in Health Care⁴ is a timely exploration of how the UK is performing in providing that care.

Rudd et al outline a number of areas where there has been improved compliance with standards of best practice in stroke care since the first stage of the national sentinel audit of stroke in 1998. They explore the degree of this improvement and evaluate whether the audit process itself has contributed to the gains. A number of interesting issues are raised in the paper, which hopefully will stimulate ongoing attention to what constitutes “good stroke care” and how best to provide it, as well as what constitutes “good audit” and how best to do it.

Given what we know about aspects of management linked to improved outcomes, it is somewhat depressing to read that the organisation and process of care for people who have a stroke remains geographically variable and less than optimum in many ways. Nevertheless, the paper describes some very positive changes between the two phases of the audit, including:

• an increase in the number of stroke units;

• an increase in the number of consultants specifically responsible for people following a stroke;

• the development of interdisciplinary documentation.

Such structural developments are impressive and, if audit has helped to bring those changes about, all credit must go to healthcare professionals and management for responding positively. There seems little doubt that the efforts to involve relevant stakeholders in developing the audit tool and the wide dissemination of results has contributed to the successes reported. Obtaining strong participant involvement is vital to the success of audit. Johnston et al⁵ propose a number of success inducing strategies including the introduction of a modern (preferably electronic) system of medical records, the appointment of dedicated staff, and provision of protected time. It seems vital that such structural components are in place or audit risks becoming merely an additional task for an already overstretched workforce—which does little to enhance the quality of care. While audit can go part of the way to ensuring that the best standards of care are delivered, actually quantifying many aspects in the care of stroke patients can be problematic. Indeed, it can be difficult to make sure there is even a shared understanding of the definition of some components of a stroke service.⁶ The fact that provision of information for patients and relatives is one component of the audit exemplifies this issue. It might seem intuitive in this situation that “more is better”, but this is not necessarily the case. Certainly, along with others we have found that the appropriateness, timeliness, and manner of information provision heavily influences whether that information is useful.⁷ To be certain that there has been a meaningful improvement in such a process, clarification of the types of information and the manner in which it should be delivered might be required.

Although this study demonstrates an overall improvement in the achievement of a number of standards of care, there is much room for further progress. Less than half of the patients received a formal cognitive assessment, only a quarter had their mood state documented, and the needs of carers were assessed separately in only just over a third of cases. These issues are crucial to the provision of a high quality service. There are also disappointing findings with regard to frequency of team meetings, the attendance of social workers at such meetings, and the in house education of staff. Such factors are usual indicators of good communication and fundamental to truly interdisciplinary teamwork.

There is no doubt that increasing the number of patients who receive the bulk of their care in a dedicated stroke unit is vital if we are to achieve the desired reduction in deaths and disability. All who provide, fund, and use healthcare services would do well to heed this message, made even clearer by Rudd and colleagues.⁴ The message is particularly relevant in areas where many patients continue to be cared for in general hospital wards or at home,⁸ and in countries (including New Zealand) where few dedicated stroke units even exist. Failure to establish such services, and failure to ensure equitable access to them, is becoming a difficult standpoint to defend.

Finally, despite increased awareness that specialist stroke care improves outcome, investment in stroke research has recently been described as woefully inadequate⁹ and many important questions about the best interventions for acute stroke remain. This is also true in rehabilitation where more research is needed into what constitutes the most effective approaches to goal setting, teamwork, and other key components of practice.¹⁰ We agree with Rudd and colleagues that there should be ongoing funding for audit, and that audit can and should be a mechanism for improving what we do. However, it is also important to avoid assumptions that we might know all we need to about which processes to audit. If we listen to people who have had a stroke and their families, we still have a lot to learn about what comprises best practice and indeed “best outcomes”.