Case Study of Decision Making: U.S. Navy Divers' Decompression Procedures. S. S. Survanshi and E. D. Thalmann, Naval Medical Research Institute, 8901 Wisconsin Ave., Bethesda MD 20889-5607; and P. K. Weathersby, Gales Ferry, CT 06335
Military occupational safety assessment of proposed new procedures involves fewer parties than those involving health of the general population. However, many of the same issues arise in both risk assessment and risk management. In 1993, a mathematical model was developed to estimate the risk of decompression sickness following dives using only compressed air and to compute air decompression procedures at a specified risk level. These air decompression procedures can get impractically long for some dives. Breathing oxygen during decompression is known to substantially reduce the required decompression time without additional risk. The 1993 mathematical model could not be used to compute decompression procedures using oxygen, since it was shown to underestimate the risk of such dives. Nevertheless, the U.S. Navy greatly desired an oxygen decompression procedure. Resources for direct experimentation and additional modeling were not available. Several candidate procedures with oxygen decompression time calculated as a fraction of air decompression time were compared to the currently used (40 years old) method. Risk management decisions were made by naval officers with the appropriate technical and occupational backgrounds. The final oxygen decompression procedure reduced decompression time by approximately two-thirds when compared to using air. There were three major elements in accepting the new procedure: 1) it was perceived as safer than the existing method, which was confirmed by a direct experimental dive trial conducted at a later time; 2) it was by design more comprehensive in scope and simpler to administer than the prior method, thus conferring an unquantified benefit; and 3) it was assessed by modeling as having comparable or greater safety over the air decompression, which would otherwise be used.
Work supported by NMRDC Work Unit #63713N M0099.01A-1002.