Objective: While there is general consensus about the need to increase access to mental health treatment, it is debated whether screening is an effective solution. We examined treatment use by inmates in a prison system that offers universal mental health screening. Method: We conducted an observational study of 7,965 consecutive admissions to Canadian prisons. We described patterns of mental health treatment from admission until first release, death, or March, 2015 (median 14-month follow-up). We explored the association between screening results and time of first treatment contact duration of first treatment episode, and total number of treatment episodes. Results: Forty-three percent of inmates received at least some treatment, although this was often of short duration; 8% received treatment for at least half of their incarceration. Screening results were predictive of initiation of treatment and recurrent episodes, with stronger associations among those who did not report a history prior to incarceration. Half of all inmates with a known mental health need prior to incarceration had at least 1 interruption in care, and only 46% of inmates with a diagnosable mental illness received treatment for more than 10% of their incarceration. Conclusion: Screening results were associated with treatment use during incarceration. However, mental health screening may have diverted resources from the already known highest need cases toward newly identified cases who often received brief treatment suggestive of lower needs. Further work is needed to determine the most cost-effective responses to positive screens, or alternatives to screening that increase uptake of services.