Accidental drug overdose continues to be a substantial cause of mortality for drug users. Characteristics of the neighborhood built environment may be important determinants of the likelihood of drug overdose mortality independent of individual-level factors. Using data from the New York City Office of the Chief Medical Examiner, we conducted a multilevel case control study using data on accidental overdose deaths as cases and non-overdose accidental deaths as controls. We used archival data from the New York City Housing and Vacancy Survey and the Mayor's Office of Operations to assess characteristics of neighborhood external (e.g. dilapidation of buildings) and internal (e.g. quality of utilities in houses) built environment. Multilevel analyses were used to assess the relations between the neighborhood built environment and the likelihood of overdose death. Six out of the eight characteristics of the external environment studied and three out of the six characteristics of the internal environment studied were significantly associated with the likelihood of fatal drug overdose in multilevel models after adjusting for individual-level (age, race, sex) and neighborhood-level (income, drug use) variables. Deterioration of the built environment, particularly the external environment, is associated with an increased likelihood of fatal accidental drug overdose. Disinvestment in social resources, psychosocial stressors, neighborhood differences in response to a witnessed overdose, and differences in vulnerability to the adverse consequences of drug use in different neighborhoods may explain the observed associations.