Select Committee on Science and Technology First Report



2.1.  Over the last two hundred years, life expectancy within industrialised nations such as the UK has doubled. More recently, life expectancy has also begun to increase across the developing world, and most nations are experiencing continuous upward trends in longevity. This astonishing feat—driven primarily by the successes of previous generations in combating early, preventable deaths—now evokes a curiously mixed response.

2.2.  It is hardly possible to open a newspaper without reading of the increase in life expectancy, and the consequently rapidly increasing proportion of older people in the population. More often than not these matters are considered for the economic impact they will have, be it on the cost of healthcare or on pensions. The underlying tone of such discussion is often negative, focusing on the "burden" of increased numbers of older people and the threat of the demographic "timebomb".

2.3.  These economic implications of changing life expectancy are without doubt of great importance, and have been the subject of a number of studies, not least by the House of Lords Select Committee on Economic Affairs.[3] However, when considering the future of ageing and its impacts upon society, it is important to understand what is actually happening to people from a biological and psychological perspective, since it is our bodies and minds—what happens to them and what we do with them—that are driving these changes. This inquiry has therefore considered these matters from the scientific perspective. We have looked at the biological processes of ageing, the application of research in technology and design to improve the quality of life of older people, and the funding and coordination of research in these areas.

2.4.  The urgency of these matters is plain from the statistics. The 2001 Census showed that for the first time the number of people in England and Wales aged 60 and over was greater than the number aged under 16. The figures among the "oldest old" are even more striking. In 1951 there were 0.2 million people aged 85 and over; by 2001 this had grown to over 1.1 million.[4] Figures given at our seminar showed that the proportion of the global population aged 65 and over in 1900 was 1% (UK 5%); in 2000 it was 7% (UK 16%); and by 2050 it is estimated to be 20%, a figure the UK will probably reach in 2020.

2.5.  Until very recently most observers and agencies concerned with forecasting future life expectancy predicted that it would soon reach a plateau, when the gains from preventing early death had been consolidated; the fixed, ineluctable reality of the ageing process would then be revealed. However, this has not happened. Life expectancy in the UK and other developed countries continues to increase by about two years per decade, so that for each hour spent in reading this report, life expectancy will have increased by 12 minutes. From studies in Sweden, where statistics have been kept since 1860, it has been found that the increase in the year-by-year maximum life-span (age of the oldest person), far from slowing down towards a plateau, has been accelerating over the last 20 to 30 years.

2.6.  No sensible work on ageing can be done without a detailed analysis of these demographic changes. We have considered the best available data, and have looked at UK data as compared with other countries. We have also looked at the very marked differences between social classes and between ethnic groups.

Life expectancy in the United Kingdom

2.7.  The most reliable statistics on expectation of life in the UK are those collected and published by the Office for National Statistics (ONS). The figures take some time to collect and evaluate, and an inevitable consequence is that the statistics are based on figures going back a few years. There is however every reason to believe that the trends these figures demonstrate are continuing trends.

2.8.  For males in the UK, life expectancy (LE) at birth has gone up from 70.8 in 1981 to 75.9 in 2002, an increase of 5.1 years; for females the figures are 76.8 and 80.5, an increase of 3.7 years. At age 70, LE for males has gone up from 10.1 to 12.6, for females from 13.3 to 15.2. However these figures mask considerable variations by geographical area. Male LE in 2002 was 76.2 in England, but in Scotland only 73.5. For females the figures are 80.7 and 78.9. Table 1 on page 12 shows life expectancy at different ages in each of the constituent parts of the UK between 1981 and 2002.[5]

2.9.  These figures mask even more startling local variations. In 1996, in 16 local authority areas the LE was still below the average UK LE ten years earlier.[6] Within London, figures for male LE in 1998-2000 show that in Westminster this was 78.4, but just across the Thames in Lambeth it was only 73.5, while in Stratford, ten stops along the Jubilee line from Westminster, it was 72.7.[7] The figures for 2001-03, in table 2 on page 13, show that male LE in East Dorset was 80.1, eleven years longer than in Glasgow City (69.1). Figures for females show similar trends, but not quite so marked.[8]


Expectation of life at birth and selected age


Life Expectancy by Local Authority, 2001-03


2.10.  Social factors play a major part in these local divergences. The following criteria were used by the Registrar-General to define social class:[9]


Definition of Social Class

  In 1996 the difference in LE between Social Classes I and V was 9.5 years for men and 6.4 years for women. The gap between the social classes is apparently narrowing; by 1999 it was down to 7.4 years for men and 3.1 years for women.[10] However care needs to be taken in interpreting underlying trends from short-term differences, when many factors may be at play. For instance, the reduction in the social-class gap LE for women between 1996 and 1999 was caused more by a decrease in the LE of women of Social Class I than by an increase in the LE of those in Social Class V.[11] The causes of these inequalities are many, complex, and interrelated; we consider some of them in the following chapter.[12]


2.11.  The 2001 Census showed that only 14.5% of minority ethnic groups were aged 50+, compared to 35.1% of whites; while only 0.3% of minority ethnic groups were aged 85+, compared to 2.1% of whites. The full figures, in percentages of older people in each ethnic group, were:[13]


Age Distribution by Ethnic Group, 2001

2.12.  This is graphically illustrated by the following chart:[14]


Age Distribution by Ethnic Group, 2001-02

These distributions are to some extent the product of differences in LE, but also reflect the ages at which immigrants enter this country, and the length of time they have been here.

2.13.  Ethnic minorities form a greater proportion of the lower social groups. It is a matter of fact that life expectancy is lower among ethnic minorities, but it is not clear to what extent this is due to the factors affecting all members of the lower social groups. The effect on ageing of distinctions attributable solely to ethnic differences is a matter we consider in Chapter 3.[15]


2.14.  International comparisons are instructive. If the percentage of the population aged 65 and over is taken as a measure, of the 30 countries with the oldest populations, all but four are European. Japan is the only non-European country in the top 17. Italy tops the list with 18.8% of its population 65 and over; the UK is fourteenth with 15.7%. By comparison the United States is a young country, with only 12.4% aged 65 and over.[16]

2.15.  A different but equally instructive measure of population ageing is provided by life expectancy at birth. The following figures[17] show LE, averaged for males and females, in the major countries of the European Union and a number of other selected nations:

New Zealand78.2
United Kingdom78.1
United States77.0
Russian Federation66.7

The significance of these figures is that in the UK, despite its sustained high economic performance, and after half a century of a comprehensive National Health Service, LE still lags slightly behind the average for the high-income OECD States, which is 78.3.

2.16.  At current rates, life expectancy within the UK is increasing at the rate of about two years for each decade that passes. The consequences of this demographic change for all aspects of life are profound. As this Report will show, we have found little evidence that policy has been sufficiently informed by scientific understanding of the ageing process.

Healthy Life Expectancy

2.17.  We believe that increases in life expectancy are truly to be welcomed only if the "added years" are years of relative good health. We therefore also looked at figures which indicate to what extent healthy life expectancy (HLE) has kept pace with LE. We immediately came up against a problem of definition. As with all statistics relating to future projections, there is a margin of error in the figures for life expectancy. But while there is no possibility of error in determining when life begins and ends, good and bad health are far from being absolutes. They are often a matter of impression, and the line between the two is not a sharp one.

2.18.  The ONS gave us the following information about methods of estimating HLE in the UK:


Methods of Estimating Healthy Life Expectancy

2.19.  Professor Sally Davies, the Director of Research and Development at the Department of Health, told us that figures obtained in this way, far from supporting the proposition that HLE is increasing at a faster rate than LE (as one might have hoped), or at least keeping pace with it, is in fact growing more slowly, so that there is a corresponding increase in what might be termed "unhealthy life expectancy" (Q 139). She subsequently drew our attention to the following graph, which strikingly illustrates this:


Life Expectancy and Healthy Life Expectancy

2.20.  If this apparent increase in unhealthy life expectancy is a true reflection of the facts, it is a matter of considerable concern. We wondered however if this might simply be a reflection of the inaccuracy of self-assessment of so vague a concept as ill-health, and we asked the ONS whether this was a reliable method of calculation. They told us:

"A morbidity measure based on self-assessed health provides valuable information on the relevance of illness to the individual. A comparison of evidence by Ilder and Benyamini established that simple measures of self-rated health consistently predict subsequent morbidity, health care use and mortality, after controlling for risk factors and diagnosed conditions. While self-assessed health status is recognised as a valid measure of population health, concerns remain about the reliability of subjective assessments in general. These are known to vary systematically across population sub-groups (by ethnicity and social class) and over time. They reflect difference in ill-health, behaviour, expectations and cultural norms for health.

We know little about the sources of health perceptions, and what people are telling us when they report 'good', 'fairly good' or 'not good' health, or what is the underlying mechanism which consistently links perceived health status with other health-related outcomes. However self-perceived health has been repeatedly shown to be predictive of health outcomes. The most dramatic influence is seen in the association between self-rated health and subsequent mortality."

2.21.  As an explanation of the apparent widening of the gap between LE and HLE, the ONS had this to say:

"There are several theories behind the increase in reported ill-health. It is possible to separate these ideas into the following six broad categories:

  • People have become more sensitive about health or have adopted higher expectations about their health and/or functioning. The result is that conditions that would not earlier have been regarded as problems are now considered to impact on daily living. Any changes in morbidity over time may reflect people's expectations of good health as well as changes in incidence or duration.
  • Economic incentives are persuading people more readily to consider or present themselves as ill. Workers are suspected of having target incomes, below which they are more inclined to work and it is suggested that rising real income has promoted more ill-health time, or higher incidence of injuries.
  • Improvements in survey methods have led to the discovery of a growing proportion of health problems.
  • Diseases, especially chronic diseases, are being detected earlier. The prevalence of those diseases has risen because the point at which sickness is said to have begun is now earlier. People examined or screened and found sick, the 'earlier patients', add to the prevalence of disease. In the case of diabetes, for example, earlier detection allows progress of the disease to be arrested and permits the sufferer to carry out ordinary activities despite to continued presence of the disease.
  • People with ill-health are living longer. Illnesses and injuries that, in former times, were resolved in death are now more often resolved in management of the disease.
  • The last century has seen a substitution in health risks. On the whole, illnesses with a short course and higher incidence, such as infectious diseases, have been replaced by diseases with a long course and low incidence, for example chronic conditions. This leads to an expansion in morbidity prevalence values, even when incidence remains the same over time.

Each hypothesis provides a plausible explanation for increase in the prevalence of ill-health. Untangling the differential trends in incidence, duration and prevalence of ill-health is only possible using longitudinal data. Many of the issues related to observed increases in ill-health rates in cross-sectional surveys will be addressed in the future using results from the English Longitudinal Study of Ageing."

2.22.  These are all possible explanations for a perceived increase in "unhealthy life expectancy". However we have received no specific evidence showing whether the increase is wholly perceived, or a reflection of a true increase in years of ill-health in later life, or (as seems more likely) a combination of the two. In the case of a perceived increase, we do not know to what extent each of the reasons suggested by the ONS is actually responsible for the increase. If there is a true increase in ill-health, no explanation for this has been suggested to us.

2.23.  It appears to us that questions based on self-assessment of such a vague concept as health, while they may enable comparisons to be made with replies to the same questions from different groups or different areas, are simply not sufficiently reliable to provide an effective objective measure of health in old age. Moreover, the questions asked often differ. While the GHS and 1991 Census asked for health to be classed as "good, fairly good, or not good", Department of Health (DoH) surveys have adopted criteria recommended in 1996 by the WHO, and have asked whether health is classed as "very good, good, fair, bad or very bad". Nor can international comparisons usefully be made, since other countries rely on yet other criteria. For example, the United States, Canada and Australia ask whether health is perceived as "excellent, very good, good, fair or poor"; in those countries, those who perceive their health to be "fair" are in the fourth category rather than the third.[18]

2.24.  Ms Susan Lonsdale, Acting Head of the DoH Policy Research Programme, equated healthy life expectancy with disability-free life expectancy (Q 140), but other witnesses drew a clear distinction between the two. Sir John Grimley Evans, Emeritus Professor of Clinical Geratology at Oxford University, told us:

"Most geratologists have since [1984] focused on disability rather than morbidity as the crucial transition, because it is disability and its associated loss of autonomy that older people fear, and which in turn leads to dependency with its cost implications for the health and social services. Moreover, much of modern medicine as applied to older people does not 'cure' disease but relieves its symptoms and other effects and, in the best case, prolongs survival. People may still have the 'disease' in a medical sense but it does not affect their quality of life. Such effects ought surely to register as success in prolonging disability-free life expectancy, rather than as failure in extending the duration of (irrelevant) 'disease' … I would urge the Committee to focus on disability-free life expectancy rather than healthy life expectancy as a measure of the well-being of an ageing population."[19]

2.25.  Sir John explained that it was generally accepted that the prevalence of disability in later life had fallen in the United States since the 1980s,[20] but that as far as the UK was concerned "the informed view is that we simply do not know what is happening, but there is certainly no evidence that disability levels in later life are falling as in the USA". Evidence that we received during our visit to the United States National Institute on Aging was to the same effect; Dr John Haaga, deputy associate director of the Behavioral and Social Research Program, suggested that in the United States, unlike the UK, HLE had been rising at the same rate as LE over previous years.[21]

2.26.  In Sir John's view, what had happened in the United States could be made to happen here through appropriate health, social and educational policy. It was therefore highly desirable that disability free life expectancy in the general population should be monitored. This would not only assess the effectiveness of relevant government policy, but would also provide a more secure basis for recognising current shortfalls and predicting future needs for health and social services. Sir John thought that the ONS would be the most appropriate body to oversee the repeated standardised surveys necessary, and that it could draw on the extensive experience in the United States in the assessment of disability.

2.27.  Although disability is easier to define, and hence to determine, than ill-health, it is still far from being an absolute. For example, research in the Netherlands has shown that in 85-year olds there can be a marked discrepancy between having an intrinsic capacity to perform an activity of daily living and the translation of this capacity into actually doing it; the factors underlying such a discrepancy may need to be taken into account in measuring "disability".[22] Here too, different countries have different definitions as to what constitutes disability. Australia takes disability to be one or more of 17 defined conditions, Japan takes disability to be confinement to bed, France includes as disabled all those in retirement homes, while in the UK disability is self-reported as a long-standing limitation on activities in any way.[23]

2.28.  Despite this, we accept and endorse the view that it is easier to assess disability accurately than it is ill-health, and that disability is less susceptible to changes in people's perceptions of themselves over time. We believe that it is important for all concerned with this question to know more accurately how the health of older people is changing over time; what the trends are; the direction in which and speed with which they are changing; and what the forecasts are for the future.

2.29.  We conclude that there is considerable uncertainty about whether healthy lifespan is increasing faster or slower than lifespan. The uncertainty comes from the variability in individual health trajectories through life, and the difficulty in applying objective measures of health and quality of life across different age groups. We believe that freedom from disability provides a more easily ascertainable objective measure of the quality of life.

2.30.  Further research should be undertaken to validate and apply appropriate measures to monitor the trends in healthy lifespan. We recommend that funds should be made available to the Office for National Statistics to enable it to carry out over a number of years the surveys needed to assess disability-free life expectancy.

2.31.  If the information provided by such a survey does ultimately show that there is a real increase in the number of years that older people experience ill-health, it will then be possible to look for the causes of that increase, and to attempt to find a remedy. If this is achieved, it will of course improve the quality of life not only of the people concerned, but also of those caring for them. But it will have a further benefit. Some 38% of NHS expenditure is spent on the 16% of the population currently over 65, and a major part of that on the last five years of their lives, whatever the age of death. Anything that can be done to increase the years during which older people are in good health and free from disabilities will therefore additionally free resources which can be used to improve treatment and care of older people.

2.32.  For the present, the fact remains that, on any measure, there are a number of years, of the order of eight in the case of men and eleven in the case of women, during which older people regard themselves as not being in general good health, or as having a limiting long-standing illness or disability. Such evidence as there is suggests that this period of perceived ill-health is not decreasing, and may well even be increasing. Perceived ill-health will on its own adversely affect the quality of life of older people, and their sense of well-being.

2.33.  The Royal College of Physicians in Edinburgh[24] warned that "Disability may be postponed but it cannot be eliminated". We accept this obvious truth. We accept also that the adverse effects of disability cannot be eliminated. We believe however that it should be possible to increase disability-free years and, if and when disability does supervene, greatly to reduce the adverse effects.


2.34.  Older people wish to feel that they are "active providers within the community"(Q 41),[25] that they are a useful part of society, and that they are not a burden on others. This is more likely to be achieved by people who live in their own homes, and who are sufficiently mobile to have access to family, friends and shops. Reasonably good health is undoubtedly a major contributing factor, but it is not the only one. There are ways of ensuring that, although health worsens, quality of life does not, or not at least to the same extent. Technological advances are at the forefront of such measures.

2.35.  The focus of this report is therefore to make recommendations which may, first, help to increase HLE, and secondly which may, if and when health does deteriorate, ensure that the quality if life is adversely affected as little as possible. Sir John Grimley Evans summed this up as "Live longer, die faster".[26]

The Need for Longitudinal Studies of Ageing

2.36.  The figures obtained from sources such as the Census and the General Household Survey are based on samples taken at a given time, and provide useful snapshots of the position at that time for that sample. Longitudinal studies, on the other hand, aim to follow the changing fortunes of individuals over the years, and thus provide continuing patterns of data, albeit based on smaller samples, from which it may be easier to discern the factors that are responsible for underlying trends. We consider this further in Chapter 7.[27]

3   House of Lords Select Committee on Economic Affairs, 4th Report (2002-03): Aspects of the Economics of an Ageing Population (HL 179-I). Back

4   Census 2001: First Results on Population for England and Wales, September 2002. Back

5   Health Statistics Quarterly 25, Spring 2005, table 5.1. Back

6   Clare Griffiths and Justine Fitzpatrick, Geographical Inequalities in Life Expectancy in the United Kingdom 1995-1997, Health Statistics Quarterly 09, Spring 2001. Back

7   Life Expectancy at Birth by Health and Local Authorities in the United Kingdom, 1998-2000, Table 2, Health Statistics Quarterly 13, Spring 2002. Back

8   Life Expectancy at Birth by Local Authorities in the United Kingdom, 1991-1993 and 2001-2003, Health Statistics Quarterly 24, Winter 2004. Back

9   These were the criteria used at the time of the study referred to. From 2001 they have been replaced by the National Statistics Socio-economic Classification. Back

10   Lin Hattersley, Trends in Life Expectancy by Social Class - an update, Health Statistics Quarterly 02, Summer 1999. Back

11   Angela Donkin, Peter Goldblatt and Kevin Lynch, Inequalities in Life Expectancy by Social Class 1972-1999, Health Statistics Quarterly 15, Autumn 2002.  Back

12   Paragraph 3.49.  Back

13   Office for National Statistics. Back

14   Office for National Statistics, Focus on Older People 2004, May 2004. Back

15   Paragraphs 3.42 to 3.45. Back

16   Older Americans 2004, Federal Interagency Forum on Aging-Related Statistics. Back

17   UN World Population Prospects 1950-2050, 2002 revision. Back

18   OECD: Perceived health status, July 2004. Back

19   p 356. Back

20   Note of the Seminar, Appendix 4, paragraph 21; Manton KG, Gu X (2001). Changes in the prevalence of chronic disability in the United States black and nonblack population above age 65 from 1982 to 1999. Proc Natl Acad Sci USA; 98: 6354-6359. Back

21   Appendix 5, paragraph 33. Back

22   A. Bootsma-van der Wiel et al, Disability in the Oldest Old: "Can Do" or "Do Do"? (2001) Journal of the American Geriatrics Society, Vol 49, Page 909. Back

23   International Health Comparisons, National Audit Office, 2003, quoting from OECD Health Data 2002. Back

24   p 391. Back

25   From the evidence of Professor Christopher Phillipson, p 23. Back

26   Note of the Seminar, Appendix 4, paragraph 21. Back

27   Paragraphs 7.36 - 7.43. Back

previous page contents next page

House of Lords home page Parliament home page House of Commons home page search page enquiries index

© Parliamentary copyright 2005