Statistically Based Decompression Tables. I. Analysis of Standard Air Dives: 1950-1970.

Abstract

A large number of air dives was examined by a new methodology. Several essentially empirical models of decompression risk were considered that predicted the probability of decompression sickness (DCS) for a given pressure exposure to avoid the indefinitely sharp threshold parameters that have characterized previous calculation of decompression tables. The candidate models used several distinct formulations of tissue gas exchange kinetics and summed tissue overpressures that are calculated during the dive to estimate decompression risk. The models were compared to decompression outcome data using the statistical principle of maximum likelihood. Reported decompression trials from American, British, and Canadian Naval laboratories were examined individually and collectively to evaluate the probabilistic models and their parameters. Only two to five parameters were found to be justified by the available data (more than 1,700 individual exposures were considered). Diving data from various sources were only partially compatible; some of the discrepancy may arise from an evolution of diagnostic criteria over several decades. Predictions were made of the outcome for additional reported diving series, and they were only partly successful. The models were then used to estimate decompression risk for current USN air diving with a finding of a wide range of hazard. Specifically, it appears that short dives are quite safe, even to a moderately deep depth, while long exposures are very risky regardless of depth. These findings will be used to produce a set of standard air tables with a uniform and low level of DCS risk. (Author)