Intended for healthcare professionals

Education And Debate

Healthy aging

BMJ 1997; 315 doi: (Published 25 October 1997) Cite this as: BMJ 1997;315:1090
  1. Kay-Tee Khaw, professor of clinical gerontologya (cthl{at}
  1. a Clinical Gerontology Unit Box 110, University of Cambridge School of Clinical Medicine, Addenbrooke's Hospital, Cambridge CB2 2QQ


    A major challenge facing society is how we can maintain health and quality of life in an aging population. Maximum life expectancy has not changed substantially, but average life expectancy has increased greatly in the past century. This reflects profound improvements in mortality in infancy and young adulthood, resulting in a much greater proportion of people surviving to older ages (tables 1, 2, and 3).

    Percentage of people surviving to,and further expectation of life from, age 55 and age 75, by year of birth,1841−19811

    View this table:
    Table 2

    Estimated number (percentage) in the population aged 60 years over in UK2

    View this table:
    Table 3

    Expectation of life for people of various ages, according to death rates assumed for remainder of their lifetimes, United Kingdom2

    View this table:

    Aging and disability

    The rise in numbers and proportion of older people has led to much concern about societal consequences, not least health consequences. Increasing age is associated with increasing disability and loss of independence, with functional impairments such as loss of mobility, sight, and hearing. In Britain in 1984-5, the estimated prevalence of those with severe disability was less than 1% in those aged 50-59 years but 13% in those aged over 80 years.3 Murray and Lopez have estimated that at age 60, we might expect to live about a quarter of our remaining years with some disability.4 5 If the average age of onset of ill health is unchanged, increased life span would mean more years of ill health before death for an individual and a greater proportion of people with disability. Much current discussion thus revolves around how best to support and care for large numbers of older people with disability. According to Fries, the age of onset of ill health might, however, rise more quickly than our life span increases, resulting in “compression of morbidity” (a shorter period of disability and ill health before death).6 7 So how far can we reduce, or postpone the onset of, disability that is associated with age?

    Summary points

    Healthy life expectancy is influenced by a relatively limited number of chronic disabling conditions

    A substantial proportion of these chronic disabling conditions can be prevented or postponed

    A greater focus is needed on prevention and health maintenance—much is already known about the impact of modifiable influences such as diet, physical activity, smoking, infection, pollution, and housing

    The social framework and policies that enable individuals to fulfil their potential and attain optimal health are crucial

    Can we prevent age related disability?

    Fries has suggested that people of high socioeconomic status, with more education, or with particular lifestyles (such as those who are physically active) seem to experience compression of morbidity.7 Indeed, even in Britain, there is evidence of secular improvement. The proportion of men unable to perform four activities of daily living at any specified age has halved between 1976 and 1994 (fig 1).1 8

    Variation in rates of chronic disease

    Healthy life expectancy is determined by a relatively limited number of chronic conditions that become more common with increasing age. Their exact contributions vary according to definitions of disability, but all estimates include cardiovascular disease, such as coronary heart disease and stroke; musculoskeletal diseases, such as arthritis and osteoporosis leading to fractures; neurodegenerative disorders, such as memory loss and dementia; neuropsychiatric orders, such as depression; cancers, including lung, breast, prostate and colorectal cancers; and other degenerative conditions, such as visual loss from cataracts, macular degeneration and glaucoma, and hearing loss. Reduction or postponement of these conditions may not only reduce premature death and increase longevity but, more importantly, may also decrease the period of illnesses so that people can remain healthy until near death.

    Fig 1
    Fig 1

    Percentage by age of men unable to perform four activities of daily living in 1976, 1985, and 19941 8

    The great variation in rates of chronic diseases in different communities shows that a substantial proportion of the chronic diseases associated with aging can be prevented, or at least postponed. Figure 2 gives examples of the relation between age and rates of some chronic disabling conditions in different countries; mortality has been used as a surrogate for incident morbidity data, which are often not available. Japanese men and women show much less increase with age for these conditions than men and women in Britain, with Spain having rates in between.9 Similar age related patterns can be seen for many other conditions, including other cancers, heart disease cataracts and glaucoma. Secular trends and changes with migration show that these large differences are likely to be due to environmental rather than genetic factors. Over the past 30 years, age specific rates for cardiovascular disease have halved in Japan and the United States but doubled in Hungary.10 Rates for hip fracture have doubled in several countries, including Britain (fig 3), Hong Kong, and Sweden.11 12 These trends have a huge impact. In Britain, for example, about 50 000 hip fractures occur annually. This number is projected to increase to 120 000 by 2020, but if age specific rates could return to those of 1950, most of this epidemic could be averted. In Britain wide regional and social class variations exist for total mortality, which is often used as an indicator of overall health status; throughout adult life, men in social class V have about twice the mortality of men in social class I.13

    Fig 2
    Fig 2

    Mortality in United Kingdom, Spain, and Japan 1992-4 for ischaemic heart disease in men, breast cancer in women, and prostate cancer9

    Fig 3
    Fig 3

    Annual incidence by age for fractured proximal femur in men in Oxford, 1954-8 and 1983

    Variation in age related physiological decline

    The rate at which many physiological functions change with age, and hence the occurrence of their clinical consequences, varies greatly in people in different circumstances. In communities worldwide, blood pressure or concentrations of cholesterol or blood glucose do not increase markedly with age; thus, rises in these factors are not necessary concomitants of the aging process. These communities also have low rates of cardiovascular disease even among elderly people. Migration studies show that the primary determinants are environmental not genetic. Members of the Luo tribe in Kenya had low blood pressures when living in a rural environment but higher blood pressures that increased with age when they moved to urban Nairobi.14 Japanese people have much lower blood cholesterol concentrations and lower coronary heart disease rates than white Americans, but Japanese migrants in the United States have much higher cholesterol concentrations and rates of heart disease.15 White people with diets of Buddhist monks had low blood cholesterol concentrations similar to those of Japanese people living in Japan, but Japanese people in the United States army with an American diet had high cholesterol concentrations resembling those of white people in the army.10 16 Dutch women living in the Antilles had high bone mass and lower rates of fractures than Dutch women of the same age in the Netherlands.17

    Genes and environment

    Many of the diseases that we associate with aging reflect deterioration of physiological functions. As we age, blood pressure and blood cholesterol concentration usually increase, leading to increased risk of heart attacks and strokes; glucose tolerance declines and insulin resistance increases, leading to diabetes; intraocular pressure increases, leading to glaucoma and visual loss; and immune function deteriorates, resulting in increased risk of infections and possibly some cancers. Loss of bone mass increases fracture risk, neuronal degeneration results in loss of cognitive function and dementia, cartilage degeneration in arthritis, and muscle loss in functional weakness.

    Numerous theories explain the aging process; most suggest that senescence results from the accumulation of unrepaired damage. Kirkwood and Wolff suggest that the different life spans of different species reflect differing distributions of investment in allocation of metabolic resources between maintenance and reproduction, which have evolved in response to extrinsic circumstances.18 Their theory predicts that, rather than any single mechanism for aging, the mechanisms responsible are those types of damage for which maintenance and repair are metabolically costly, such as damage to and mutations of DNA, defective mitochondria, oxidative damage by free radicals, and accumulation of aberrant proteins. Whatever the possible mechanisms, these seem to result in a progressive generalised impairment of functions and loss of adaptive response to stress, a growing risk of chronic diseases, and an increase in the probability of dying. The biological process of aging and its clinical manifestations reflect the interaction between our genetic inheritance and the environment. Indeed, we know that huge variation in phenotype exists in people with the same genes and that gene expression and function is profoundly modified by environmental factors. Thus, while the maximum life span is probably genetically determined, the likelihood of reaching that life span in good health seems to be largely determined by environmental and lifestyle factors.

    Model for intervention strategies

    Evidence confirms that throughout life our chances of aging successfully can be increased in various ways. In early life the intrauterine and early postnatal environment may programme basic metabolic processes and hence susceptibility to various conditions such as cardiovascular disease and diabetes in later life, and Barker has suggested that maternal nutrition plays a critical role.19 Growth and development of vital organs such as brain, muscles, bone, and blood vessels during childhood and early adulthood build reserves that may affect later capacity. For example, pattern of fatty acid intake, or cognitive exposures in infancy influence brain development, and calcium intake and physical activity influence bone mass in youth, the consequences of which may extend well into old age. In later life, strategies may be to reduce damage (for example, from infections or toxins), to increase protection against damage (for example, by increasing antioxidant defences or strengthening immune function), or to prevent loss through lack of use (for example, by remaining physically and mentally active) (box).

    Potential health interventions at different stages of life

    View this table:

    The “free radical” and “immune function” theories of aging both give clues as to possible interventions. Free radicals are reactive molecules produced as byproducts of cell metabolism that cause oxidative damage to cell components, including proteins, nucleic acids, and membranes. Damage by free radicals in different tissues has been postulated to be responsible for conditions as diverse as cancers, respiratory disease, dementia, cardiovascular disease, and eye diseases including macular degeneration and cataracts. Many exposures such as infection, toxins, smoking, or high dietary saturated fat load are associated with increased production of and damage by free radicals; antioxidants such as vitamin C, carotenoids, vitamin E, or selenium may mitigate such damage.

    The immune theory suggests that functional capacity of the immune system declines throughout life, with involution of the thymus gland and deterioration of stem cells; this is associated with an increased incidence of infections, cancer, and other immune-complex type diseases. Exercise, smoking, and nutrition such as zinc, vitamin A, or pyridoxine and riboflavin can affect immune function.20

    Preventing cardiovascular disease and osteoporosis

    Cardiovascular disease

    Risk of cardiovascular disease rises incrementally with increasing level of blood pressure and cholesterol concentration.21 22 23 Intervention trials have shown that reduction of blood pressure by 6 mm Hg reduces risk of stroke by 40% and of heart attack by 15% and that a 10% reduction in blood cholesterol concentrations will reduce risk of coronary heart disease by 30%.24 25 Reducing blood pressure and cholesterol concentrations in older people could have a substantial effect on reducing the burden of cardiovascular disease. The exact blood pressure or cholesterol concentration at which drug treatment is considered warranted is still debated. Observational studies and trials, however, have implicated high dietary saturated fat as a cause of high cholesterol concentration and rates of coronary heart disease, and high dietary sodium intake in the aetiology of the age related rise in blood pressure and stroke, and have shown that relatively modest reductions in saturated fat and salt intake may greatly affect cardiovascular disease.26 27 28 The feasibility and achievability of such changes in diet and cardiovascular disease are well supported by the existing large international and secular variations. Dietary changes seem to affect risk factor levels throughout life and may have even more impact in elderly people. For example, reducing dietary sodium lowers blood pressure more in elderly people than in younger people. Many other dietary and other factors such as cigarette smoking, physical activity, infection, and psychosocial stress also seem to influence cardiovascular risk throughout life through various mechanisms, such as haemostatic tendency, inflammatory processes, or homocysteine metabolism, not all of them well understood or tested in trials. Again, the available evidence shows that the potential impact may be substantial: increasing fruit and vegetables by 1-2 servings daily may decrease cardiovascular risk by 30%.29 Fruit and vegetables may act through various mechanisms including increasing folic acid and lowering homocysteine levels, increasing vitamin C and flavonoids and thus antioxidant defences, and increasing minerals such as potassium and magnesium, possibly reducing blood pressure. Secondary prevention trials may be more generalisable to elderly people, many of whom may have established diseases. A secondary prevention trial of advice to eat fatty fish twice a week reduced cardiovascular death by 30%30; another secondary prevention trial of Mediterranean diet (essentially substituting foods rich in α linolenic acid for dairy and animal fats and increasing intake of bread, fruit, and vegetables) reduced mortality by 70% after four years.31

    Embedded Image

    Embedded Image


    Low bone mass increases risk of fracture,32 but bone loss and subsequent risk of fractures including vertebral and hip fractures in elderly women or in those with existing fractures can be reduced by 30-50% by various treatments including vitamin D and calcium supplementation or bisphosphonates.33 34 As with cardiovascular disease, preventive drug treatment in the whole older population is debatable because of risk-benefit issues, but risk of fracture can be reduced by ensuring that older people have adequate dietary intake of calcium and vitamin D.

    Resolutions when I come to be old

    Not to marry a young Woman.

    Not to keep young Company, unless they really desire it.

    Not to be peevish, or morose, or suspicious.

    Not to scorn present Ways, or Wits, or Fashions, or Men, or War, &c.

    Not to be fond of Children.

    Not to tell the same Story over and over to the same People.

    Not to be covetous.

    Not to neglect decency, or cleanliness, for fear of falling into Nastiness.

    Not to be over severe with young People, but give allowances for their youthful follys and weaknesses.

    Not to be influenced by, or give ear to knavish tattling servants, or others.

    Not to be too free of advice, nor trouble any but those that desire it.

    To desire some good Friends to inform me which of these Resolutions I break, or neglect, & wherein; and reform accordingly.

    Not to talk much, nor of myself.

    Not to boast of my former beauty, or strength, or favour with ladies, &c.

    Not to hearken to Flatteries, nor conceive I can be beloved by a young woman, et eos qui hæreditatem captant, odisse ac vitare.

    Not to be positive or opiniative.

    Not to set up for observing all these Rules, for fear I should observe none. Jonathan Swift, 1667-1745

    Interventions to improve health in later life

    Cardiovascular disease and osteoporosis are examples of conditions for which substantial trial evidence exists of the effectiveness and magnitude of impact of preventive interventions in later life; however, since many conditions coexist in elderly people, the impact of interventions on overall health is also crucial.

    For most other chronic diseases, a wealth of evidence implicates the substantial role of environmental (including lifestyle) factors, though for most, data from trials are not available. The box summarises some of these factors. Tobacco smoking must be the single most preventable cause of ill health and disability35; the benefits of not smoking in terms of respiratory function and cardiovascular disease are apparent even at older ages.

    Interventions to improve health in later life

    View this table:

    Nutrition clearly has a key role in health throughout life—from maternal nutrition and fetal metabolic programming, to childhood nutrition affecting growth and development, to diet in later life influencing maintenance of health. Caloric restriction is often believed to delay aging, because of experiments reporting that severe food restriction increased longevity in surviving rats. However, findings from studies of rats in strictly controlled and protected experimental conditions are not easily generalisable to humans,36 and prospective population studies have shown an inverse relation between mortality and caloric intake.37 Higher caloric intake associated with better health outcome in humans may reflect higher levels of physical activity (which may be beneficial) or higher intake of protective nutrients.38 39 Indeed, aging may be associated with less efficient processing of essential nutrients—for example, poorer ability to synthesise vitamin D in the skin, and poorer ability of the gut to absorb nutrients—so older people may need higher intakes of particular nutrients.

    Numerous other nutrients have been the focus of interest, including the B vitamins such as B-12 and pyridoxine (which have been implicated in neurological function) and folate and possibly riboflavin (involved in homocysteine metabolism). Unfortunately, trials of antioxidant vitamin supplementation have been largely discouraging; several ß carotene supplementation trials show no effect or adverse effects on cardiovascular disease, cancer, or total mortality40 41; and a trial of vitamin E in secondary prevention of coronary heart disease which reduced non-fatal events by 70% had no, or possibly, an adverse, effect on mortality.42 Nevertheless, high fruit and vegetable intakes have been most consistently associated with protective benefits in several conditions, including macular degeneration, visual loss, cataracts, respiratory disease, and cancers such as breast, stomach, and colorectal.43 44 45 The discrepancy between foods and isolated supplementation may be that many other nutrients in food or their interactions are responsible for the clinical effects.

    Although we may not know precisely which nutrients are responsible for which particular actions, the evidence is overwhelming that particular dietary patterns do seem to relate to good health.46 47 48 49 The 1994 recommendations of the Committee on Medical Aspects of Food Policy emphasise the importance of adequate intake of nutrients such as vitamins and minerals and of −3 fatty acids, which can be achieved by diets high in fruit, vegetables, and complex carbohydrates or plant polysaccharides such as rice, bread, and pulses.49 Conversely, reduction in dietary sodium and saturated fat or transfatty acid intake can be achieved by reductions in intakes of certain processed foods (their major source in Western diets) or replacement with low fat foods or oils rich in unsaturated fatty acids. These are diets characteristic of Japan and Crete, which now have greatest average life expectancies.

    Obesity is a risk factor for many chronic diseases. Although studies give varying values for ideal weights in older people, extremes of weight (very high (>30) or very low (<18) body mass indices (kg/m2) are adversely associated with health in most populations; to maintain adequate nutrition while keeping body mass index within the (wide) desirable range implies regular physical activity.

    Indeed, physical activity seems to protect against many diseases, such as cardiovascular disease, osteoporosis and fractures, diabetes, and breast and colon cancer.50 51 Exactly how much and what sort of physical activity at different ages has particular effects is still unclear, but even moderate activities such as walking, gardening, and keeping generally mobile seem to promote physical and mental functioning and wellbeing.

    Prevention of cognitive loss or dementia poses a particular challenge in elderly people. Some deterioration can be attributed to atherosclerotic disease, and thus interventions such as aspirin or particular dietary patterns that reduce cardiovascular risk may also prevent dementia. High educational status early in life and, intriguingly, continued mental stimulation, also seem protective.

    Many other extraneous factors offer future possibilities for interventions. For example, chronic infections such as chlamydia and helicobacter have been implicated as risk factors for cardiovascular disease and cancer of the stomach, and air pollution is now believed to affect cardiovascular health. Thus, general public health measures in other, not hitherto directly related, areas may have additional benefits for age related chronic diseases.


    For many diseases we do not yet have sufficient evidence to make highly specific recommendations for prevention, but we know the general environmental and lifestyle patterns that can help. However, individuals' ability to make changes to improve their health is determined by the social and cultural context and circumstances including choice, access, availability, information, and income. Environmental determinants such as adequate housing and clean air and water, fundamentals for health, cannot always be taken for granted. The large inequalities in health experienced by social class and region in Britain reflect these varying circumstances.

    Successful aging of course encompasses social as well as physical and psychological wellbeing. The social framework and policies that may enable individuals to fulfil their potential and attain optimal health are crucial.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    14. 14.
    15. 15.
    16. 16.
    17. 17.
    18. 18.
    19. 19.
    20. 20.
    21. 21.
    22. 22.
    23. 23.
    24. 24.
    25. 25.
    26. 26.
    27. 27.
    28. 28.
    29. 29.
    30. 30.
    31. 31.
    32. 32.
    33. 33.
    34. 34.
    35. 35.
    36. 36.
    37. 37.
    38. 38.
    39. 39.
    40. 40.
    41. 41.
    42. 42.
    43. 43.
    44. 44.
    45. 45.
    46. 46.
    47. 47.
    48. 48.
    49. 49.
    50. 50.
    51. 51.
    View Abstract