from 132 in Northumberland to 75 in NW Hertfordshire (national SMR =100)8 . This implies certain areas would have more difficulty in achieving mortality reduction targets and local knowledge of incidence and case-severity will need to be addressed. In terms of years of life lost as a result of stroke, in England and Wales in 1993-1994, an average of 28 years of life were lost per 10,000 population and this varied from 21.6 in the South and West region to 34.9 in the North West region8 . There would appear to be inequities in mortality from stroke between social classes. Kunst et al showed that in all countries for men, manual classes had higher stroke mortality rates than non-manual classes, this inequity being relatively large in the UK, Ireland, Finland and small in Sweden, Norway, Denmark, Italy and Spain. These differences probably represent differences in the prevalence of risk factors and access to health services in the different groups9 . Incidence The incidence of stroke is defined as the number of first in a life-time strokes occurring per unit time. It is a sensitive measure of the need for stroke services, but is difficult to estimate without considerable resource. The incidence of all acute strokes (first and recurrent) is in the region of 20-30% higher than the first in a life-time. Bonita has estimated that the risk of a person 45 years of age having a stroke within 20 years is very low (about 1 in 30)10 . However, almost one in four men and nearly one in five women aged 45 years can expect to have a stroke if they live to their 85th year. Although the life-time risk of having an acute stroke is higher in men than women, the converse is true for the life-time risk of dying of a stroke. Thus about 16% of all women are likely to die of a stroke compared with 8% of men; this difference is largely attributable to the higher mean age at stroke onset in women, and to their greater life expectancy. by guest on 17 April 2022 Stroke Sociodemographic influences on incidence The incidence of stroke doubles with each successive decade over the age of 55 years, with an overall rate 0.2/1000 in those aged 45-54 years and 10/1000 in those aged over 85 years. Men have a 25-30% increased chance of having a stroke. African-Caribbean and African men and women have approximately double the risk of stroke compared to the Caucasian population. People in the lowest social class have a 60% increased chance of having a stroke compared to those in the highest social class11 . Subarachnoid haemorrhage The incidence is about 9-14 per 100,000 per year3 . Other published estimates are as high as 33 per 100,000 per year for men and 25 per 100,000 per year in women. Cerebral infarction There have been many population-based studies of stroke, most having had significant methodological flaws12 . Although the Oxford Community Stroke Project is the gold standard for incidence studies in the UK, it commenced in the mid-1980s when mortality rates were higher and the study area was predominantly in rural Oxfordshire with no ethnic minority groups. The overall crude incidence of first in a life-time stroke was 2.4 per 1000 per year3 . A south London register reported an overall crude incidence rate of 1.3 per 1000 population (1.28 male. 1.33 female) with a 2.2-fold increased risk in the Black population in 199513 . Studies published or presented at conferences covering the years 1995-1997 indicate the incidence rates to be between 119-203 per 100,000 adjusted to the European population, but with significant differences between and within countries. Incidences of subtypes of cerebral infarction Intracerebral haemorrhage (excluding SAH) accounts for just over 10% of all stroke, the remainder being cerebral infarction. Using the Bamford classification, the following proportions of first strokes can be expected: cerebral infarction 76% (partial anterior circulation 56%, lacunar 20%, total anterior circulation 15%, posterior circulation 8%, unclassified 1%), primary intracerebral haemorrhage 10%, subarachnoid haemorrhage 4%, not known 10%6 . 278 Bntoh Medical Bulletin 20OO;56 (No 2) Downloaded from https://academic.oup.com/bmb/article/56/2/275/303250 by guest on 17 April 2022 The impact of stroke Recurrence The cumulative risk of recurrence over 5 years is high, ranging from 15—42% in community studies and the pathological subtype of recurrence is the same as the index stroke in 88% of cases14 . Case fatality Case fatality measures the proportion of people who die within a specified period after the stroke; comparisons are based on the first-ever stroke in a life-time since recurrent strokes have a higher case fatality. One month case fatality rates are dependent on the age structure and health status of the populations studied and vary from 17-49% amongst men in the MONICA studies and 18-57% in women with an average of about 24% from the literature15 . In the UK, the Oxford 28-day case fatality was 19% overall, that for cerebral infarction being 10%, primary intracerebral haemorrhage 50%, and subarachnoid haemorrhage 46%3 . Studies reporting rates in the 1990s estimate one month case-fatality of 19-28% and one year case-fatality as 34—41 %n . Trends in stroke incidence and case fatality Incidence Case fatality There have been few stroke incidence registers that have been maintained over long enough periods of time to document a change in