Current Trends in Decompression Development: Statistics and Data Analysis

DCIEM has been involved in decompression research for over 30 years, and has successfully developed and validated a number of decompression tables for use by the Canadian Forces. These include decompression for air diving, surface-supported helium-oxygen diving, and self-contained semi-closed circuit rebreather diving for mine countermeasures. This article is a review of the traditional deterministic approach to decompression modelling and the newer probabilistic approach. The new approach to decompression modelling has the potential for developing safer and more efficient diving procedures and tables for CF diving applications. Since the development of the first decompression tables in 1906 by J.S. Haldane, considerable research and effort have been expended in the development of safer and more rapid decompression procedures and tables. Most models of decompression that have been used to generate decompression tables have taken a deterministic approach where the boundary between 'safe' and 'unsafe' dives is governed by a fixed set of rules, depending on the gas exchange model and 'safe ascent' criterion that are selected. These models are essentially empirical and not physiological models, providing 'safe' decompression only over a limited range of depth and bottom times. Because decompression illness (DCI) is considered a binary event, it becomes logistically and financially impossible to conduct enough dives to show that a given dive profile is safe within statistical significance. DCI should actually be considered as a probabilistic event. Decompression profiles are not just a case of being 'safe' or "unsafe" but should be considered as a time-depth dosage.