Objective: The objective of this study was to evaluate predictors of unplanned readmission to a specialized hospital addiction unit within less than 30 days, between 30 and 60 days and over 60 days post-discharge among individuals with a diagnosis of substance use disorder.

Methods: Cox proportional hazards regressions were used to test the effects of potential risk factors on time-to-onset for unplanned readmissions. The outcome (survival time) was the length of time to hospital readmission and the predictors were age, sex, duration of the first hospital stay, Health of Nation Outcome Scales score and Brief Symptom Check List.

Results: Of the 750 readmissions analyzed for the reported period 28.0% took place in less than 30 days, 12.0% between 30 and 60 days and 60.0% after 60 days of discharge. Length of the first hospitalization was a statistically significant predictor of readmission between 30 and 60 days and over 60 days but not for less than 30 days. A 10% increase in length of the first hospitalization, holding all other variables constant, was associated with a 5.0% decrease in unplanned readmissions occurring between 30 and 60 days and a 2.2% decrease in readmissions over 60 days post-discharge.

Conclusion: Length of the first hospitalization was found to be a protective factor of readmission between 30 and 60 days and over 60 days but not for less than 30 days post-discharge. The longer the duration of the first hospitalization, the less quickly patients were readmitted to hospital.