Age Standardised Rates

Comparison of crude mortality rates between areas which may have different age structures would be inappropriate, because
the age structure of the population can affect the number of deaths and thereby the crude death rate. To overcome this problem,
the common approach is to adjust or standardise the mortality rates to take account of differences between the age structure of
the populations. The two main methods of standardisation are Standardised Mortality Ratios (SMRs) (also called indirect
standardisation) and Age Standardised Rates (ASRs) (also called direct standardisation).

Back to top ^^

Standardised mortality ratio (indirect standardisation)

An SMR is essentially a comparison of the number of the observed deaths in a population with the number of expected deaths
if the age-specific death rates were the same as a standard population. It is expressed as a ratio of observed to expected
deaths, multiplied by 100.

SMRs equal to 100 imply that the mortality rate is the same as the standard mortality rate. A number higher than 100 implies
an excess mortality rate whereas a number below 100 implies below average mortality.

A SMR is calculated as the number of deaths observed within an area divided by the expected number of deaths within that
area. This ratio is then multiplied by 100. To arrive at the expected number of deaths, for each age group, the standard
age-specific death rate is multiplied by the local population in that age group. The number of expected deaths in each age
group are then summed across all ages to arrive at the expected number of deaths for the local population.

The template below provides details of the data required and the calculations performed to calculate a standardised mortality ratio for a particular area.

Back to top ^^

Age-standardised mortality rate (direct standardisation)

The ASR for an area is the number of deaths, usually expressed per 100,000, that would occur in that area if it had the same
age structure as the standard population and the local age-specific rates of the area applied.

Directly standardised mortality rate is calculated by dividing the number of deaths by the actual local population in a
particular age group multiplied by the standard population for that particular age group and summing across the relevant age
groups. The rate is usually expressed per 100,000.

The template below provides details of the data required and the calculations performed to calculate a standardised
mortality rate for a particular area. It also provides links to some other information on methods of standardisation that may
be useful.

Back to top ^^

95% confidence intervals are usually calculated for both SMRs and ASRs to give an indication of the level of uncertainty of
the calculation. Statistical uncertainties usually arise because these the rates or ratios are based on a random sample of
finite size from a population of interest. Confidence intervals are then used to assess what would happen if we were to repeat
the same study, over and over, using different samples each time. The precise statistical definition of the 95% confidence
interval states that on repeated sampling, 95 times out of 100 the true population value would be within the calculated
confidence interval range and 5 times the true value would be either higher or lower than the range.

However, when calculating SMRs or ASRs for different PCTs or electoral wards, the information is not based on a sample and is
therefore not subject to sampling error. It is, however, subject to random fluctuations over time or between local PCTs\electoral
wards. In this case the 95% confidence interval is a way of conveying the stability of the rates. The smaller the confidence interval,
the more stable the rate. More events lead to a smaller interval, so mortality rates from rare causes of death have quite wide the
intervals and the rates fairly unstable.

Back to top ^^

Age Standardised Rates - Datasets

Age Standardised Rates - Resources

Breslow N, Day N. Statistical methods in cancer research, volume II. The design and analysis of cohort studies. Lyon: International agency for research on cancer. WHO., 1987.

Goldblatt P, Jones D. Methods. In Goldblatt P, ed. Longitudinal Study. Mortality and social organisation, London: HMSO, 1990.

Back to top ^^