Guided and influenced by a famous follow-up study in criminology focused on desistance from crime, this dissertation studies desistance from crime, homelessness, and substance abuse. In the early 1990s, The New Orleans Homeless Substance Abusers Project (NOHSAP) was founded as an experiment funded by the National Institute on Alcohol Abuse and Alcoholism (NIAAA) to uncover optimal treatment strategies for homeless alcoholics and drug addicts. The program ran for three years (1991-1993) and in those years, 670 homeless New Orleans men and women were admitted into treatment. Some of the original clients were followed for as long as 18 months, but none of them had been re-contacted since the mid-1990s. This dissertation involves finding these individuals and re-interviewing them, to discover what life trajectories they have taken some 17-19 years later. Guided by social bonding theory, this project shows what baseline factors and conditions explain variability in life outcomes. The methodology for this study consists of three main parts: 1) a quantitative analysis of mortality data; 2) a historical analysis of criminal histories and 3) in-depth interviews. Nested logistic regression models explained differences among those who have died (n = 91) and those still living. The same method was used to explain differences among those currently incarcerated (n = 56). Follow-up interviews were conducted with 32 individuals in a variety of settings including at their homes and in prisons. Findings from the quantitative results show that social bonding theory seems to be a weak explanation scheme among this population. Results from the qualitative data, however, are contrary and show social bonds to be crucial in the desistance process. Like Laub and Sampson’s study, marriage and employment were strong predictors of desistance. Individuals iv interviewed tended to be sober, but disaffiliated with twelve-step meetings. Other themes from the interviews involve presentation of self, the importance of religion, and a process of aging out of crime. Policy implications from these results focus on the importance of choosing a good life partner, the reduction of alcohol and drug use among abusers, and emphasizing stable employment.