Guidelines for conducting community surveys on injuries and violence

Global mortality data indicate that in 2002 an estimated 5.2 million people died as a result of some form of injury. This number, however, does not reflect the numerous individuals who are non-fatally injured, many of whom suffer lifelong health consequences and disabilities. Combined data from high-income countries, such as Australia, the Netherlands, Sweden and the United States, indicate that in these countries for every person killed by injury, approximately 30 times as many people are hospitalized and roughly 300 times as many are treated in hospital emergency rooms and then discharged. When disability resulting from injury is also taken into consideration, injuries represent an even more significant public health problem. Developing countries account for approximately 90% of all injury fatalities world-wide. It is in these environments where the least has been done to prevent injuries. This is because of the lack of visibility of the problem, multisectorality and a lack of ownership and a failure to realize that injuries can be prevented through organized efforts of society. The present paper reports on the development of guidelines for conducting community surveys, which will contribute to increasing the visibility of the injury problem, especially in lowand middle-income countries, by providing a reliable methodology for their documentation. In many lowand middle-income countries vital statistics and routine health information may be lacking or, at best, patchy. Furthermore, in these settings demographic data may be incomplete or out of date, because it has been some years since a population census has been held. This situation may be compounded by the effects of wars, which lead to large population displacements, as has happened in countries of subSaharan Africa. Additionally, in post-conflict conditions the problem of injuries may be greater due to the ready availability of firearms and the loss of social cohesion in local and displaced populations. Routine health service information may be incomplete or inaccurate and reliance on it may lead to underestimates of the numbers of injured people seeking treatment in hospitals and other health facilities. As a result of a combination of these factors, estimates of the burden of injuries in lowand middle-income countries may be unreliable. Estimates for these countries are often based on projections from countries with more comprehensive injury data or on projections from population laboratories. Community-based, household surveys are one way of obtaining data on injury occurrence and deaths; in some settings such surveys also provide a means of collecting baseline population denominator data, necessary for the calculation of prevalence or incidence rates and that otherwise would not be available. In less resourced environments, a community survey to assess injuries can be a stand-alone method for the surveillance of injuries or a valuable adjunct to hospital-based injury surveillance systems. Adequate data about types of injuries, their causes and consequences are vital to understanding the scale and nature of the local injury problem and, subsequently, the implementation and evaluation of effective prevention programmes. Reliable information on injuries is also important to build up a picture of the extent of the national, regional and global problem. To address the need for improved injury surveillance, in 2001 the World Health Organization (WHO),