Injury prevention programmes in primary care: a high risk group or a whole population approach?

OBJECTIVE: To examine the relationship between risk factors for childhood unintentional injury and injury outcome and to assess the feasibility of using risk factors to identify children at high risk of injury. SETTING: One general practice in Nottingham, UK. METHOD: A postal questionnaire survey to all parents of children registered with the practice (n = 771) to obtain data on risk and sociodemographic factors. All children still registered with the practice one year later were followed up for occurrence of a medically attended injury. RESULTS: The response rate was 78%. The injury rate over the follow up year was 246 injuries per 1000 children. Previous medically attended injury was associated with each of the injury outcomes (odds ratio (confidence interval) for all attendances, 2.33 (1.37 to 4.05); for accident and emergency attendances, 2.27 (1.15 to 4.4); and for primary health care team attendances, 2.58 (1.33 to 5.0)). Male sex was associated only with accident and emergency department attendance (odds ratio 2.13 (1.06 to 4.2)). Maternal age and previous injury were associated with a higher number of injuries in the subsequent year on univariate and multivariate analyses. The sensitivity and positive predictive value of the risk factors were low, except for previous injury and male sex. The number of children needing an injury prevention intervention to prevent one injury as identified by the risk factors was not significantly different from that required if a whole population approach were to be used. CONCLUSION: Primary care based injury prevention programmes, at present, should not be targeted at children identified as being at 'high risk' of injury. Nevertheless, a larger study using a wider cross section of the population is needed to address this issue further.