Understanding of informed consent is often limited in prison, due to unique challenges faced by detained people. This study will test a modified "teach-to-goal" informed consent process compared to standard procedures. Conducted within a larger study on psychiatric outcomes, it aims to enhance ethical practices for vulnerable populations.

Background and rationale. Informed consent is an ethical pillar of health-related research involving human subjects, but participants' understanding of informed consent is often limited. Informed consent is also a critical issue in prison research.

First, detained persons are deprived from liberty, a context that may undermine voluntary participation and autonomy. Second, detained persons have disproportionally low levels of education, literacy, and high rates of cognitive or substance- and psychiatric-related impairments. In Switzerland, a study conducted in the general population concluded that informed consent forms are too long and complex. Another study conducted in prison showed a full understanding of 5% in detained persons. Of note, a revision of the Federal Human Research Act (HRA) has just begun and includes propositions to strengthen informed consent. Nonetheless, few studies focused on understanding of informed consent in prison research and to our knowledge, no study with a robust methodology tested whether the most effective interventions in the general population are also effective in prison. Objective of the study. The overarching aim of this study is to estimate and improve understanding of informed consent in prison research, and better understanding characteristics associated with a low understanding. The informed consent process is a time-consuming process and studies may lack resources to ensure that participants provide fully informed consent. To fill in this critical research gap, this study will be embedded in a larger randomized controlled trial (RCT) ("parent-RCT") designed to improve psychiatric outcomes. We will use an RCT to compare two informed consent processes. Methods. This is a cross-sectional RCT (ratio 1:1) testing the effect of informed consent processes on understanding (primary endpoint) and evaluation (secondary endpoints) of informed consent. The research team and will not be blinded to the participants' group, but the statistician will be. Participants will be partially blinded. Data will be collected in two prisons among sentenced men (n=100). The intervention will be modified "teach-to-goal" vs. standard informed consent process. In the modified teach-to-goal intervention, the participant is asked to describe the research procedures. Misconceptions are corrected by a study team member and the participant is asked to describe again the corresponding components. The procedure is iterated until correct and full understanding is reached. Analyses will include intention-to-treat bivariable and multivariable regressions. We will also explore associations between understanding and socio-demographic variables using logistic, linear or negative-binomial regressions, to identify characteristics associated with lower understanding. Expected results and impact. There is a growing recognition that prison research is needed. Improving informed consent in prison research therefore constitutes a critical but neglected issue, both as a separate research topic and for future research projects focusing on detained persons and more broadly, on vulnerable populations. Embedding a RCT on informed consent in an already existing RCT is an unconventional research process. We believe that it will provide "real (study) life" information on how informed consent is informed and ways to improve it. It will also ensure that ethical requirements are fully met in the parent-RCT. This study will thus provide scientific evidence on how to improve informed consent in prison research.

Results will be valuable for other vulnerable populations, including people with low levels of education and literacy and high burden of disease. We plan to publish results in a high-impact biomedical journal and to share them in international conferences and with Swiss cantonal Ethics Committees to raise awareness on informed consent as a core component of medical ethics in research, and more specifically, research involving vulnerable populations. The project may provide meaningful information for the revision of the HRA and help improving informed consent processes in Swiss research.

Inclusion Criteria:

• inclusion in the parent-RCT
• age ≥ 18 years
• good command of French
• absence of acute psychiatric disorder
• providing written informed consent for study participation in the parent-RCT

Exclusion Criteria:

- none.