Elsevier

The Lancet Neurology

Volume 6, Issue 12, December 2007, Pages 1106-1114
The Lancet Neurology

Personal View
Lifetime risk of stroke and dementia: current concepts, and estimates from the Framingham Study

https://doi.org/10.1016/S1474-4422(07)70291-0Get rights and content

Summary

The main neurological causes of morbidity and mortality are stroke and dementia. We contend that the most relevant and readily communicated risk estimate for stroke and dementia is lifetime risk, which is the probability of someone of a given age and sex developing a condition during their remaining lifespan. Lifetime risk estimates describe the population burden; however, they can be refined with risk-stratified models to enable individual risk prediction. Community-based data on a group of North Americans of European descent indicate that the lifetime risk of stroke for a middle-aged woman is 1 in 5 and for a middle-aged man is 1 in 6. The lifetime risk of stroke was equal to the lifetime risk of dementia and equal to or greater than the lifetime risk of Alzheimer's disease (1 in 5 and 1 in 10 for women and men, respectively), and the lifetime risk of stroke or dementia was greater than 1 in 3. Thus, the lifetime burden attributable to common neurological disease is immense.

Introduction

In the past century, the worldwide average life expectancy at birth rose sharply from less than 40 years in 1900 to more than 65 years in 1975; more of the population now live to be 80 years or older. A corollary of increased life expectancy is an increase in the individual and population burden of age-related diseases. Stroke and dementia are the most widely feared age-related neurological diseases, and are also the only neurological disorders listed in the ten leading causes of disease burden.1 Stroke is the third highest cause of mortality and the highest neurological cause of disability-adjusted life-years lost in the developed world; in developing countries that have a low overall mortality rate, stroke ranks second in disability-adjusted life-years lost.1 The 2007 statistical update from the American Heart Association estimates the lifetime risk of stroke and other cardiovascular diseases on the basis of data from the Framingham Heart Study.2 But what is the lifetime risk of a disease, and do we need this additional descriptive statistic for stroke when we already have the widely used measures of incidence, prevalence, relative risk, and cause-specific mortality?

People should be aware of the risk of a disease at some point in their life; similarly, such statistics are essential for public health planners to estimate the projected disease burden in a population during its expected lifespan. The estimation of lifetime risk enables a long-term perspective, which is particularly important for conditions such as stroke where exposure to risk factors in midlife can alter the incidence of disease later.3 The lifetime risk of a condition can be defined as the probability that a person who is currently free of the condition will acquire it at some time during the remainder of their expected lifespan. The concept of lifetime risk was pioneered by cancer epidemiologists4, 5, 6 and later extended to chronic bone7 and mental health disorders.8 The term lifetime risk was used to describe the diagnosis of a specific disease (such as breast cancer5 or Alzheimer's disease [AD]9), a specific disease event (such as hip fracture10 or stroke11), or the development of a risk factor (such as hypertension12 or obesity13). Lifetime risk was used to define risk from birth14 or, more frequently, to define the residual or remaining lifetime risk from an age when that risk becomes of relevance and of concern to a person or to a society, such as the remaining lifetime risk (or the lifetime risk) of breast cancer in adults, the risk of stroke from age 55 years, or the risk of dementia from age 65 years.

The concept of lifetime risk is underused. Most neuroepidemiological papers address disease incidence and prevalence but do not provide any estimate of the lifetime risk of the condition in their study samples. In this Personal View, we present a conceptual definition of lifetime risk and distinguish it from measures of cumulative incidence and prevalence. We describe how the lifetime risk can be operationally defined and computed, and discuss data on the lifetime risks of stroke, dementia, and AD from one well characterised, community-based sample. In addition, we discuss the use of such estimates for health planning, public education, and the prediction of a person's risk. We also set out the reasons and preliminary evidence for the expected regional and temporal differences in lifetime risk, and summarise some of the available data on the lifetime risk of various neurological diseases. The lifetime risk estimates for other common age-related diseases, such as cancer and cardiovascular disease, are compared with the lifetime risk of neurological disorders. With this Personal View, we aim to persuade more investigators to use lifetime risk estimates.

Section snippets

Why define the lifetime risk of a condition?

A brief historical perspective that describes the development and application of lifetime risk in the context of the Framingham database might be helpful. We realised the usefulness of the concept of lifetime risk in the early 1990s when neurologists were grappling with a simultaneously exciting and yet disappointing observation. An unexpected new risk factor—the apolipoprotein E (APOE) ɛ4 allele—was found to have a powerful effect on the age of onset and risk of AD in familial and apparently

What is the lifetime risk of a disease?

The lifetime risk of a disease is clearly not synonymous with the age-specific annual incidence of the disease because lifetime risk is the risk over a long period rather than the risk over a year. The lifetime risk differs from conventional estimates of cumulative incidence (or cumulative risk) whenever we regard a disease that is common—with a prevalence greater than 10%—or a group in which there is a high risk of competing events (typically death by a different cause) that remove people from

How can we estimate the lifetime risk?

Operationally, the lifetime risk or, more accurately, the mortality-adjusted residual or remaining lifetime risk at any age is the conditional probability of developing a disease when a person has reached the baseline (index) age and is free of that disease. The conditional probability can be estimated on the basis of the mean age-adjusted incidence of the disease and the average residual life expectancy for each sex at a particular age. In 1994, when we first approached this question, there

What is the lifetime risk of stroke?

The Framingham Study has prospectively screened participants for incident stroke since the original group was enrolled in 1948–50.21, 22, 23 With data from the 4897 original participants in the Framingham Study who were stroke-free and dementia-free at 55-years-old, 875 incident strokes were recorded during a follow-up period of up to 51 years (115 146 person-years); therefore, we estimated that the average 55-year-old man has a 1 in 6 risk of stroke during his lifetime, whereas the risk for an

Lifetime risks of stroke versus dementia

We compared the lifetime risks of stroke and dementia and noted that lifetime risk of stroke is greater than lifetime risk of AD until people reach an advanced age (∼85 years old). We made this comparison because dementia, which is mostly due to AD, is the only other neurological disorder listed in the ten leading causes of total disease burden in developed countries, and these are the two most widely feared age-related neurological disorders in the public perception. A group of 2794

Lifetime risks in women

The greater life expectancy for women translates into a greater lifetime risk of several diseases, which is true not only for disorders that are known to affect women at least as frequently as men (such as AD) but is also true for vascular diseases (such as stroke), where premenopausal, endogenous oestrogen concentrations might protect women. The mean age of first stroke is later in women than it is in men;29, 30, 31 however, because the average adult now spends more than 25% of their life in

Patterns in lifetime risk estimates of disease

During the past decade several investigators have extended the concept of estimating lifetime risks to various cardiac disorders, such as coronary heart disease,33 congestive heart failure,34 and atrial fibrillation.35 Lifetime risk estimates fall into two broad categories on the basis of the pathophysiology of the underlying disease. One group of diseases has a peak incidence within a given age group, with decreased age-specific incidence in older people. For these diseases, the lifetime risk

Individualised risk prediction

People who are apprehensive about the risk of a disabling and potentially fatal condition, such as a stroke, should be rightly concerned with not just the annual or 5-year risk, although these short-term risks are important, but also with the risk that they will develop the condition at any time in their life. Someone who is worried about the risk of stroke could use a predictive scale, such as the Framingham stroke risk profile, to predict the 10-year risk of stroke.31 However, risk prediction

Population differences in lifetime risk estimates

Lifetime risk estimates are a useful public health tool that can be estimated in any population for which we have data on disease incidence and cause-specific mortality. More estimates of lifetime risk from racially, geographically, and socioeconomically diverse populations are needed because there are acknowledged differences in stroke incidence and life expectancy among ethnic groups, countries, and, possibly, socioeconomic groups.45, 46, 47 Lifetime risk can vary with the differences in life

Temporal trends in lifetime risk of stroke

Lifetime risk is not static over time. Temporal trends in these risks should be expected because life expectancy rises or falls, the number of risk factors change, and as diagnostic criteria and the sensitivity of diagnostic tests for a condition change. With the Framingham data,32 we noted that there was a decreasing trend in the lifetime risk of stroke at age 65 years across the three periods analysed (1950–1977, 1978–1989, and later than 1990) that did not reach statistical significance for

Lifetime risk of risk factors for stroke

Lifetime risk estimates are a useful measure of risk factor burden in populations. For some of these reversible disorders, such as obesity and hypertension, the prevalence can change as people revert from a disease condition to a healthy state. However, people who have been exposed to these risk factors have an increased risk even after they revert to a healthy state.3 Studying the lifetime risk of developing risk factors for stroke and vascular disease, such as hypertension,12 obesity,13

Lifetime risk of other neurological disorders

The lifetime risks of other neurological conditions apart from stroke and dementia have been reported; these include Parkinson's disease55 and carpal tunnel syndrome in people with diabetes.56 The lifetime risks of common neurological disorders have been compared with the risks of coronary heart disease and all cardiovascular diseases, the risks of common cancers (such as breast, prostate, and lung cancer) and all deaths from cancer, and with other age-related conditions, such as hip fractures,

Conclusions

The lifetime risk of stroke, AD, or other disorders is an aggregate estimate of actual risk during the remaining lifespan that is applicable to people of a particular age and sex. The risk can be refined, however, for an individual on the basis of their risk factor profile. For an individual, the lifetinme risk is the appropriate a priori risk estimate to determine the usefulness and cost-effectiveness of additional prognostic tests and algorithms. As multiple genetic, metabolic, imaging, and

Search strategy and selection criteria

References for this Personal View were identified by searches of PubMed from 1966 until August 25, 2007, with the terms “lifetime risk” and either “stroke” or “dementia”, which found 151 and 119 articles, respectively. Articles were also found by a review of all publications on the topics of “stroke” and “lifetime risk” from the Framingham Heart Study, through a review of citations in the primary articles and the Science Citation Index, and a review of articles that cited the identified

References (60)

  • World Health Organization. Major burden of disease–leading 10 selected risk factors and leading 10 diseases and...
  • W Rosamond et al.

    Heart disease and stroke statistics—2007 update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee

    Circulation

    (2007)
  • S Seshadri et al.

    Apolipoprotein E epsilon 4 allele and the lifetime risk of Alzheimer's disease. What physicians know, and what they should know

    Arch Neurol

    (1995)
  • RE Cashman et al.

    The probability of developing cancer

    J Natl Cancer Inst

    (1956)
  • EJ Feuer et al.

    The lifetime risk of developing breast cancer

    J Natl Cancer Inst

    (1993)
  • AP Bender et al.

    A standard person-years approach to estimating lifetime cancer risk. The Section of Chronic Disease and Environmental Epidemiology, Minnesota Department of Health

    Cancer Causes Control

    (1992)
  • M Kramer et al.

    The lifetime prevalence of mental disorders: estimation, uses and limitations

    Psychol Med

    (1980)
  • S Seshadri et al.

    Lifetime risk of dementia and Alzheimer's disease. The impact of mortality on risk estimates in the Framingham Study

    Neurology

    (1997)
  • LJ Melton

    Lifetime risk of a hip fracture

    Am J Public Health

    (1990)
  • S Seshadri et al.

    The lifetime risk of stroke: estimates from the Framingham Study

    Stroke

    (2006)
  • RS Vasan et al.

    Residual lifetime risk for developing hypertension in middle-aged women and men: the Framingham Heart Study

    JAMA

    (2002)
  • RS Vasan et al.

    Estimated risks for developing obesity in the Framingham Heart Study

    Ann Intern Med

    (2005)
  • H Bryant

    Breast cancer in Canadian women

    BMC Women's Health

    (2004)
  • EH Corder et al.

    Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer's disease in late onset families

    Science

    (1993)
  • DA Drachman

    If we live long enough, will we all be demented?

    Neurology

    (1994)
  • LJ Schouten et al.

    Cancer incidence: life table risk versus cumulative risk

    J Epidemiol Community Health

    (1994)
  • JC Breitner

    Clinical genetics and genetic counseling in Alzheimer's disease

    Ann Intern Med

    (1991)
  • A Beiser et al.

    Computing estimates of incidence, including lifetime risk: Alzheimer's disease in the Framingham Study. The Practical Incidence Estimators (PIE) macro

    Stat Med

    (2000)
  • TR Dawber et al.

    Epidemiological approaches to heart disease: the Framingham Study

    Am J Public Health

    (1951)
  • WB Kannel et al.

    Vascular disease of the brain - epidemiologic aspects: the Framingham Study

    Am J Public Health

    (1965)
  • Cited by (251)

    • Exploring cognitive progression subtypes in the Framingham Heart Study

      2024, Alzheimer's and Dementia: Diagnosis, Assessment and Disease Monitoring
    View all citing articles on Scopus
    View full text