Accident and incident data

Other techniques for accountability are useful. A later section will discuss leading and trailing indicators for safety. Accident and incident data, for example, are types of trailing indicators. They provide delayed results and trail many important preventive activities. Leading indicators measure safety activities which occur independent of accidents and incidents. Regular tailgate talks for constmction crews is one form of leading indicator. 

The collection of accident and incident data and the systems which have to be in place if accident and incident data are to be collected and recorded effectively. 

The main uses of accident and incident data such as the techniques for learning from the analysis of aggregated accident and incident data, using trend analysis, comparisons of accident and incident data, and epidemiological analyses. 

The relationship between accident and incident data and risk assessment data. In addition, there is an appendix dealing with the UK legal requirements to notify accidents causing injuries of particular types and levels of severity. 

Table 2.5.1 The main types of accident and incident data...

TTiis means that the first stage in any work on accident and incident data should be a clear definition of the particular subset which will be used. To be of practical value, fhe definition should deal with the following ... 

In this section a more detailed look is taken at how accident and incident data can be used to learn from what has gone wrong in the past so that risk control measures can be implemented or improved. There are three main aspects ... 

Making comparisons using accident and incident data. 

The collection and use of accident and incident data 283 Table 2.5.2 Comparisons using incidence, frequency and severity rates... 

Accurate accident and incident data will provide a measure of what has gone wrong in the past, and allow comparisons over time (trend analyses) and comparisons between one organisation and another. What these data will not do, even if they are accurate, is to provide a measure of risk. 

True levels of risk in an organisation can only be determined accinately using appropriate risk assessment methodologies, details of which will be found elsewhere in this book. However, more detailed discussion of the relationship between accident and incident data and risk assessment data will be found in section 2.5.8 of this chapter. 

The techniques of epidemiological analysis were first applied to the study of disease epidemics and historical example will be looked at by way of illustration to show how epidemiological techniques can be applied to accident and incident data. 

Accident and incident data and risk assessment data... 

There are two types of accident and incident data to be considered, the aggregated data used for trend and epidemiological analyses and the data on single accidents and incidents collected during investigahons. Each of these data t) es is dealt with separately. 

As has already been mentioned accident and incident data do not provide a measure of risk. This is because the number of accidents and incidents depends on three factors ... 

The techniques are based on probabilities which many people find difficult to understand. In essence, the fact that an accident happens does not mean that the risk assessment was incorrect. For example, if it is correctly calculated that there is a very low likelihood of a multiple fatality, the fact that the multiple fatality occurs does not necessarily mean that the estimate of likelihood was incorrect. Rather, it is the third of the factors listed above, i.e. the operation of chance. Whaf is required in the longer term are numerical techniques for risk assessment which identify the xmderlying level of risk and fhe exfenf to which risk control measures will reduce the risk. It will then be possible to predict the number of accidents and incidents that will occur by chance and this can be compared with the numbers of accidenfs and incidents that do occur. It may then also be possible, by examining the accident and incident data in more detail, to determine whether any problems are due to an xmderestimate of fhe underl5ung level of risk or a failure to select or implement appropriate risk control measures. These are discussed by Boyle. ... 

This section consists of a brief description of the sorts of computer software which are available for the recording and analysis of accident and incident data, and for a range of related data handling tasks. The criteria to be used in selecting software are also briefly discussed. 

Setting up the database so that it will do the recording and analysis required. If, for example, a general database is to be used to record and analyse accident and incident data it would be necessary to set up the fields for recording such things as name of person injured, time of... 

Measurement is only the first step in using survey techniques. Collected data, including accident and incident data, should be used for discussion with operators. Utilising the data in this way is as important as getting it in the first place. The involvement of operators in this way raises expectations - operators expect more from managers and managers expect a greater contribution from operators. 

---