• Home
  • About Us
  • Contact Us
  • Links
  • Newsletter
  • Triplane to Typhoon
  • What's New
• Lancashire Overview
• Area Profiles by Area
• Area Profiles by Subject
• Research Monitors
  • Archive
  • Business Activity
  • Crime, Justice and Safety
  • Earnings, Income and Benefits
  • Economic Performance
  • Education and Training
  • Employment
  • Environment
  • Health and Wellbeing
  • Housing and Households
  • Industry Sectors
  • Manufacturing
  • Population
  • Transport
  • Unemployment
• Data Downloads
• Lancashire Focus
• Hot Topics
  • Economic Intelligence Bulletin for NW England
  • State of Lancashire Report
  • Supporting Continued Manufacturing Success
  • JSNA Health and Wellbeing
  • Small Area Profiles
  • Deprivation
  • Unemployment

Introduction

Standardised Mortality Ratios (SMRs) have been calculated by National Statistics for Wards in England and Wales for deaths under age 85 based on mortality data from 1999-2003. The SMRs are available for males, females and all persons for all Census Wards. Figures for local authorities and Government Office Regions are also provided. Full downloads of all the data by ward for England and Wales are available from National Statistics. Data relating specifically to Lancashire can be obtained from the Lancashire Profile Data Download Centre.

Background to the Data and SMRs (Based on information published by ONS)

Standardised Mortality Ratios are a means of assessing the mortality experience of local areas, in this case, wards. The numbers of deaths in a particular area do not by themselves allow the mortality of that area to be measured for comparative purposes. The first step towards this aim is to divide the total number of deaths for an area by the population of that area. This provides a crude mortality rate but this alone can be misleading as the age structure of populations can differ widely. As mortality rates generally increase with age the crude death rates of areas in which the population is predominately elderly could be expected to be worse than an area with a much younger population age structure.

To overcome this difficulty mortality rates are "age-standardised". In the case of the data shown below, the SMRs have been standardised using death rates for England and Wales. The national death rates in each age group (up to 85) were used to calculate how many deaths would be expected in a particular ward given the size and age structure of its population. This gives a total number of "expected" deaths in each area. This figure is then compared with the actual number of "observed" deaths which did take place. For example, given national death rates the number of deaths in a particular ward might be expected to be 80 but only 40 were observed. The SMR is then the ratio of the observed to expected or 40 divided by 80. For presentational purposes the ratios are normally multiplied by 100.

Figure 1 Standardised Mortality Ratio 1999-2003 for All Persons Aged under 85 in Lancashire NUTS-2 Wards

Source ONS - Standardised Mortality Ratios for Wards in England and Wales, 1999-2003

An SMR can therefore be defined as: "the ratio of the observed number of deaths in a ward to the number expected if the ward had the same age-specific rates as England and Wales".

As national death rates are used as the standard for the results, the SMR for England and Wales is 100. The results for wards therefore allow their mortality experience to be compared with the national average. If an SMR is less than 100 that means the number of deaths for a ward was less than would have been expected. Conversely if an SMR is greater than 100 the number of deaths was greater than expected.

Lancashire Results

As context information, Table 1 presents SMRs for Lancashire Districts. Just three of these districts (Fylde, Ribble Valley and Wyre) have a SMR of less than 100, meaning that the number of deaths in these areas was less that the number that might have been expected if they had the same age-specific rates as England and Wales as a whole. Conversely, in all other Lancashire districts the SMR was greater than 100, meaning that the number of observed or actual deaths was greater than expected. This differential was greatest in Blackburn with Darwen and Blackpool which in terms of their SMRs ranked 10th and 14th in England and Wales. Overall, five Lancashire districts (also including Burnley, Hyndburn and Preston) had SMRs in the top 10% in England and Wales.

Lancashire Wards

Even from a cursory examination it is apparent that overall Lancashire has a marked skew in its distribution of ward SMRs compared with England and Wales with a much higher proportion of wards exhibiting higher SMRs. No less than 36 Lancashire wards are ranked within the top 5% for mortality in England and Wales whilst only three wards fall within the 5% range recording the lowest SMRs. There are 64 Lancashire wards in the top 10% and only eight wards in the lowest 10% in England and Wales.

The highest ward SMRs are recorded in Bloomfield and Waterloo wards, both in Blackpool, with SMRs of 171 and 170 respectively and all the other high ranking areas are based within the larger urban areas and mainly those which are also known to rank poorly in terms of numerous deprivation indices (Table 2 and Figure 1). The lowest SMRs are to be found in Blacko & Higherford in Pendle and Mellor in Ribble Valley. It is no surprise to find that the wards with the lowest mortality rates are in general in more rural locations and are characterized by the presence of higher socio-economic groupings and rank highly in terms of household incomes and other indicators of well-being.

Figure 2 Standardised Mortality Ratio 1999-2003 for Females Aged under 85 in Lancashire NUTS-2 Wards

Source ONS - Standardised Mortality Ratios for Wards in England and Wales, 1999-2003

Figures 2 and 3 map the distribution of ward SMRs for males and females separately. The accompany data for these figures may be found in the Data Download Centre.

Notes Pertaining to Interpretation

Care needs to be taken in the interpretation of the above ward-based distribution of SMRs:

• Ward-level SMRs are often based on rather small numbers of events. As these will be affected by some random variation the results are therefore subject to some margin of error. Consequently, ONS have calculated 95% confidence intervals for the figures. For comparisons between areas, the confidence intervals allow those differences which are statistically significant at the 95% confidence level (confidence limits do not overlap) to be distinguished from those more likely to result by chance (confidence limits do overlap).
• SMRs have been calculated only for people who died before the age of 85. The reason for this is to try and ameliorate the impact that can happen in the case of wards that have particularly large proportions of their total population in nursing homes and other communal establishments. As mortality in nursing homes is higher than in the general population it may appear that death rates in these areas are unexpectedly high. There are a handful of wards in Lancashire that appear to have been affected by this. In some areas other local factors may influence the SMRs. Local knowledge may therefore be an important factor when interpreting results.
• SMRs based on the "indirect standardisation" method used here by National Statistics allow (subject to confidence limits) comparisons to be made with a national average as the results take into account differing age structures. However, if two areas have identical death rates in every age group, but different population age structures, their SMRs may differ. In consequence, direct comparisons between wards or between sexes can sometimes be misleading. Alternative summary measures may need to be used to overcome these limitations such as life expectancy or directly age-standardised mortality rates. Life expectancy data for districts is outlined elsewhere on the Lancashire Profile Site and ONS are proposing to publish figures showing life expectancy at birth for wards in England and Wales at a later data.

Figure 3 Standardised Mortality Ratio 1999-2003 for Males Aged under 85 in Lancashire NUTS-2 Wards

Source ONS - Standardised Mortality Ratios for Wards in England and Wales, 1999-2003

For further details, please contact:
Peter Kivell
Tel 01772 534157
Email Peter.Kivell@lancashire.gov.uk