|
South West Public Health Observatory
|
|
|
SWPHO Bulletin September 2003
General health Limiting long-term illness (LLTI) Carers Further information |
Click on the thumbnail images to see full-size graphs and maps.
Introduction
In February 2003 the Office for National Statistics (ONS) released the first results from the 2001 Census. These comprised key statistics for all local authority areas. This document is the response of the South West Public Health Observatory to the health content of these results. It is not intended as an in-depth analysis, but sets out to give an overview of these new data and some of the potential impacts on the NHS.
As more data become available further analysis and reports will be produced, with more detail (e.g. age and sex breakdowns) and a lower geographic unit (e.g. electoral ward and output area) due later in 2003. Output areas are small geographical areas created for the purpose of the 2001 Census. Smaller than wards, they will be designed to be around 125 households in size, have regular shapes, and where possible 'natural' boundaries, with populations which tend towards homogeneity. They will 'nest' within wards and parishes, and normally comprise whole unit postcodes.
The 1991 Census saw for the first time a question specifically about health. It asked whether the respondent had any long-term disability or illness that affected their ability to carry out their work or everyday tasks. This was a valuable source of information about morbidity in the population, although it was limited in scope.
The 2001 Census offers those working in public health more information than in previous years. Firstly it repeated the limiting long-term illness question, which means that comparisons can now be made to see how things have changed over the last decade. Secondly an additional health question was included. This asked respondents to assess their own personal health as "good", "fairly good" or "not good", giving a more general view of everyone's self-perceived health as opposed to a simple measure of those with chronic health problems. Finally it also asked about the provision of unpaid care for long-term physical or mental ill-health, disability or problems related to old age.
Questions relating to communal establishments, demographics and socio economic characteristics are also included in the Census. These, together with the specific health questions, will allow public health professionals to examine health and the determinants of health using the 2001 Census.
The general health question on the 2001 Census asked people to assess their own health as ‘good’, ‘fairly good’ or ‘not good’. Figure 1 shows the percentage of residents who rated their health as good or fairly good, for each region in England. The South West region (91.5%) is roughly middle of the range, with a higher percentage of people responding positively to the question than England as a whole, but a lower percentage than the South East, East and London.
Figure 1: Percentage of people with 'good' or 'fairly good' self reported health by region
Although the South West’s population seems to think they are relatively healthy, there is substantial variation within the region. Penwith for example shows the lowest percentage of positive responses in the region (88.24%) compared to the Isles of Scilly where 94.9% of people responded positively. This variation by Local Authority can be seen in Figure 2.
Figure 2: Percentage of 'good' or 'fairly good' self-reported health (across South West Local Authority areas)
This variation may, in part, be due to the demography of the area. Figure 3 shows the relationship between the percentage of people responding "not good" to the general health question and the percentage of people aged 65 or over for all Local Authorities in the South West. Although there is a weak relationship between the two variables this isn't strong enough to account for the size of the variation in the health question.
Figure 3: Scatterplot showing the relationship between the percentage of people aged 65 or over and the percentage of people with 'not good' self-reported health (across South West Local Authority areas)
Another possible reason for this variation is that people with the same actual health experience may assess their health differently. It may be, for example, that people assess themselves in comparison to the population as a whole or to their peers or to some perceived expected health for someone of their age and status. In practice, the answers will be a combination of perception and actual health status and both factors are important in relation to health services. This is a concept that will need to be explored further. In particular the relationship between actual health measures (mortality and ill-health measures) and perceived health (general health question) will be analysed as and when ward and output area level data become available.
The variation could also be influenced by deprivation. Figure 4 illustrates the relationship between the general health question and the percentage of people who are classed as "employment deprived" by the DETR Indices of Deprivation (2000). There is a strong relationship between employment deprivation and negative responses to the general health question. These issues will need to be examined more closely at a future date, as more data becomes available.
F
igure 4: Scatterplot showing the relationship between the percentage of people with 'not good' self-reported health and the percentage of people who are employment deprived (as defined by DETR 2000 Indices of Deprivation) (across South West Local Authority areas)
Figure 5: Percentage of residents with 'not good' self-reported health (across South West Local Authority areas)
Limiting long-term illness (LLTI)
In both 1991 and 2001 Census data was collected on any "long term illness, health problem or disability which limits your daily activities or the work you can do" (2001 Census). The questions were not identical but are thought to be similar enough to allow comparisons between 1991 and 2001 (Figure 6).
Bennett et al. state that, "caution needs to be exercised when interpreting changes in the prevalence of self reported morbidity as changes over time may reflect changes in people's expectations of health as well as the prevalence or duration of sickness". (Bennett et al. 1996,
Living in Britain: results from the 1994 General Household Survey
(London: HMSO))
With this in mind, it is still useful to compare and this comparison can be seen in Figure 6. It shows that the percentage of people suffering from LLTI has increased considerably over the last ten years in every region of the country.
The rate of change within the South West can be seen in Figure 7, where the change ratio for each local authority in the region has been calculated. A value of 1 would mean that the values for 1991 and 2001 are equal, whereas a positive value indicates an increase).
Figure 6: Percentage of residents with limiting long term illness in 1991 and 2001 by region
Figure 7: Ratio of change in limiting long term illness 2001:1991(across South West Local Authority areas)
The relative change is variable, with Isles of Scilly scoring 1.13, compared to 1.51 in South Gloucestershire. The relative change in LLTI seems to correlate well with the relative change in the percentage of people aged 65 or over (Figure 8), indicating that the variability in the increase in LLTI is at least in part due to the extent of change in the percentage of the population made up by the elderly. The main anomalous area is the Isles of Scilly, with a lower increase in LLTI than expected given the increase in their percentage of people aged 65 or over. This could be due to a healthy elderly population effect, and/or the small sample size in the Isles of Scilly.
Figure 8: Scatterplot showing the relationship between change in limiting long term illness and residents aged 65 and over (across South West Local Authority areas)
Although there does seem to be a relationship, the geographic patterns of LLTI and residents aged 65 or over differ (Figures 9 and 10). The general patterns are similar, but the areas with highest percentage of residents aged 65 and over (predominantly south coast) are not necessarily the same areas that exhibit the highest levels of LLTI (predominantly in the far west). It would seem therefore that although the age structure of an area has a significant influence on LLTI, other influences such as deprivation, and the influence of pockets of healthy elderly people are also having an effect. These issues will be examined in more detail when more data becomes available.
Figure 9: Percentage of residents with limiting long term illness across South West Local Authority areas)
Figure 10: Percentage of residents aged 65 or over (across South West Local Authority areas)
Carers are people looking after or giving help or support to family members, friends, neighbours or others, because of long term physical or mental ill-health or disability, or problems related to old age. The 2001 Census was the first to ask people about this.
In the South West 10.5% of the population provide unpaid care, which is comparable with the levels in England as a whole (see Figure 11). This may seem surprising given that the South West is the region with the largest percentage of older people, but when we look at individual Local Authorities (Figure 12), variation within the South West is apparent.
Figure 11: Percentage of people providing unpaid care by region
Figure 12: Percentage of residents providing unpaid care (across South West Local Authority areas)
The relationship between limiting long term illness and levels of unpaid care is clear cut and not surprising. Figure 13 shows that there is a strong relationship between the levels of unpaid care and the levels in limiting long term illness in a given area.
Figure 13: Scatterplot showing the relationship between limiting long term illness and unpaid care (across South West Local Authority areas)
Figure 14 shows the relationship between unpaid care and the percentage of the population aged 65 or over. There does seem to be a relationship, but this is less strong than might have been expected given that the requirement for care should increase with age.
Figure 14: Scatterplot showing the relationship between provision of unpaid care and percentage of population aged 65 or over (across South West Local Authority areas)
Figure 15 shows the relationship between provision of unpaid care and deprivation. The DETR Concentration Score evaluates the level of deprivation for the most deprived 10% of each Local Authority. It was chosen over other Local Authority based deprivation scores such as Income or employment as these scores are calculated using benefit claims, which could be a major confounding factor when examining unpaid carers. The scatterplot indicates that the relationship between deprivation and levels of unpaid care is weak.
Figure 15: Scatterplot showing the relationship between provision of unpaid care and the DETR concentration score (across South West Local Authority areas)
Another contributing factor could be a relationship between the percentage of people living in communal establishments (as defined by Census 2001) and the levels of unpaid care. Figure 16 shows that there may be a weak inverse relationship.
Figure 16: Scatterplot showing the relationship between residence in communal establishments and unpaid care (across South West Local Authority areas)
These relationships suggest either that where there is a lack of communal establishments the need has to be met by the community, or where the community structure is not in place to provide care, communal establishments are needed to meet the need. However, the relationship is not strong, suggesting that there are other factors that may help to explain the distribution of unpaid carers.
The maps of the levels of unpaid care (Figure 17) and levels of nursing and residential home residents (Figure 18) show similar patterns with general east west gradients and high levels of both variables on the south coast. The main differences between the two seem to be in Cornwall where there are high levels of unpaid care and only moderate levels of nursing or residential home uptake. The opposite seems to be the case in Devon.
Figure 17: Percentage of residents providing unpaid care (across South West Local Authority areas)
Figure 18: Percentage of population resident in a nursing or residential home (across South West Local Authority areas)
Indications are that levels of unpaid care are not solely determined by the availability of communal establishment facilities, by the levels of LLTI, by the age structure or by deprivation levels of the area. Until more detailed data are available, broken down by age group and at a smaller geographical level we will not be able to explore this further, but it is obvious that the relationships are complex.
The Office for National Statistics and South West Observatory websites both hold Census 2001 data as well as further information about the Census and planned dissemination activity.
