A probabilistic model was used to predict decompression sickness (DCS) outcome in pig (70 and 20 kg), hamster (100 g), rat (220 g) and mouse (20 g) following air saturation dives. The data set included 179 pig, 200 hamster, 360 rat, and 224 mouse exposures to saturation pressures ranging from 1.9–15.2 ATA and with varying decompression rates (0.9–156 ATA • min−1). Single exponential kinetics described the tissue partial pressures (Ptiss) of N2: Ptiss = ∫(Pamb – Ptiss) • τ−1 dt, where Pamb is ambient N2 pressure and τ is a time constant. The probability of DCS [P(DCS)] was predicted from the risk function: P(DCS) = 1−e−r, where r = ∫(PtissN2 − Thr − Pamb) • Pamb–1 dt, and Thr is a threshold parameter. An equation that scaled τ with body mass included a constant (c) and an allometric scaling parameter (n), and the best model included n, Thr, and two c. The final model provided accurate predictions for 58 out of 61 dive profiles for pig, hamster, rat, and mouse. Thus, body mass helped improve the prediction of DCS risk in four mammalian species over a body mass range covering 3 orders of magnitude.