Prevalence of Stroke and Stroke-Related Disability
To the Editor:
Bonita and colleagues1 provide valuable information on the prevalence of stroke and the proportion with persisting sequelae. Their conclusions, drawn on the basis on an actuarial model, agree well with the results from a population-based survey from Nord-Trøndelag County, Norway.2 This study comprised 74 977 persons, including permanent nursing home residents; the attendance rate was 88.1%. The raw prevalence rate of stroke was 1850 per 100 000 in the population aged ≥20 years and 960 per 100 000 when standardized to the entire European population. However, when the sensitivity and specificity of the screening question3 are taken into account, one may have to adjust the prevalence estimate downward to approximately 1100 per 100 000.4 In the Table⇓, age-specific prevalence estimates from the Nord-Trøndelag and the Auckland studies are compared.
The Nord-Trøndelag study2 provides figures for self-reported motor impairments, whereas the Auckland study1 reports the prevalence rate of patients with self-reported incomplete recovery and those who need help in activities of daily living. The results of the two studies (for both sexes combined) are summarized in the Table⇓. Despite focusing on different sequelae, the results are similar, especially in the older age groups. In the younger age groups, both studies have low numbers of events and thus less confident estimates.
The credibility of such estimates increases when similar results are obtained by two different methods. The prevalence of stroke seems to be considerably higher than reported in older studies with more highly selected populations,5 and it increases from about 1% at age 50 to about 10% in the age group over 80 years. Approximately one in three of the younger patients and three in four of the older patients have persisting impairments and disabilities from the combined effect of stroke and other chronic diseases.6
- Copyright © 1998 by American Heart Association
Bonita R, Solomon N, Broad JB. Prevalence of stroke and stroke-related disability: estimates from the Auckland Stroke Studies. Stroke.. 1997;28:1898–1902.
Wyller TB, Bautz-Holter E, Holmen J. Prevalence of stroke and stroke-related disability in North Trøndelag county, Norway. Cerebrovasc Dis.. 1994;4:421–427.
Wyller TB, Ranhoff AH, Bautz-Holter E. Validity of questionnaire information from old people on previous cerebral stroke. Cerebrovasc Dis.. 1994;4:57–58.
Wyller TB. Egenrapportering av hjerneslag: validitet og konsekvenser for prevalensestimater. [Self-reported cerebral stroke: validity and consequences for prevalence estimates. In Norwegian with summary in English.] Nor J Epidemiol.. 1995;5:84–87.
Malmgren R, Bamford J, Warlow C, Sandercock P, Slattery J. Projecting the number of patients with first-ever strokes and patients newly handicapped by stroke in England and Wales. BMJ.. 1989;298:656–660.
It is encouraging to see an increasing number of well-designed population-based studies measuring the incidence rate and case fatality of stroke. However, determining the prevalence of stroke, a more useful measure for estimating the impact of stroke on community services and family caregivers, presents particular problems.
Increasing efforts are being made to validate self-reported history of stroke. As the Nord-Trøndelag study indicates, information elicited by self-report overestimates stroke prevalence by about one third.R1 Because many people recover from stroke, estimates that do not incorporate measures of ongoing disability related to stroke sequelae further overestimate the services required for stroke patients.
While it would be comforting to conclude that similar results between the Nord-Trøndelag study and the Auckland Stroke study reflect the true prevalence of strokeR1 R2 it is equally possible that the similarities could have occurred by chance. The lack of confidence intervals around the rates presented in the Table⇑ do not allow precise comparisons. In addition, differences in study design as well as definitions of disability also hamper comparisons. For example, the Auckland study used conservative cut points. Incomplete recovery included people who had no ongoing motor deficit; people requiring assistance in any one activity of daily living were included in the more disadvantaged group. The Nord-Trøndelag study included only patients with severe motor deficit as the more severe group. Although both studies have attempted to refine earlier stroke prevalence estimates, neither have succeeded in separating stroke-related disability from disability associated with other concomitant disabling conditions.
The challenge is for stroke epidemiologists to develop a consensus concerning the measures that best reflect stroke-related disability and how such definitions can be usefully incorporated into future studies in a standard manner. Dr Wyller and colleagues have contributed to the need for a debate on the issue.
Wyller TB, Ranhoff AH, Bautz-Holter E. Validity of questionnaire information from old people on previous cerebral stroke. Cerebrovasc Dis. 1994;4:57-58.
Bonita R, Solomon N, Broad JB. Prevalence of stroke and stroke-related disability: estimates from the Auckland Stroke Studies. Stroke. 1997;28:1898-1902.