For healthcare professionals only

Education And Debate

Using data from the 1991 census

BMJ 1995; 310 doi: (Published 10 June 1995) Cite this as: BMJ 1995;310:1511
  1. F Azeem Majeed, lecturer in public health medicinea,
  2. Derek G Cook, senior lecturer in epidemiologya,
  3. Jan Poloniecki, lecturer in medical statisticsa,
  4. David Martin, lecturerb
  1. a Department of Public Health Sciences, St George's Hospital Medical School, London SW17 0RE
  2. b Department of Geography, University of Southampton, Southampton SO17 1BJ
  1. Correspondence to: Dr Majeed.
  • Accepted 13 April 1995
View this table:

Approximate 95% confidence intervals for the difference between a sum from a table containing modified cells and the corresponding actual value

Box 3—Effect of data modification on small area statistics

SAS table 12—Long term illness in households: residents in households with limiting long term illness. Data taken from an enumeration district in Wandsworth Health Authority

View this table:


Small area statistics are released only for enumeration districts with 50 or more residents and 16 or more households. For enumeration districts below either of these thresholds, the only data that are available are the total number of people present, the total number of residents, and the total number of resident households. All other small area statistics for these enumeration districts are combined with those for a neighbouring enumeration district, provided that the combined total number of people and households is greater than the minimum thresholds. Local base statistics are released only for those wards with 1000 or more residents and 320 or more households. As with enumeration districts, data from wards which fall below either threshold are combined with data from a neighbouring ward. Enumeration districts or electoral wards that contain data from neighbouring areas are categorised as importing zones.


Enumeration districts that were expected to contain 100 or more people living in one or more communal establishments (defined as an establishment in which some kind of communal catering is provided, such as nursing homes, hospitals, and prisons) on the night of the census were defined as special enumeration districts. The residents of such establishments often have different social and demographic characteristics from people living in the surrounding area. Therefore, to prevent the residents of these establishments distorting the small area statistics for the enumeration district in which they are located, these areas were treated as special enumeration districts. The total population and total number of households for special enumeration districts are published, …

View Full Text

Log in

Log in through your institution


* For online subscription