TRANSPORT RESEARCH LABORATORY TITLE by Road accident data recording I A Sayer Overseas Centre Transport Research Laboratory Crowthorne Berkshire United Kingdom XA 11 12(B[P
HR
SAYER, I A (1 994). Road accident data recording. Paper presented to the International Course on Prevention and Control of Traffic Accidents and Injuries. New Delhi, India, 8- 16 December 1994. Prclvcllion & Cour rof of Traffic 1njrrics December 1994 ROAD ACCIDENT DATA RECORDING 1 A SAYER TRL, UK OBJECTIVES To demonstrate the importance of having reliable data collection and analysis systems. Withou t a reliable system, little, progress can be made in either understanding the factors that comprise road accidents or developing effective remedial measuire programmes. INTRODUCTION Developing countries have a serious and growing road accident problem. Traffic accidents have become a major problenm in many Third WVorld countries and result in significant social and economic losses that they can ill afford. Little progress can be made in improving the accident situation until the problem has been clearly defined and an understanding gained on where, why and to whom road accidents are happening. Road accident data therefore needs to be collected to meet the broad range of purposes and groups involved in the I eld of road safety. In many developing countries road accident data collection, storage and analysis systems are non existent or are largely inadequate. Existing systems can range from those where only the barest details are collected to those where an enormous amount of data are collected, much of which is totally irrelevant to the needs of accident investigators, engieers etc or it is inaccessible because of the way in which it is stored. Road accidents are complex occurrences which makes the collection of reliable data difficult. Thus;. it is important that data collection is carried out in a systematic and uniform manner using standardised fo~r-mis to'ensure that each accidents is ircated in the same way. Once appropriate fords are available, police responsible for reporting road accidents need to be trained in their use and application. In somne developing countries illiteracy and poor training in the traffic force can be overcome by the use, of specially designed report booklets an.rd the training of police selected to specialise in such work. ACCIDENT REPORTING IN TiHE UK. Road accident st atistic's were first collected in a form which could yield national data in 1909 -when there were 1070 fatal accidents. The first serious, and substantial publication of road accident statistics was by the Ministry of Transport in 195 1. In order to keep pace with changing times over the years, the original accident report form(s) has been modified sever al times. Only in 1982 did all the authorities use the specially designed 'Stats 19' form. These form are completed by reporting officers details of which are sent to the Department of Transport every month for national collation and analysis. Since 1974, UK local authorities have had a statutory responsibility for road safety and as a result collected information about accidents in their are-as for the past 20 Years. Information collected is usually based on the national data together with any extra items collected locally for their own use. In the vast majority of cases in the UK, the accident report booklet is filled in by a Police reporting officer visiting thie scene of the accident. The police or local authority enter the accident details on to computer where tlie records are validlated belore being mnade available for both local and national analysis. ACCIDENT REPORTING IN DEVELOPING COUNTRIES 63 )',~cic~iiioit & Controlt 7'ri Trnfite Injurics Iccbr19 A:'; Ca' Iv stud)' (Jacobs et al 1975) of datat collection and analysis in developing countries found that v'ery few Third 'orl.J countries operated an adequate road accident data collection band analysis system. Although most of the co;untries surveyed collected data there wvas a tendency for the repo:it forms to be either too comiplicated or i:.d ru~ ot the tap;k- of accident reporting. At the time of thie study none of the countries represented transferred d 0uaont a computer based storage and analysis system. '.:hat can be achieved in a country's road safety programme, is in many ways dependent upon the quality of its accident reporting. In the vast majority of developing countries thie data used in accident prevention programmes are obtained from reports prepared by police officers. However, a police accident report form, such as a sheet of paper w~ith few headings that give little or no assistance to the reporting officer discourages the creation of a reliable and consistent datei base. Reporting systems requiring accident details to be. recorded on sheets of paper that contain only a minimum of guidance for the reporting officer, make it difficult to establish the common core of data essential needed for accident investigation. Equally, a form or booklet that is too long or complex wvill discourage officers from filling it in accurately. With developing countries becoming aware of their accident problem, an increasing number are seeking ways to prevent and reduce their occurrence. Developed countries confronted with the same problem found it necessary to adopt a scientific approach to uncover the underlying factors involved in road accidents. Such an approach requires a level of detail and accuracy in its accident reporting and recording system that many Third world countries do not haive. Shortages of experience, professional skills and money also contribute to the problems of providing adequate le-vels of accident recording and analysis in industrialising countries. TRL RESEARCH A TRL- research programme into developing an accident report booklet of use by police in Third World countries found that booklets designed as a desk exercise, without adequate provision for evaluation by police officers. using them as part of their daily routine, will almost certainly contain faults in design, layout and content. Important in any evaluation programme is the opinion of future booklet users. In general booklets or forms wvith the least number of questions were preferred. This implies that very careful thought should be given to the amount of information asked for in a reporting system. Keeping the number of questions to a minimum should be conducive to accurate reporting and improving reporting rates. However, booklets must still contain sufficient information to meet local, regional and national data needs, accident investigators and courts requirements. Anmpotn part of the booklet design was the pre coding of the reported accidens data. Pre coding reduced the tuine needed to input the accident data into a microcomputer system and reduced inputting errors. Symtbols wvere found to be useful in a limited number of situations and where they were used few errors were reporting errors were found. THEl NEED FOR ACCURATE ACCIDENT RECORDS Many (lifferent groups and organisations use accident data and it is important that data collection is carried out in a systeniatic anrd uniform manner using standardised forms or booklets. Accident statistics are of value to: I. [lie p)oliCe 2. government departments 3. local Ihighwas~y authorities 6-4 December 1994 Frree'nfion & Control of Traffic Injurles December 1994 5. national re-c'ords offices 6. research institute-sii~ 7. road engineers and planners 8` coinmerciagl;19 insurance organisations. It is important that accident information is recorded as accurately a's poslible. Lack of reliable data makes it impossible to compile effective and reliablp statistics. Inaccurate reporting may seriously affect the overall accident picture, estimates of accident costs and remedial measure investment programmes. Reliable data not only helps ensure reliable statistics but the details may be required as evidence in court proceedings. Thus, whether for national, broad statistics or for use in detailed research. there is a need for the accident data collected to be both accurate and uniform. Without agreement for such terms as 'fatal' injury or 'intersection' little can be done in the w ay of producing or maintaining long term accurate road accident information. Accuracy of data begs the question of what should be collected. Prepared forms or booklets are now used in many countries for data collection but the design and content vary markedly. Some countries use the same form for data collectionand subsequent analysis while others use different documents for the two stages of the process. Obviously a ba'lancehbas to be struck between what is the minimum- needed for bar e national statistics and the much more detailed information needed for accident reduction programmes. Formlbooklet design. and content are not considered here except to say that they should be easy to complete, contain data that is relevant to the subsequent analysis and not require reporting skills and knowledge that recording officers' may not have.. DATA ANALYSIS To make full use, of the road accident data collected it is essential that an adequate data processing system be u-sed. Analysis by hand is time consuming, labour intensive, limited and prone to error. Main frame computers have had a chequered history when used in the analysis of road accident data in iDeveloping countries. The rapid development * and availability of microcomputers has opened up new possibilities for tanalysing road accident data efficiently and imaginatively. lIn 19S I the Overseas.Centre of the Transport Research Laboratory begn developing its Microcomputer Accident * Analysis Package (MJAAP). The package, originally developed as patf a cooperative road safety research programme iii Egypt has recently been enhanced to include a mouse driven pull down menu system, with extensive graphical outputs. Advantages c~f micerocomputers include them being relatively cheap, ph~ysically robust, tolerant of difficult environments,. portable, relatively easy to maintain, generally accessible with 'user friendly' programs designed for thie job in hand.. 1 losvever, successful analysis will not come from the hardware alone. ~rrained, experienced personnel are still needed to adnminister. a~nd execute the collection of data, to analyse, interpret and present it in a meaningful manner. To mlake the mnost e fficient use of microcomputer outputs and other road safety resources, an effective institute (or some other such organisation) is required. Such an institute wvill initiate integrated road safety action pla-ns containing clear objectives and goals. P EFERENCE~S 65 Pre).rilloi & Control of Traffic Injuries December 1994 D)EPART1N!ENT OF TRANSPORT, 1986. Accident Investigation Manual. Department of Transport, London. DEPARTMENT OF TRANSPORT, 1987. Accident Investigation and Prevention. Traffic Advisory Unit Leaflet .4/87, DOT, London. l! LLS, B L and ELLIOT, G, 1986. A microcomputer accident analysis package for developing countries. Proc of Seminar, 1ndian Road Congress Road Safety Seminar, Siringar. G.