Investigating Accidents in the Home *A paper read to The Bristol Medico-Chirurgical Society January 1972.

The value of on-the-spot study of accidents as a research method has been well established for road traffic, for aircraft and for railway accidents (States and States 1968, Robertson, Mclean and Ryan 1966, Jamieson et al. 1971, Kolbuszewski et al. 1969, Hadon, Suchmann and Klein 1964, De Haven 1942). The methods were developed in these fields as a result of the need for closer scientific scrutiny of Srowing accident problems at a time when the other major causes of premature death and disability were declining because of advances in public health and medical care. As accidents were studied more closely underlying causal factors were detected in the physical and psychological environment; remedies of re-design ?f equipment and structures and of re-training and reeducation of key people?drivers, pilots, navigators? Were begun. The basis for developing the prevention Programmes was on-the-spot investigation. The most satisfactory method of studying an event is to observe and measure its characteristics as it occurs. This cannot be done with accidents because ?f their comparative rarity and their unpredictability in time and space. But if we can set up a fast information system so that when an accident occurs we can as quickly as possible visit the scene with a team of trained investigators, observe the results of the accident, examine the phenomena involved, interview witnesses, attempt to determine the sequence of events 'eading up to the accident and to the injury and damage, we may be able to construct a sufficiently realistic model of the factors involved. This model may then be used in developing general theories of causation and for prevention in other similar situations. Further laboratory experiment and control surveys may be necessary to test the model and subsequently community experiments or field trials must be done to explore the cost and value of the preventive measures. By giving an event the label 'Accident' we generally consider it an event of momentary duration which happens suddenly "out of the blue". Chance or the gods determine its action and its victim probably considers it "just one of those things". Primitive man, who witnessed lightning, attempted to explain it in terms of the super-natural; but with knowledge of electricity we can now better appreciate the events leading UP to a flash, which is only a single incident in a successive chain of physical events, involving cloud, rain and electricity extending both into the pre-flash and post-flash periods of time; and we can explore mathematically the costs and benefits of lightning conductors on buildings of known height and value. Our model must embrace measures of predictability, preventability and cost and benefit. A similar approach to other so-called accidents?for example in transport?has resulted in changes in our understanding of human environmental control (De Haven 1952) and in particular has established the important conceptual distinction between accident causation and injury causation. The prediction of fatal injury in aircraft accidents has depended on detailed comparative study of survivors and non-survivors in given crashes and careful identification of injury causation (Hasbrook 1953). A collision between two cars appears to the casual bystander as a sudden momentary event. However, the pre-collision factors could extend for a long period into the past, quite unknown to the witness. The factors may include poor vehicle maintenance, alcohol taken by the driver, a family quarrel or a poor road surface. All these factors require investigation and study before scientific prevention can be envisaged. Information is obtained by studying the scene and the vehicle, and interviewing those involved. The sooner these measures are carried out the more reliable and complete the data that will be collected.