Trends in rates and seasonal distribution of sudden infant deaths in England and Wales, 1988-92

BMJ 1995; 310 doi: (Published 11 March 1995) Cite this as: BMJ 1995;310:631
  1. E A Gilman, research fellowa,
  2. K K Cheng, senior lecturera,
  3. H R Winter, senior registrara,
  4. R Scragg, senior lecturerb
  1. a Department of Public Health and Epidemiology, Institute of Public and Environmental Health, Medical School, University of Birmingham, Birmingham B15 2TT
  2. b Department of Community Health, School of Medicine, University of Auckland, Private Bag, Auckland, New Zealand
  1. Correspondence to: Dr Gilman.
  • Accepted 30 December 1994

In the United Kingdom around half of all deaths between 1 month and 1 year of age are sudden—that is, cot death, the sudden infant death syndrome, or a similar description is recorded on the death certificate with or without any other cause. Epidemiological features suggest that infections, sleeping prone, exposure to cigarette smoke, and overheating of infants, particularly in the winter, may be associated with sudden infant deaths.1 2 A campaign launched in October 1991 in the United Kingdom encouraged parents to avoid putting infants to sleep on their fronts, smoking near them, and overheating them. A similar campaign in New Zealand was followed by a fall in sudden infant death rates, and a noticeable decrease in the winter peak.3 We studied the trends in rates and the seasonal distribution of sudden infant deaths in England and Wales, 1988-92.

Methods and results

We used published statistics.4 They showed that sudden infant death rates rose more or less continuously from 1971 to a peak of 2.30 deaths per 1000 live births in 1988. Rates then fell steadily to 1.44 in 1991 and abruptly to 0.70 in 1992.

The seasonal distribution is shown in the figure. Linear regression of each quarter's rates for 1988-91 showed that all quarters except the second (April-June) had a significant negative slope (b=-0.13, 95% confidence interval −0.38 to 0.12). For deaths in July-September b=-0.19 (−0.04 to −0.35), particularly steep slopes being seen in October-December (b=-0.52, −0.39 to −0.65) and January-March (b=-0.31, −0.24 to −0.39). Differences between slopes were significant (P<0.001). Between 1991 and 1992 the rate in April-June fell sharply for the first time since 1989 (by half), while the decline in January-March steepened abruptly. All quarterly rates converged over the period, until in 1992 there was little difference between them.


Sudden infant deaths in England and Wales in each quarter from 1988 to 1992


Although sudden infant death rates were falling before the campaign, the rates fell dramatically in 1992. Large falls occurred in 1992 in the first and second quarters, those immediately after the campaign's launch, while sudden infant death rates in July-September and October-December were within the range expected on the basis of the slopes during 1988-91.

In Avon the proportion of infants laid to sleep on their backs increased from 5% in September 1991 to 23% in December 1991 and 40% in June 1992.2 If the campaign had caused national changes in infant care practices which persisted throughout 1992, the larger drop in sudden infant death rates in the first two quarters of 1992 could indicate that the interaction between prone sleeping position and the underlying pathogenic mechanisms that it affects (overheating or respiratory obstruction, possibly exacerbated by infection) is more important in the early part of the year. Alternatively, the lack of a noticeable drop in rates in the last two quarters of 1992 could indicate a waning of the campaign's impact on infant care practices. Less likely, the observed trend might have been dependent on other factors.

Although it is satisfying to see such dramatic reductions, monitoring of the infant care practices featured in the campaign is needed to identify the effect of the campaign on the trend in sudden infant death. Continued monitoring of rates, and the seasonal distribution of sudden infant deaths will also be interesting.

This study formed part of a programme of work funded by the West Midlands Regional Health Authority.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
View Abstract