Introduction Currently advice about an OSAS patient's fitness to drive is based upon the severity of the condition, with or without objective measure of daytime sleepiness and their account of their driving. Although there is a trend towards increased likelihood of accidents with more severe OSAS, this is not sufficiently robust data. There are conflicting data about the relationship between perceived sleepiness and the likelihood of being involved in an accident. Recently we have established that it is possible to identify with high degree of certainty a group of OSAS sufferers who perform significantly worse than others using specific simulator parameters on an advanced office based driving simulator (miniSim). We now explore the impact of patients' perception of problem driving, demographic, clinical, and polysomnographic characteristics on the outcomes of the simulator test.
Methods 133 (52±10 yrs, ESS 12±5, AHI 29±21) patients completed a detailed driving related questionnaire and performed a 90 km motorway driving scenario on the miniSim. Two events were programmed to trigger evasive actions, one subtle (Veer event) where an alert driver should not crash, while with the other (Brake event) even a fully alert driver might crash. There were three possible outcomes of the simulator runs; “fail”, “indeterminate” and “pass”. The questionnaire responses, demographic, clinical and polysomnographic characteristics were compared between the three outcome categories using one way ANOVA. Logistic regression was performed to explore whether a “fail” could be predicted from any of these data.
Results The results of one way ANOVA are described in Abstract P20 table 1. Patients who fail the simulator test tend to report more sleepiness while driving with a higher ESS & ODI. They also have more, but statistically insignificant, near misses and history of accidents. None of this information could predict a “fail” accurately in the logistic regression analysis.
Conclusions These data confirm that patients' accounts and perception of their own driving and the severity of their OSAS may not be reliable predictors of safe driving. Whether poor performance on an advanced driving simulator is predictive of poor on road performance needs to be established.