Stroke strategy: how are we doing?

It is the most common form of cerebrovascular disease and accounts for 10 per cent of deaths in England as well as being a significant and preventable cause of disability.

The risk of stroke and death from stroke rises steeply with age, as does the percentage of deaths accounted for by stroke (see first chart). Age-standardised death rates from stroke in the UK are higher than France, Germany and the Netherlands, but lower than Italy, Greece and Portugal.

No single indicator is ideal for measuring the frequency of stroke, and no two indicators put regions in exactly the same rank order (second chart). Death rates in those aged under 75 (the age range used in public service agreement and cancer targets) depend not only on the frequency of stroke but also on the proportion of cases that are fatal. Emergency hospital admission rates depend on frequency of stroke but also on primary care and hospital admission practices.

However, quality and outcomes framework prevalence (estimated from quality and outcomes framework returns) is affected not only by frequency of stroke but also by ascertainment and recording by GPs. Framework prevalence is not a reliable measure of prevalence as the data is not adjusted for age, but it generally shows a similar rank order to emergency admissions. London has a much lower framework prevalence (10 per 1,000) than others chiefly due to the lower age of its population - although this hides the importance of stroke as a priority in some ethnic groups.

Indicaton of outcomes

The strategy suggests 20 markers with which the quality of services can be assessed. These markers will require audit and special data collection, but routinely published statistics give some indication of the outcome of treatment.

The proportion of patents admitted for stroke who die within 30 days of admission, the proportion of patients discharged after admission for stroke who are readmitted within 30 days and the proportion of patients discharged to their former place of residence after admission for stroke are all relevant outcomes we can measure now. The bottom chart shows that these vary between regions, but at primary care trust level the variation is much greater. If we are to meet the challenges set by the next stage review, these are some of the outcomes where we must reduce unacceptable variations at every level.