previous | contents | next

Life Expectancy for areas in Scotland, 2008-2010

2. Background methodology

2.1 Life expectancy at birth

2.1.1 The figures quoted in this publication are a three year average. They are produced by aggregating deaths and population data for the three year period. Three years of data are needed to provide large enough numbers to ensure that the figures published in this report are sufficiently robust.

2.1.2 The figures are all period life expectancies. Period life expectancies are calculated using age specific mortality rates for a given period. They do not make allowance for any actual or projected future changes in mortality after that period. This means that life expectancy at birth for a given time period and area is an estimate of the average number of years a new born baby would survive if he/she experienced the particular area’s age specific mortality rates for that time period throughout his/her life. The figure reflects mortality among those living in the area in each period, rather than mortality among those born in each area. Life expectancy at birth is not simply the number of years a baby born in the area during the three year period is expected to live (although the term ‘can expect to live’ is used throughout this publication for ease of reading), both because death rates are likely to change in the future and because many of the newborns may live elsewhere for at least some part of their lives.

2.1.3 The abridged life tables in Table 1, Table 2, Table 3, Table 4 and Table 5 are constructed from the estimated populations in 2008, 2009 and 2010 and the total number of deaths registered in these years. The column headed Ix shows the numbers who would survive to the exact age of x out of 100,000 persons who, from birth, were subject to the mortality probabilities indicated by the death records for 2008-2010. The column headed ex0 shows the expectation of life - the average number of years of life left to persons aged exactly x who are subject to the 2008-2010 mortality probabilities from age x onwards. For example, in Table 1, for Glasgow City, 73.1 per cent of the male population and 84.5 per cent of the female population could be expected to survive to age 65. In Table 2, a male aged 65 in Fife NHS Board could be expected to live on average another 17.1 years, and a female aged 65 could be expected to live on average another 19.7 years.

2.2 Accuracy of results

2.2.1 Life expectancy is an estimate which is subject to a margin of error. The accuracy of the results can be indicated by calculating a confidence interval within which the true underlying life expectancy would be expected to lie. In this publication, confidence intervals are set at the 95 per cent level, with the true underlying life expectancy being found within its upper and lower limits with 95 per cent probability. Life expectancy for an area can usually be said to be significantly different from another area if the 95 per cent confidence intervals do not overlap. Also, errors in the population estimates used for calculation purposes must be considered.

2.2.2 The 95 per cent confidence intervals for life expectancy at birth are illustrated in Figure 3, Figure 4, Figure 5, Figure 6 and Figure 7 for Council, NHS Board and Community Health Partnership areas within Scotland as well as for the Urban/Rural classification and Deprivation deciles.

2.2.3 There is no simple ‘rule of thumb’ for the size of confidence intervals, although it largely depends upon the size of the population: areas with small populations tend to have wider confidence intervals. It is also worth noting that life expectancy results in these areas can be affected by the random variation in the annual number of deaths. This means that estimated life expectancy can vary from year to year. For example, in Table 6, Table 7, Table 8 and Table 9, changes in rank may simply be due to random variation from year to year in the population and number of deaths (particularly in the smaller areas such as Orkney and Shetland) as opposed to a change in the long term trend life expectancy. Rankings are usually only significantly different if the confidence intervals do not overlap e.g. in Figure 4, life expectancy at birth for females in Orkney is only significantly different from that of females in Greater Glasgow and Clyde and Lanarkshire; all other Health Boards have overlapping confidence intervals and therefore do not have significantly different female life expectancies.

2.2.4 A description of the methods used to calculate life expectancy for different geographical areas and other classifications within Scotland can be found in the Life Expectancy for Scotland: Methodology Guide on the NRS/GROS website.

More information about the methods used to calculate the standard errors and confidence intervals used in this report can be found in the Office for National Statistics publication, number 33 in the GSS Methodology Series. The report details the research undertaken by the Office for National Statistics (ONS) to compare methodologies for calculating life expectancy figures and confidence intervals. It can be found on the ONS website.

previous | contents | next