The evaluation of fitness to drive on medical grounds raises important issues both for the individual and for society in general. At a time when the road accident rate still compares unfavourably with other major forms of transport, there may be a general reluctance to relax the existing criteria, together, possibly, with pressure for even more stringent criteria to be applied. An example is the EU directive of 1998 which restricted further the range of vehicles that those with insulin-treated diabetes were entitled to drive. However, the desire for improved levels of safety needs to be balanced by the rights of the individual in a society where the inability to drive may bring severe disadvantages even for the non-vocational driver. These considerations have led to the conclusion that the decision-making process relating to medical licensing should be based on a risk analysis approach, and the consequences of this were discussed in a recent report. There are two stages in the application of risk analysis to this area. The first is concerned with the estimation of risk based on the available information, while the second is concerned with the management of risk and the definition of the level of risk that can be tolerated, which is essentially a political decision. This report is concerned with the first stage, and specifically with the sensitivity of the process to variations in the parameters used in the estimation of risk. The previous report drew attention to various limitations in the risk analysis approach, including the lack of credible evidence, at least in some areas, on which an assessment could be based. This raises the issue of whether such an approach can be justified when the information on which is founded may be unreliable. The present report, therefore, investigates the sensitivity of the earlier conclusions to the assumptions that were made. The objective is to establish the extent to which risk analysis can be usefully applied, and to identify those parameters that are most critical and where additional information would be of greatest value. (Author/publisher)
Abstract