Exploring accident causation in the construction industry
The construction industry has a longstanding reputation for offering dangerous work and has above average rates of occupational injuries and fatalities. Although fatalities have more than halved in the last 20 years, there continues to be more than one construction worker death on average per week. Earlier construction research has generated a wealth of data portraying a clear profile of accident types, but has failed to reveal what happened - the causal factors. In response to this lack of information the Health and Safety Executive (HSE) sponsored this research, the main component of which has been to undertake detailed examination of construction industry accidents. Lacking any precedent of earlier or related work, an important precursor to data collection was a review of the resources that might inform development of the methodology. The path of progress in accident research was explored through evaluation of accident causation models. The need to identify active and latent factors using a systems approach was identified. Desirable features of the systems approach were isolated and, against these criteria, construction accident models were evaluated. Construction accident models were found to be too technically orientated and focused upon human failings to fulfil the criteria for the systems approach. Nevertheless some gave good representation of failure potential through the project lifecycle, and these features were isolated for later inclusion during development of the data collection methods. To complement the theoretical development, perceptions of accident causation were gathered from groups of construction industry practitioners', by the use of focus groups. Appraisal of the accident investigation processes used in industry identified numerous problems with reporting and interpretation, and it was concluded that they are not efficient ways to explore latent conditions. Recommendations for improvements include the development of `performance assessments'; a supplementary system (to synchronise with the risk assessment process) for assessment of factors that affect performance and which are contributory in accident causations, the latent conditions. Benchmarking with industries that have moved away from the traditional organisational and safety management approaches is also advised. It will be an additional challenge to devise changes that are compatible with the unique construction management and contracting methods - any interventions will need careful management, leadership, participatory processes and cross-disciplinary development.