Key points

  • Over the last decade, the overall proportion of babies in England and Wales who are of low birthweight (less than 2,500 grams) has remained broadly unchanged.
  • The proportion of babies from manual backgrounds who are of low birthweight has, however, fallen over the last decade, from 8½% in 1999 to 7% in 2009.  As a result, babies from manual backgrounds are now only a bit more likely to have a low birthweight than those from non-manual backgrounds: 7% compared to 6½%.
  • Babies of lone parents are more likely to be of low birthweight than babies of couples: 10% compared to 7%.
  • The proportion of babies who are of low birthweight is similar in all regions of Great Britain.
  • See the equivalent analysis for Scotland.
  • There is a very strong relationship between low birthweight (this indicator) and the subsequent likelihood of infant death (see the infant death indicator).  For example, those of extremely low birthweight (less than 1,000 grams) are 200 times more likely to die as an infant than those of normal birthweight, whilst those just below the low birthweight threshold (2,000 to 2,500 grams) are 5 times as likely to die as an infant as those of normal birthweight.  Reflecting this, two-thirds of all infant deaths are among those borne of low birthweight, and more than half of these were borne of extremely low birthweight.

Why this indicator was originally chosen

The rates of death and illness associated with low birth weight reflect both its immediate and its long-term health risks to the infant.  It is closely correlated with poor health in the first four weeks of life, and with death before the age of two years; there are also associations with premature death from coronary artery disease. 1

Low birth weight is also associated with delayed physical and intellectual development in early childhood, and in adolescence. 2  Cerebral palsy, sight and hearing defects, and hernias are all more common in low birth weight babies. 3

Definitions and data sources

The first graph shows the proportion of babies born each year who are defined as having a low birth weight, i.e. less than 2½ kilograms (5½ lbs).  The proportions are shown separately for babies according to the social class of the father.  The social class classifications for 2002 onwards are different from those used in earlier years and range from 1 (higher managerial and professional) to 8 (never worked and long-term unemployed).

The data is for live-births only (i.e. it excludes still-births).  It is based on a 10% sample coded to father’s occupation and excludes sole registrations by mothers.

The second graph shows, for the latest year, how the proportion of babies who are of low birthweight varies according to the parents’ living status at the time of the registration of birth.  The data is based on a 100% count of live births.

The data source for the first two graphs is ONS child mortality statistics (from 2006 onwards, although the data is not publicly available) and DH3 childhood, infant and perinatal mortality statistics (prior to 2006).  The data relates to England and Wales.

The third graph shows how the proportion of babies who are of low birthweight varies by region.

The data sources for the third graph are ONS vital statistics (for England and Wales) and ISD Scotland (for Scotland).  To improve its statistical reliability, the data is the average for the latest three years.

The fourth and fifth graphs both concern the relationship between low birthweight and infant deaths (deaths which occur at ages under one year).  The fourth graph shows the number of infant deaths per 1,000 live births for each birthweight.  The fifth graph shows the proportion of infant deaths that were borne of each birthweight.

The data source for the fourth and fifth graphs is ONS child mortality statistics and relates to England and Wales.

Overall adequacy of the indicator: medium.  The data itself is large and reputable, but classification by the social class of the father may be problematic since those where no details are known about the father are not included at all.

External links

Relevant 2007 Public Service Agreements

None directly relevant.

The numbers

Graph 1

Year Social classes I to IIINM Social classes IIIM to V
1995 6.2% 7.9%
1996 6.3% 7.7%
1997 6.3% 7.8%
1998 6.3% 8.3%
1999 6.3% 8.5%
2000 6.4% 8.2%
2001 6.4% 8.5%
Social classes 1-4 Social classes 5-8
2002 6.6% 8.5%
2003 6.7% 8.5%
2004 6.6% 8.2%
20056.4% 8.6%
20066.6% 8.1%
20076.3% 7.7%
20086.5% 7.8%
20096.4% 7.2%

Graph 2

Joint registration, same address 6.9%
Joint registration, different addresses 8.9%
Registered solely by the mother 9.6%
Total 7.1%

Graph 3

East 6.6%
East Midlands 7.2%
London 7.5%
North East 7.3%
North West 7.3%
Scotland 7.0%
South East 6.4%
South West 6.2%
Wales 7.4%
West Midlands 8.4%
Yorkshire and The Humber 7.6%

Graphs 4 and 5

Birthweight (in grams) Infant deaths per 1,000 live births Proportion of all infant deaths
Less than 1,000g335 42%
1,000-1,500g52 8%
1,500-2,000g20 7%
2,000-2,500g8 9%
2,500g or more2 34%
1. Spencer, N, Poverty and child health, Radcliffe Press, 1996, page 112. 
2. Carr-Hill, R, The Measurement of Inequalities in Health: Lessons from the British Experience, Social Science and Medicine, 31(3), 1990, pages 393-404. 
3. Botting, B (ed) The health of our children, Decennial Supplement Series DS No 11, 1995, page 71.