In an interesting original paper in this issue of the Journal of Epidemiology and Community Health, the question is asked: how radical extinction of poverty would be expected to affect injury incidence and distribution per type and age (see page 899).1 In the authors’ retrospective cohort, there is a statistically significant association between lower mean household income and increased risk of injury in children’s history (especially a clear distinction at the extreme ends of the distribution). This would appear to justify support schemes for disadvantaged families based on experience gained from a variety of studies undertaken in different countries and for different accident types. However, the evaluation of the extent of poverty contribution to injury incidence is anything but simple, owing to the fact that very low-income populations are frequently under-represented in epidemiological studies on account of educational and access restraints. Moreover, the concomitant presence of additional injury risk factors renders it complicated to distinguish the independent role of poverty in injury aetiology. On the other hand, the feasibility of poverty-targeted interventions at national/international scale is questionable because there is no clear evidence that the expected benefit would justify the cost. Recent reports have indicated that strategies focusing on the general population may be more advantageous when combined with those referring to specific priority groups.2^–4

The reduction in the burden of injuries has relied on the implementation of any of two alternative options, namely passive and active measures or, in an optimal scenario, a combination of both. In the first approach, emphasis is placed on behaviour change and promotion …