Age standardisation
We use age-standardised data in the general health section of our Health, disability and unpaid care topic report.
Age standardisation
Age is a risk factor in health outcomes, as many illnesses/conditions occur more frequently as people get older. Age-Standardised Percentages (ASPs) allow for comparisons to be made in health outcomes between populations living in different geographical areas and over time that may contain different age structures. ASPs adjust observed rates for a health outcome (e.g. disability) to a standard population with a known age structure. This means that the populations being compared are adjusted to have the same age structure as the standard population. ASPs therefore allow for comparisons to be made using a single figure which accounts for different age structures, without the need to present and compare age-specific percentages.
For example, Aberdeen City has a slightly higher raw percentage (observed count/population) of people with very good general health at 49.2% than Na h-Eileanan Siar at 48.7%. However, Aberdeen City has an ASP of 48.2%, lower than the ASP of 53.5% for Na h-Eileanan Siar. The raw percentage implies that the populations in both local authorities have similar health experiences/outcomes. Presenting the ASPs allows us to both identify the difference in health outcomes between Aberdeen City and Na h-Eileanan Siar and, and to say that this difference is due to prevalence of people with very good health, rather than a difference in the age of their populations.
European Standard Population 2013
ASPs were calculated using the European Standard Population (ESP) 2013. The ESP is a theoretical population based on an average of 2011-2030 population projections for EU27 and EFTA countries. Table 1 shows the age-standardised percentages from the ESP. They reflect the proportions that would occur if the observed population structure in the Scotland Census 2022 was the same as the ESP. ASPs have been produced for Scotland and by sex and local authority.
Census data on unpaid care was only collected for those aged 3 and over. To age-standardise this, the lowest age-band used was 3 to 9 years old. The ESP percentages were also adjusted to reflect the removal of those under 3 from the population.
Table 1. European Standard Population
|
Age band |
European Standard Population 2013 (%) |
|
0 |
1.0 |
|
1 to 4 |
4.0 |
|
5 to 9 |
5.5 |
|
10 to 14 |
5.5 |
|
15 to 19 |
5.5 |
|
20 to 24 |
6.0 |
|
25 to 29 |
6.0 |
|
30 to 34 |
6.5 |
|
35 to 39 |
7.0 |
|
40 to 44 |
7.0 |
|
45 to 49 |
7.0 |
|
50 to 54 |
7.0 |
|
55 to 59 |
6.5 |
|
60 to 64 |
6.0 |
|
65 to 69 |
5.5 |
|
70 to 74 |
5.0 |
|
75 to 79 |
4.0 |
|
80 to 84 |
2.5 |
|
85 to 89 |
1.5 |
|
90 or over |
1.0 |
Calculation used for age standardisation
Age standardisation was applied in three steps.
1. Age-specific percentages were calculated for each age group:
Pi=(Ci ÷ Ni) × 100
Where:
P = percentage of people
C = the number of people in a category
N = the number of people in the target population
i = age group
For example, to get the age-specific percentage of disabled people in the 30-34 age group, we would divide the number of disabled people aged 30-34 by the number of people aged 30-34 in the Census data, then multiply by 100.
2. The age-specific percentages were then multiplied by the proportion of the corresponding age group in the ESP:
Mi=Pi× Ei
Where:
E = European Standard Population
For example, the age group 30-34 represents 6.5% of the ESP, so we would multiply the age-specific percentage of disabled people in the 30-34 age group by 6.5.
3. The age-standardised percentage is then calculated by adding these together to obtain one number, then dividing by the total ESP. The calculation for ASPs is therefore:
ASP=∑(Pi × Ei) ÷ ∑ Ei