The Scottish Government

« Previous | Contents | Next »

Listen

Chapter 4: Population, demographics and morbidity

Future demand for care for older people will be determined by changes in the numbers of older people and any changes in ill-health and dependency. Future numbers of older people will be determined by demographic factors such as historical fluctuations in the birth rate and inwards and outwards migration and, crucially, future trends in life expectancy.

The impact of the increasing numbers of older people on the demand for care will depend upon future trends in age-specific levels of ill-health or dependency. In other words will an 80 year old in 20 years time be more or less healthy, more or less dependent, than an eighty year old at present. A core assumption of the projections in the first report of the Range and Capacity Review Group was that age-specific levels of ill-health or dependency would remain constant. This was a reflection of a lack of strong evidence of a sufficiently quantifiable nature to be incorporated in the model rather than of any evidence that age-specific levels of ill-health and dependency would remain constant.

How such changes will translate into actual levels of provision of care will depend upon a range of other factors including the availability of carers; the availability of services; care standards; the future of pensions, and government funding including free personal care; and how health and care markets develop. Apart from the first, these are not relevant to this chapter.

Increase in the number of older people

The latest population projections for Scotland were published by the Registrar General for Scotland in October 2005. The projections concentrate on the period up to 2031, although for the first time the Government Actuaries Department is making available projections up to 2074. Projections that far ahead become increasingly more uncertain, however, because assumptions are being made in some cases about the behaviour of people not yet born.

The key points are that:

• the population of Scotland is projected to rise, peaking at just over 5.1 million in 2019 and then slowly declining, falling below 5 million in 2036 and reaching 4.86 million by 2044;
• the number of children aged under 16 is projected to decrease by 15% from 940,000 in 2004 to 790,000 by 2031;
• the number of people of working age is projected to fall by 7% from 3.18 million in 2004 to 2.96 million in 2031;
• the number of people of pensionable age is projected to rise by 35% from 970,000 in 2004 to 1.31 million in 2031;
• the number of people aged 75 and over is projected to rise by 75% from 370,000 in 2004 to 650,000 by 2031 (due in part to the baby boomers after the Second World War entering their early eighties by 2031 and the effect of improved mortality rates.)

Scotland's population by age group over the years from 2004 to 2031, on the basis of the 2004 projections, is expected to be:

* Pensionable age is 65 for men, 60 for women until 2010; between 2010 and 2020 pensionable age for women increases to 65.
Note: not all figures sum due to rounding.

These changes are illustrated graphically in the following chart (2004 projections):

Scotland's older population by 5 year age groups, actual and projected, 1911-2044

Interestingly the dependency ratio - the ratio of persons aged under 16 or over pensionable age to those of working age - is projected to remain around 60 per 100 from 2004 to 2021. It is only after 2021, and the completion of the change to the state pension age, that the dependency ratio rises, to 71 per 100 working age population in 2031, and eventually to 75 in 2044.

Projections are available from the Registrar General for Scotland by local authority area. This will be very helpful for health boards and local authorities as they undertake the capacity planning we recommend at Chapter 8.

Increasing life expectancy

Compared with the position in 1900, when men and women who survived to 65 were a minority, life expectancy has increased greatly.

The Registrar General for Scotland's 2004 population projections contain assumptions about life expectancy. They base future improvements in mortality rates on the trend observed in the period 1961 to 2003. It is assumed that annual rates of reduction in mortality rates will tend towards a common reduction at each age of 1% a year by 2029. Thereafter the mortality improvement is assumed to continue at this rate (in contrast to previous projections where it was assumed to half every subsequent 25 years). In line with the long-term trends, it has been assumed that the mortality rates for Scotland will continue to be higher at most ages than those for England and Wales. Based on these rates, expectations of life at birth are projected to increase as follows:

This is around 0.3 years more for males over 27 years (2003-2030), and 0.1 years more for females, compared to the assumptions used in the 2002-based projections.

It should be noted these are the expectations of life at birth, so do not impact on the numbers of older people living in 2030.

The Actuarial Profession published its latest Continuous Mortality Investigation in September 2005. This shows improvements in pensioner mortality in the UK since the work was last done in 1994. The work is based on data analysis for the 4 year period 1999-2002 and shows mortality rates around 30% lower than the 1994 work (based on the 4 years to 1994) for both males and females in their late 60s. They noted that

"Life expectancy has improved dramatically over recent decades but all estimates of future mortality carry considerable uncertainty. Issues of individual choice, such as diet, smoking or drugs have the potential to slow down or even reverse mortality improvements. Individuals, by making choices, have a big impact on how long they live."

The report suggests that if recent trends are simply extrapolated over the next decade, dramatic increases in life expectancy are likely. But while life expectancy is improving for all, it is going up fastest among those born in the mid 1930s - a so called golden cohort. These people are now just into their 70s, and may be skewing the results - a "cohort effect" on the mortality data that illustrates the uncertainty and difficulties of projecting mortality.

Factors contributing to longer lives are the virtual abolition of once common childhood illnesses such as scarlet fever, diphtheria, polio, measles and mumps, a decline in smoking and a general improvement in diet and housing. It is possible these factors have contributed particularly to the increasing life expectancy of those in their mid 60s during 1999-2002. No-one can confidently predict whether or not they will continue into the future.

It will be seen from this that the projections of mortality by The Actuarial Profession and the Registrar General for Scotland are made on quite different bases.

Life expectancy: future trends

There are doubts about whether future generations of older people will be as fit and healthy as those who are currently pensioners. There is some evidence that the cohort of older people born into the privations of the 1920s and the Great Depression are healthier - and have greater longevity - than more affluent post-war cohorts. The increase in obesity and unfitness among younger people today is likely to affect their health status and longevity - and their prosperity - as they grow older. We had some discussion around this, thinking of the increased consumption of fast food and soft drinks, and the rise of a generation involved in computer games and other sedentary pursuits.

We were interested, therefore, when the New England Journal of Medicine reported on 17 March 2005 that while it had been assumed that life expectancy in the U.S.A would rise indefinitely, new data analysis suggests that this trend is about to reverse itself due to the rapid rise in obesity, especially among children.

The review concluded that, on a conservative estimate, obesity now reduces average life expectancy by about 4 to 9 months in the U.S.A. The researchers further conclude that if the current epidemic of child and adolescent obesity continues unabated, life expectancy in the U.S.A. could be shortened by 2 to 5 years in the coming decades.

UK specialists noted that the same trends could happen here. We note the point, and that concerns about future health are already leading to action to remove soft drinks from schools and to increase the nutritional value of school meals.

Trends in age-specific levels of ill-health and dependency among older people

The previous section has described trends in life expectancy in Scotland. Future changes in life expectancy will be the most important determinants of how quickly the numbers of older people in Scotland increase in years to come. The first report of the Range and Capacity Review was a systematic attempt to model the impact of projected changes in the number of older people on the demand for care. A core assumption in these models was that age-specific levels of ill-health and dependency would remain constant. In other words an 80 year old in 20 years time would, on average, experience the same level of ill-health or dependency - and thus generate the same demand for care - as an 80 year old at present.

How realistic is this assumption? What is the balance of evidence relating to whether older people of a given age will be more or less healthy in future? The first point to make is that there is a remarkable lack of definitive evidence about past trends in levels of age-specific ill-health and dependency among older people. About future trends there is even less certainty. Experts are extremely unwilling to make predictions about future trends.

Some useful context may be provided by assessments carried out as part of the analysis of drivers for change in the NHS in Scotland for the National Framework for Service Change. This was part of an attempt to assess the extent to which
long-term increases in the levels of emergency inpatient admission among older people were a reflection of increased levels of age-specific ill-health among older people.

The balance of evidence, at an international, British and Scottish level, was that age for age older people have been getting healthier.

A recent analysis based on the relationship between hospital expenditure and proximity to death provided the most powerful evidence that future demand for health care is unlikely to increase as quickly as would be the case if age-specific levels of ill-health were to remain constant in future (Seshamani, 2004)

A high proportion of health care expenditure and in particular hospital expenditure occurs in the years immediately preceding death. For example, the 5% of patients in the last year of life generated approximately half the hospital expenditures for the population aged 65 and over in 2002 (Seshamani, 2004). Proximity to death is a much more powerful predictor of hospital expenditure than age alone.

This relationship between hospital expenditure and proximity to death provides the basis for an elegant calculation of the extent to which the impact of an ageing population on health-care expenditure is likely to fall far short of being proportional. Seshamani (2004) compared two methods of projecting future hospital expenditure. The first assumed that each age group would generate a constant per capita amount of hospital-based health care expenditure over the years 2002 to 2026. The second method incorporated per capita levels of expenditure specific to proximity to death. Incorporating the effects of proximity to death in the expenditure model halved the estimated increases in real national hospital expenditure over time.

This is another way of making the point that improved age-specific levels of health in the older population are likely to mitigate to a considerable extent the impact of growing numbers of older people on the demand for health care.

Of course, these analyses were carried out with reference to ill-health and health care costs. Caution should be exercised in extending their implications to the area of social and residential care.

However, in Scotland, Stearns and Butterworth (2001) carried out analyses directly relevant to the broader definition of care for older people as background to the work of the Care Development Group. They analysed data from the 1985 OPCS Survey of Disability among Adults in Private Households and the 1996/7 Disability follow-up to the Family Resources Survey. They concluded that

"Overall, disabled elderly people (in all living conditions combined) decreased as a proportion of the population between 1985 and 1996/7. The best estimate of the reduction overall is of 0.2 or 0.3 percentage points per year in the UK, though evidence indicates that the rate of reduction may have been slightly greater in Scotland".

Healthy life expectancy and impaired health at the end of life

In recent years, there has been increasing interest in a complementary perspective on levels of ill-health and disability among older people. This concerns the relationship between increasing life expectancy and changes in healthy life expectancy. In other words how many of the additional years of life contributed by increasing life expectancy will be lived in good health and how many will be years of ill-heath or disability.

There are range of views about current trends and the likely pattern in the future. At the optimistic end of the spectrum is the view is that the years of ill-health (and high requirements for care) at the end of life will become ever more 'compressed' into the final years (the 'compression of morbidity' hypothesis). A more pessimistic view is that medical advances will lead to greater overall life expectancy but not be able to delay the onset of disability or ill-health - and therefore most or all of the additional years of life will be years of ill-health and disability rather than good health.

Where the truth lies along this spectrum, we do not know for certain. The best current estimate is that healthy life expectancy is increasing - in another words some of the years of additional life will be healthy - but unhealthy life expectancy is also increasing - some of the additional years of life will be years of ill-health or disability. This view is reflected in a recent analysis of trends in healthy life expectancy carried out in Scotland.

However before looking at these results in more detail we must be very clear about the relationship between analyses couched in terms of trends in healthy life expectancy and the kinds of analyses couched in terms of trends in age-specific levels of ill-health and disability which have already been discussed.

In particular, the projections outlined in the First Report of the Range and Capacity Review were based on an assumed trend of no change in levels of age specific ill-health and disability.

In terms of the spectrum of views relating to trends in healthy life expectancy this assumption of no change in age-specific levels of ill-health is at the pessimistic end of the spectrum. An assumption of no change in age-specific levels of ill-health is broadly equivalent to an assumption that there is no change in healthy life expectancy and that all the extra years of life contributed by increasing life expectancy will be years of ill-health. Any increase in healthy life expectancy means that a person of a given age is, on average, more likely to be healthy than they were before the increase in healthy life expectancy - or in other words that there has been an improvement in age-specific levels of health.

Demonstration of any improvement of healthy life expectancy has, in broad terms, the corollary that age-specific levels of ill-health are declining and is evidence for a more positive out-turn than the assumptions incorporated in the model presented in the First Report.

The first published estimates of levels and trends in healthy life expectancy for the Scottish population appeared in Healthy Life Expectancy in Scotland by ISD Scotland in March 2004. This found that:

• the current life expectancy of a 65 year old man is Scotland is 14.8 years: 7.7 of these years (on average) will be spent free from limiting long term illness and 11.5 years will be in good or fairly good health.
• the current life expectancy of a 65 year old woman is Scotland is 17.9 years: 8.9 of these years (on average) will be spent free from limiting long term illness and 13.4 years will be in good or fairly good health.
• in general terms over the last 20 years healthy life expectancy at birth has not kept pace with increasing life expectancy - the proportion of years of life spent in less than good health has increased.
• the figures for those who have reached the age of 65 are better, however - their healthy life expectancy is generally keeping pace with increasing life expectancy, so the proportion of time spent in less than good health has remained the same.

Healthy Life Expectancy in Scotland did not make any projections into the future.

Perhaps of greatest relevance to issues of social, personal and residential care are the findings relating to trends in life expectancy with unassisted Activities of Daily Living. For women at age 65 the expectation of the number of years they would be able to live without assistance in Activities of Daily Living increased from 14.6 years in 1980 to 16.0 years in 1998. For males the increase was from 11.6 years to 12.6 years. At the same time there was an increase for both males and females in the number of years mostly at the end of life where they would require assistance with Activities of Daily Living. The implication is also that an older person of a given age in Scotland was less likely to require assistance with Activities of Daily Living in 1998 than they were in 1980.

This exemplifies the general tenor of the ISD Report which lies towards the middle of the spectrum of views outlined earlier. The extra years of life contributed by increasing life expectancy have contributed increasing years of ill-health and increasing years of health.

In terms of implications for trends in age-specific morbidity, it is the finding of increasing life expectancy which is important. This implies that age-specific levels of health have been improving. If this trend continues into the future, the out-turn will be more positive than that implied by the assumption of no change in age-specific levels of ill-health incorporated in the First Report.

Other points to note

Finally it needs to be borne in mind that there is an increasing number of older people with long-term conditions such as diabetes, dementia, and learning disabilities. The number of people with dementia in Scotland was around 60,000 in 2001, and it is projected to increase to around 82,000 in 2020 and to around 125,000 by 2041: see Annex B. This is a function of both increasing life expectancy and the demographic increase in the number of older people expected over the next 20 years. And there have been quite dramatic increases in the life expectancy of people with learning disabilities.

Conclusion

As its baseline model, the First Report of this group related future demand for care among older people to projected changes in the sizes of the age groups concerned - with an assumption that age-specific levels of ill-health and disability among older people would not change.

The actual out-turn will depend on future trends in age-specific levels of ill-health and disability among older people.

The balance of evidence relating to past trends is that age-specific levels of ill-health and disability among older people have been improving.

This is confirmed by evidence of increasing healthy life expectancy.

To the extent that these trends continue, the projections contained in the first report of this Group can be regarded as conservative ( i.e. pessimistic).

The major determinant of increasing future demand for care for older people in Scotland will be demographic change involving increasing numbers of older people and in particular the oldest old. This is the factor which was modelled in the First Report of the group.

The precise out-turn may well be affected by changes in age-specific levels of
ill-health and disability in the population and increases in healthy life expectancy. If past trends in age specific levels of ill-health and disability and past increases in healthy life expectancy continue, they will serve to mitigate to some extent the impact of an ageing population on the demand for care.

« Previous | Contents | Next »