COMPARISON OF MEASURES FROM THE SDAC, THE SHORT DISABILITY MODULE AND THE CENSUS
In order to explain some of the differences between results from the different collections, the measures that are derivable from the 'Severity of disability' variable are presented. The section examines the relationship between the 'Severity of disability' measures from the different collections and age.
The analyses include the prevalence of various levels of impairments across the collections by age and the prevalence of the different Disability types across the collections. The aim is to identify where differences in prevalence occur to understand how the three sources of disability information differ in terms of who they identify as having a disability and the implications this has on the use of the resulting data.
In order to present the most comparable statistics from each collection, non-private dwellings and children under 18 years were excluded from all of the analysis presented in this section.
Severity of disability
The tablebelow presents the statistics for severity of disability from the Survey of Disability, Ageing and Carers (SDAC 15), the Survey of Income and Housing (SIH 13-14), the General Social Survey (GSS 14), National Health Survey (NHS 14-15), the Personal Safety Survey (PSS 12), and the Census 2016. The Census ‘core activity need for assistance’ variable is only discussed in one section of the analysis, the ‘Profound or Severe core activity limitation’ section as it only provides data about this composite group.
For the purposes of this analysis, the term ‘Short Disability Module’ is used as a cover phrase for all surveys other than the SDAC, as a disability module was used in each of these, even though there were sometimes differences between the module used in the different surveys. Where these differences have had an effect on the data, the differences are noted.
The data in the table below are for people aged 18+ years living in private dwellings (the 18+ year age range is common to all of the collections). As noted earlier, the SDAC outputs differentiate between people with a disability and those with a long-term health condition that is not disabling. In contrast the Short Disability Module’s standard output is to combine these two groups. This means different data is reported in the top line of this table for different surveys. The 2014 GSS output varied from the standard for the Short Disability Module and the figure in the first data row for that survey is also only for those with a disability.
Disability Severity by Survey, people aged 18+ years
Disability or long-term health condition
20.6(a) ± 0.5 %
27.2 ± 0.8 %
31.9(a) ± 1.2 %
36.7 ± 1.2 %
31.7 ± 1.0 %
Specific limitation or restriction
17.8 ± 0.5 %
18.5 ± 0.6 %
20.5 ± 1.0 %
22.4 ± 1.0 %
20.4 ± 0.8 %
Core activity limitation
16.2 ± 0.4 %
16.2 ± 0.6 %
17.9 ± 0.9 %
18.8 ± 0.9 %
17.7 ± 0.8 %
5.4 ± 0.2 %
5.1 ± 0.4 %
4.8 ± 0.5 %
4.6 ± 0.4 %
4.0 ± 0.4 %
2.5 ± 0.1 %
2.4 ± 0.3 %
2.1 ± 0.4 %
1.8 ± 0.3 %
1.4 ± 0.3 %
3.0 ± 0.2 %
2.7 ± 0.3 %
2.7 ± 0.4 %
2.8 ± 0.3 %
2.6 ± 0.3 %
3.2 ± 0.2 %
4.5 ± 0.3 %
5.4 ± 0.6 %
6.1 ± 0.5 %
5.8 ± 0.5 %
7.5 ± 0.3 %
6.6 ± 0.4 %
7.8 ± 0.7 %
8.0 ± 0.6 %
7.9 ± 0.6 %
Education/employment restriction only
1.6 ± 0.1 %
2.3 ± 0.2 %
2.6 ± 0.4 %
3.6 ± 0.4 %
2.6 ± 0.4 %
No specific limitation or restriction
2.8 ± 0.1 %
8.7 ± 0.5 %
11.5 ± 0.9 %
14.4 ± 0.8 %
11.4 ± 0.6 %
(a) Population with disability excludes the population with no disability but have a long-term health condition.
(b) Proportion based on total which includes responses not stated people living in occupied private dwellings only. Excludes remote areas.
Examination of the table indicates that:
· Surveys using the Short Disability Module show consistently higher rates of overall disability/long term health conditions than the SDAC.
· the SDAC rate for the 'Profound/severe' category is higher than for surveys using the Short Disability Module, however it is lower in the ‘moderate’ category.
· the SDAC has a markedly lower rate for ‘No specific limitation or restriction’ than the surveys using the Short Disability Module.
To explain these differences, it is useful to separately examine each of the disability severity measures:
Disability or long term health condition
The main disability measure published from social surveys is that of 'total disability'. In the SDAC, this measure is known as 'Disability' or 'All reported disability'. In the Short Disability Module the broadest measure is known as 'Disability or long-term health condition'. The broadest measure in the Short Disability Module however (as will be shown later), incorporates a large proportion of people who have non-restricting health conditions as well as those with disability. The Short Disability Module output, ‘Disability or long term health condition’, was chosen deliberately to reflect this mixed population.
The graphbelow shows that the SDAC measure of disability is noticeably lower for all ages compared to the corresponding measures from surveys using the Short Disability Module.
However, when long-term health conditions and disability are combined in the SDAC, the measure is consistently higher than the corresponding measures from the other surveys. While some of the estimates from the surveys using the Short Disability Module approach the proportion of people in the SDAC with a ‘disability or long term health condition’, the Short Disability Module appears to include some, but not all, of those with a non-restricting long-term health condition.
· In the SDAC, disability is defined by whether a person has a long-term health condition that restricts their ability to do everyday activities. The SDAC includes a number of screening questions aimed at differentiating between those who have long-term health conditions only and those whose long-term health conditions limit their activities (i.e. those with disabilities). Therefore in the SDAC, the broadest measure ‘All reported disability’ only includes those who have a disability and excludes those whose conditions are non-restricting.
· in the Short Disability Module, the filters used to differentiate between these populations have operated differently than in the SDAC.
Users need to be aware, that not all people with long-term health conditions are identified by the Short Disability Module measure. More accurate information about people with long-term health conditions can be obtained from the 2014-15 NHS. The NHS asks expressly whether a person has been diagnosed with a number of health conditions.
While there were some small variations in the way the Short Disability Module was asked in different surveys, these changes will not account for, for example, the statistically significant difference in moderate core activity prevalence in the 2014-15 NHS when compared with the moderate core activity prevalence from the 2013-14 SIH. The variability in prevalence rates across surveys using the Short Disability Module demonstrates why users should not use disability data estimates produced by the different surveys as a proxy for monitoring disability prevalence over time.
Specific limitation or restriction
In the SDAC, the sequence of survey questions first determines whether there are people in a household who have conditions which have lasted or are likely to last for at least 6 months. These people are then personally interviewed to determine whether their conditions cause them restrictions in their everyday lives. If they are found to have restricting conditions, they are identified as having disability. They are then asked to identify the level of restriction caused by this disability. If they are found to have a core activity restriction or limitation (i.e. a limitation in their ability to look after themselves, move around freely and communicate with others) or if they are found to have a restriction that affects their schooling or employment capability, they are said to have a "specific restriction". All other people with disabilities who do not have a specific restriction are said to have "no specific restriction or limitation".
A similar process is followed with the Short Disability Module, but there may be someone else answering the questions on behalf of the respondent.
The graph belowshows that, in the category of ‘specific restriction’, the measure is in reasonable agreement across all the different surveys.
'Core-activity limitation' is classified by the severity of limitation a person may experience in one of the core areas of self-care, mobility or communication. In the SDAC:
· ‘Profound’ means a person always need help;
· ‘Severe’ means they sometimes needs help;
· 'Moderate' means a person may have difficulties in at least one of the core activities, but needs no help in performing the activity; and
· 'Mild' means the person needs no help and has no difficulty with core activities, but they use aids and/or cannot easily do certain tasks such as walking 200m, negotiating stairs, bending to pick up objects or using public transport.
The Short Disability Module uses similar definitions to SDAC for the core-activity limitations with the exception of the 'Mild' category. In the SDAC, questions on a range of activities are asked of respondents to determine whether they have a mild limitation. In the Short Disability Module, the only criterion used to define 'Mild core-activity limitation' is that a person uses aids to assist with tasks.
The graphbelow shows that survey estimates for core activity limitation from the Short Disability Module relate well to the corresponding SDAC estimates.
This is the most severe restriction category and indicates when a person cannot do, or always needs assistance with, one or more of the core activity tasks.
Severe core activity limitation
A person has a severe core activity limitation when they sometimes need help with a core-activity task, if they have difficulty being understood by family or friends, or when they communicate more easily using sign- language or another non-verbal form of communication.
The combination of profound and severe core activity limitations is a measure commonly used because it represents the population with the greatest disability and therefore the greatest need for assistance with core activities. The measure is conceptually comparable to the 2016 Census measure "Has need for assistance with core activities" and is available from all the most recent ABS social surveys.
The graph below indicates that estimates are reasonably consistent with those of the SDAC for people up to age 80 beyond which estimates become more varied.
Both SDAC and the Short Disability Module are designed to collect data on the full range of disability severity. The Census, however, uses a collapsed question set that specifically targets those with a need for assistance in at least one of the core activity areas. ‘Need for assistance’ with these activities is conceptually related to the profound/severe end of the disability spectrum in the SDAC and the Short Disability Module.
The more the concept 'Need for assistance' is collapsed into reduced question sets, the fewer the opportunities for people to be identified and correctly categorised, resulting in smaller populations. The Short Disability Module, with its 16 question set, produces lower estimates of people with profound or severe disability than the SDAC with its 166 question set. Likewise, the Census, with an even more limited 4 question set, results generally in a lower proportion of people being identified as needing assistance than either the SDAC or the Short Disability Module for most age groups, until ages 80 and over, when it results in a similar proportion of people being identified as needing assistance to SDAC.
In contrast, the PSS has the lowest proportion of people being identified with profound or severe core activity limitation in the age groups from 70+, most likely due to the profoundly or severely disabled in the older age groups who were unable to manage a personal interview.
As previously stated in the Overview of this paper, a possible further factor influencing the size of the population identified as needing assistance, is the way in which information is collected. Personal interviews tend to elicit higher prevalence rates than self-completion paper questionnaires.
A person is defined as having a ‘Moderate core activity limitation’ if they have difficulty with at least one of the core activities but do not require assistance to be able to manage by themselves. The graph below indicates that, moderate core activity limitation show a linear rise in prevalence as age increases. While some age groups appear to have statistically different prevalence rates in some surveys, there are no consistent patterns of difference between the SDAC questions and the Short Disability Module.
The proportion of people identified as having a mild core-activity limitation is higher in surveys using the Short Disability Module than in the SDAC. A complex set of factors account for the differences noted across all ages. Analysis of the results for mild core activity limitation show that at least some of these differences can be attributed to the fact that fewer questions were asked about this category than in the SDAC, however the vast majority of differences between SDAC and the surveys using the Short Disability Module are not significant.
The SDAC estimates of 'Education/ employment restriction only' are consistently lower than those collected using the Short Disability Module. It is possible that some of the mild core-activity limitation not identified by the Short Disability Module is instead identified as an education/employment restriction. It is also possible that the more general wording on the Short Disability Module prompt cards elicit a higher proportion of positive responses than the equivalent SDAC questions.
The major difference between the statistics collected in the SDAC and those from the Short Disability Module occurs in this category.
It is possible that in surveys using the Short Disability Module, this category includes some of the people who would have been classified as having 'Mild core activity limitation' in the SDAC. These would have been people who did not have difficulties with any of the core activities nor used aids, but who had problems with walking distances, negotiating stairs, bending to pick up objects or using public transport.
The graph belowshows the extent of possible over-reporting and how this is most marked for those aged 65 or more. This is a time of life when people may be experiencing the onset of conditions, but not necessarily be restricted by them.
The peak prevalence of 'No specific limitation or restriction' is in the age range 65-74 after which people are more likely to have a defined core-activity limitation.
This information paper has examined the differences between the disability population identified by the SDAC and that identified by the Short Disability Module for most of the Disability Severity measures, whether for the entire age range or for parts of it. Given the differences found, we can then look deeper into Disability groups to see where the Short Disability Module differs, when compared with the SDAC.
The proportion of people reporting a sensory or speech impairment is higher in the selected surveys using the Short Disability Module than in SDAC, as is the proportion of people reported as having a physical impairment. This suggests the module may be over-counting people with these types of impairments. With regard to the over-estimation of people with a sensory or speech impairment, this will be due to the SDAC asking more questions about a person’s hearing impairment than the Short Disability Module. These extra questions are used to establish whether a person is impaired by their hearing loss and reduce the number of people counted as having a hearing impairment. The reason for the difference in physical impairments is less clear and may be related to the different way in which the information is collected.
Further analysis looking at the prevalence of these impairments across age groups showed the over-reporting was consistent across the lifespan.
Footnote(s): (a) Head injury, stroke or acquired brain injury