How have social care needs and long-term conditions changed over time?

Key points

  • This chapter considers how the relationship between age and the rate of social care need and long-term conditions shown in chapter 2 has changed over time.
  • Between 2006 and 2018, there was a fall in the share of people with social care needs in each age group. The change was greatest for those aged 80 and older, meaning a higher share of people are now able to live independent lives.
  • By contrast, the share of people 75 and older with two or more long-term conditions (according to primary care records) increased and the share of those with no recorded long-term conditions fell. For those younger than 75, rates of long-term conditions remained largely unchanged.
  • The reduction in the rate of social care need by age has counteracted some of the increase in the total number of people with need which is the result of an ageing population. We estimate that the number of people with two or more social care needs is 0.2 million lower, and the number of people with no social care needs 0.6 million higher in 2018 than they would have been if age-related prevalence had remained at 2006 rates.
  • The change in the rate of long-term conditions by age exacerbated the increases in the total number of people with long-term conditions because of a rise in the number of older people. The estimated number of people with two or more diagnosed long-term conditions was 0.4 million higher, and the number of people with no diagnosed conditions 0.4 million lower, than if prevalence by age had remained at 2006 rates.
  • The total number of people living in residential settings, those with the highest social care needs, grew 9% between 2007 and 2019, while the population aged 85 and older grew by a quarter. This suggests that a smaller fraction of this age group now uses residential care and is consistent with a lower fraction – but a higher total number of people in their 80s and 90s – having social care needs. The number of people in residential care will also be affected by factors other than need, including local authority policies and funding and the availability of unpaid care provided by family.

This chapter considers how social care need, as measured by ADL limitations, and long-term conditions have changed over the past 15 years. We begin with an individual perspective, looking at how the chances of having a social care or health care need have changed over this period. We then consider what this means at the population level, estimating the total number of people who need care, and how this has changed.

We focus on two ways of measuring how ageing might affect an individual’s life. The first looks at social care need, measured by no ADL limitations (‘no need’) and two or more ADL limitations (‘high social care need’) from ELSA. The second is long-term conditions, as diagnosed and reported in primary care CPRD data, measured by zero long-term conditions and two or more long-term conditions. We use CPRD administrative data rather than self-reported long-term conditions from ELSA to allow for larger sample sizes, and because this measure is likely to have more direct implications for NHS service use. We do not focus on one ADL limitation or one long-term condition, as trends are typically less clear or informative. All trends have been age-sex standardised using the 2018 population as the reference. This means for every year for our measures, we take the proportion by 5-year age bands and sex, and apply it to the corresponding population for that age band and sex in 2018. This means that we adjust for changes in population structure over time.

The rate of social care needs within age groups has fallen over time

Figures 4 and 5 show the share of ELSA respondents who have no ADL limitations and two or more ADL limitations in each of the biennial surveys between 2006 and 2018. Figure 4 shows that the share of the population with no ADL limitations is increasing in all age groups and that the share of people with any ADLs is declining. The trends are not very smooth due to the relatively small sample size. Changes are particularly large for those aged 80 and older; the share of those aged 80–84 with no ADL limitations rose from 68% in 2006 to 75% in 2018, while the same share for those aged 85 rose from 51% to 57%.

Figure 5 shows that the share of those with two or more ADL limitations is trending downwards for those younger than 85. There is no real trend for those aged 85 and older.

Both these figures only represent people who are living in the community – they exclude those living in residential care homes. They will therefore overestimate the true population prevalence of ‘no need’ (as people in care homes will have social care needs), and underestimate the true prevalence of need. The final section of this chapter documents trends in the residential care home population to provide a more complete picture of how need across the older population as a whole has changed.

Figure 4: The percentage of the older population with no ADL limitations (no social care need) by age group, 2006–2018

Source: ELSA, 2006–2018.

Figure 5: The percentage of the older population needing help with two or more ADL limitations (high social care need) by age group, 2006–2018

Source: ELSA, 2006–2018.

Rates of long-term conditions are increasing for those 75 and older

Figures 6 and 7 show how the shares of people with no long-term conditions and two or more long-term conditions have changed between 2006 and 2015, by age group.

For the 65–69 and 70–74 age groups, there was an increase in the rate of diagnosed conditions between 2006 and 2008. The share of people with no long-term conditions fell, and there was an increase in the share of people with two or more long-term conditions. After 2008 (our ELSA data runs 2006–2018), these trends halted, and there was very little change in the rate of diagnosed conditions among these age groups. For those aged 75–79, rates of long-term conditions increased up until 2010, but remained stable thereafter.

For the 80–84 and 85+ age groups, the rates of diagnosed long-term conditions increased at a faster rate than for younger cohorts, and continued for the whole period between 2006 and 2015. For those aged 80–84, the proportion of people with no long-term conditions fell from 40% to 32%. For those aged 85+, this share fell from 39% to 31%.

The increases in long-term conditions for those older than 80 could be due to real changes in the underlying prevalence of disease, or changes in the frequency of diagnosis (for example due to changes in diagnostic guidance or an improvement in diagnostic tools). What is clear is that the increase in diagnosed long-term conditions for those aged 80 and older occurs at the same time as a reduction in social care need.

Figure 6: Percentage of the older population with no diagnosed long-term conditions by age group, 2006–2015

Source: CPRD, 2006–2015.

Figure 7: Percentage of the older population with two or more diagnosed long-term conditions by age group, 2006–2015

Source: CPRD, 2006–2015.

What do the changing rates by age mean for the population?

The number of older people with a social care need is defined by two things: the prevalence of need in the population (ie how common it is), and the size of the population. As populations age, the number of older people will increase and – unless there are reductions in the prevalence of need – so too will the number of people with social care needs. In this section we disaggregate these two factors.

To do this, we start by estimating what the total number of people with needs would have been if prevalence of need by age had not changed. We estimate this by fixing our rates of need by age at their 2006 values and applying actual population numbers in each age group over time. We then compare these projected totals with estimated needs using the actual rates of need by age, which change over time, and the actual population numbers.

Figure 8 shows the results of these estimates and projections for those with no ADL limitations and no long-term conditions. In both cases, the estimated number of people with needs increases over time. However, the fall in ADL limitations over time, shown in Figure 8, means that the projected population living without ADLs, based on the 2006 rate and shown by the dotted line, is below the actual estimate shown by the solid line. In other words, there are more people living without social care need than we would have expected based on the change in the size and age structure of the population alone. For long-term conditions, the opposite is true: there are fewer people living without long-term conditions than we would have expected purely from demographic changes. This is because the rates of diagnosed long-term conditions within each age group have risen.

Figure 9 presents the same analysis for the population with two or more ADL limitations and two or more long-term conditions. The number of people with two or more ADL limitations has increased slightly over time, but is lower than we might have expected based on population growth alone. Conversely, the number of people with two or more long-term conditions is higher than we might have expected.

Table 1 summarises the change in the estimated number of people with no needs and high needs between 2006 and 2018 for ADL limitations, and 2006 and 2015 for long-term conditions. It also shows the relative contributions of demographic change (the size and age structure of the population) and the change in the rate of need by age.

For low need, the demographic changes increase both the number with no long-term conditions and those with no ADL limitations. In the case of ADL limitations, the change in the rate of ADL limitations by age acts to further increase the population with no needs. By contrast, demographic changes alone (with the 2006 prevalence rates) would have led to a 0.9 million increase in the number of people living with no conditions between 2006 and 2015. Instead, due to an increased rate of diagnosed long-term conditions, the actual increase in those living with no conditions was 0.5 million.

Turning to those with two or more ADL limitations or two or more long-term conditions, we would have expected to see a growth of 0.3 million in the number of people with two or more ADL limitations between 2006 and 2018. Instead, the reduced rate of social care need has led to an actual growth of 0.1 million (0.2 million lower than expected). Conversely, we would have expected the number of people with multiple long-term conditions to have increased by 0.3 million between 2006 and 2018 had rates of diagnosed conditions by age remained at 2006 levels. Instead, the actual growth was 0.8 million (0.5 million higher than expected) as a result of the increased prevalence of individuals with two or more long-term conditions.

Figure 8: Estimated population living without ADL limitations and long-term conditions, 2006–2018

Source: ELSA and CPRD, 2006–2018.

Figure 9: Estimated population living with multiple ADL limitations and long-term conditions, 2006–2018

Source: ELSA and CPRD, 2006–2018.

Table 1: Changes in the total number of people with ADL limitations and diagnosed long-term conditions between 2006 and 2018: the contributions of demographic change and changes in rates of prevalence by age (millions)

0 ADL limitations

0 long-term conditions

2+ ADL limitations

2+ long-term conditions

2006 estimate

5.8

4.0

1.1

1.8

Change in population size and age structure

+ 1.5

+0.9

+0.3

+ 0.3

Change in prevalence by age

+ 0.6

-0.4

-0.2

+0.4

Final year estimate (2018 ADL/2015 long-term condition)

7.9

4.4

1.2

2.6

Note: Figures may not add up as a result of rounding. ADL limitations estimated using ELSA. Long-term conditions based on diagnosed long-term conditions in CPRD. Estimates given by applying prevalence to ONS population estimates.

The residential care home population

ELSA is representative of older people living in the community, but not those living in residential homes. The decline in ADL limitations in ELSA for the oldest in our population may therefore not reflect what is happening for the overall population – if the threshold for who is admitted to a residential home has changed. We therefore assess how the ELSA results match up with the residential home population.

Figure 10 shows that the total number of people in residential settings in the UK (both state and self-funded) has risen by around 9% between 2007 and 2019. At the same time, the number of people aged 65 and older, and number of people aged 85 and older, has risen by around a quarter. This indicates that a smaller proportion of older people are living in residential settings. We also know that recent data on social care need in residential settings shows increased levels of severe disability, as well as rising average age., The median age in care homes rose from around 84 to 88 between 1992 and 2012.

This pattern of care home use appears consistent with our ELSA results. The share of people with social care needs in each age group is falling over time, but the total number of people with needs is increasing due to the rise in the number of people living to advanced ages. There are likely to be other contributing factors to the care home occupancy rate, including preferences of care recipients and supply side constraints (such as limited capacity). However, if this were the dominant factor, we would expect to see an increase in the prevalence of ADL limitations in ELSA, which is not the case. The patterns we observe for care homes and ADL limitations are also consistent with trends in dementia, where rates for a given age group have declined but the total number with dementia has risen.

Figure 10: Number of older people in residential settings in the UK by source of funding, 2007–2019

Source: LaingBuisson. Care homes for older people, 30th edition.

Overall, this chapter’s findings suggest that the number of older people needing social care has not risen as quickly as the population of older people. However, this relates to what people report as their social care needs, which is not the same as the care they receive, nor where they get this care from. Demand for different types of care – state-funded, privately funded, or unpaid care – will have grown at different rates, as each is affected by different factors. For example, because social care is means tested, users of state-funded social care tend to be poorer, and may not have experienced the same reductions in social care need as the population as a whole. The demand for state-funded care will also be influenced by the availability of unpaid care provided by family and friends.

Previous Next