Trend analysis of socioeconomic differentials in deaths from injury in childhood in Scotland, 1981-95

BMJ 1999; 318 doi: (Published 27 February 1999) Cite this as: BMJ 1999;318:567
  1. Anita Morrison, research officer (Anita.Morrison{at},
  2. David H Stone, director,a,
  3. Adam Redpath, principal statisticianb,
  4. Harry Campbell, senior lecturer in epidemiologyc,
  5. John Norrie, enior statisticiand
  1. a Paediatric Epidemiology and Community Health (PEACH) Unit, Department of Child Health, University of Glasgow, Glasgow G3 8SJ
  2. b Information and Statistics Division, Scottish Health Service, Trinity Park House, Edinburgh EH5 3SQ
  3. cDepartment of Public Health Sciences, Medical School, University of Edinburgh, Edinburgh EH8 9AG
  4. dRobertson Centre for Biostatistics, Boyd Orr Building, University of Glasgow, Glasgow G12 8QQ
  1. Correspondence to: Ms Morrison
  • Accepted 15 September 1998

Injuries are a leading cause of death and disability among children in the United Kingdom, costing an estimated £200 million annually in direct costs to the NHS.1 Recent research from England and Wales suggests that death rates from injury and poisoning have fallen in both sexes and all social classes.2 The decline in those of lower social class has been much smaller, however, resulting in widening socioeconomic differentials in mortality from injury. To determine whether this phenomenon has also occurred in Scotland we examined time trends in mortality from injury in Scottish children for 1981-95.

Subjects, methods, and results

National data on external cause of death (injury and poisoning; E800-999 of the ninth edition of the International Classification of Diseases) for children aged 0-14years were obtained from the registrar general for Scotland for the period 1981-95.Socioeconomic status was measured with Carstairs' deprivation index on the basis of characteristics of Scottish postcode sectors.3 This involves a continuous scale of 1-7,where 1is most affluent and 7is most deprived. The per cent reduction in fatalities from 1981to 1995 was estimated with Poisson regression models for deprivation categories 1-2,3-5,and 6-7and for all deprivation categories combined. In all models a log linear time trend was found to be adequate. Confidence intervals were adjusted for overdispersion.4 A further Poisson regression model compared the least and most deprived groups.

In 1981-95,1728Scottish children (65% boys) aged 0-14years died as a result of an injury or poisoning. Overall, there was a decline of 58% (95% confidence interval 45% to 68%) from 1981 to 1995with observed death rates per 100000 averaging 13.2for 1981-93compared with an average of 7.4from 1993-5.The proportion of deaths due to injury decreased only slightly, from 14% of deaths in children in 1981-3to 12% in 1993-5.

Throughout the study period children residing in areas of relatively greater deprivation (6-7) experienced higher mortality from injury than children living in more affluent areas (1-2and 3-5) (figure). Similar proportional decreases were observed in all deprivation categories over the study period. Between 1981 and 1995mortality from injury decreased by 60% for categories 1-2and 3-5(27% to 78% and 47% to 70%, respectively) and by 53% (27% to 70%) for categories 6-7.The observed average mortality per 100000 for 1981-3compared with 1993-5was 9.9versus 5.6, 13.3versus 7.0,and 17.0versus 10.7for categories 1-2,3-5, and 6-7,respectively. The risk ratio between the most and least deprived groups was 2.29(1.82to 2.88). There was no evidence that this had changed over the study period (P=0.56).


Mortality from injury per 100 000 children by category of deprivation, 1981-95


Despite overall improvements in childhood mortality from injury in England and Wales, children in the lower socioeconomic groups seem to be experiencing progressively increasing relative risks of injury in comparison with those in higher socioeconomic groups,2 possibly as a result of a differential impact of health education efforts. Our data suggest that this may not be the case in Scotland. The reason for this difference north and south of the border is unclear. One hypothesis is that, unlike in England and Wales, measures to prevent injury in children are exerting an equal effect on all socioeconomic groups in Scotland. An alternative explanation is that the decline has occurred independently of specific preventive efforts.

We cannot rule out the possibility that the cross border differences are due to the way socioeconomic status is measured. The Carstairs' deprivation scores adopted in Scotland and the social class definitions used in the analysis of England and Wales data are not necessarily directly comparable. The Carstairs system adopts a measure of “area” deprivation, whereas the measure of social class used in England and Wales categorises the population into five classes based on individual occupational status. This may be important as there is some evidence that the characteristics of communities, as well as those of individuals, exert an impact on the health of the population.5 Further investigation of the apparently divergent socioeconomic patterns of mortality from injury in children within the United Kingdom is necessary.


Contributors: AM discussed core ideas, participated in data analysis, data interpretation, and writing of the paper. DS contributed to the initiation of the study, discussed core ideas, participated in data analysis and interpretation, and contributed to the writing of the paper. AR contributed to the initiation of the study, provided the data, participated in data analysis and interpretation, and edited the paper. HC contributed to the initiation of the study, participated in data interpretation, and edited the paper. JN provided statistical advice. DHS is guarantor for the study.

Funding: None.

Competing interests: None declared.


View Abstract