For almost 50 years, attempts have been made to account for the pronounced solvent effect on the lifetime of singlet molecular oxygen, O-2(a(1)Delta(g)). This process is dominated by the O-2(a(1)Delta(g)) -> O-2(X-3 Sigma(-)(g)) nonradiative transition. Given the comparatively low O-2(a(1)Delta(g)) excitation energy of similar to 7880 cm(-1), existing models have been built upon a foundation of electronic-to-vibrational (e-to-v) energy transfer in which C-H and O-H stretching modes in the solvent act as the dominant energy sink. The latter accounts for large H/D solvent isotope effects on the O-2(a(1)Delta(g)) lifetime. However, recent experiments showing a pronounced temperature effect on the O-2(a(1)Delta(g)) lifetime in some solvents reveal limitations in these models. We have developed a general and computationally tenable model that accounts for both temperature and H/D solvent isotope effects on the O-2(a(1)Delta(g)) lifetime. A key feature of our approach is the need to strike a balance in the oxygen-solvent interaction between weak and strong coupling. In the weak coupling limit, the O-2(a(1)Delta(g)) -> O-2(X-3 Sigma(-)(g)) transition probability is determined by the overlap of vibrational wave functions, and this is the main component defining the H/D isotope effects. In the strong coupling limit, the transition probability is determined by an activated process and thus accounts for the observed temperature dependence. In addition to resolving a long-standing oxygen-dependent problem, our model may provide useful insights into a wide range of bimolecular interactions that involve e-to-v energy transfer.