Objective: The aim is to develop a reliable and valid questionnaire for adult offenders in order to assess risk and needs of them, including mental health and psychosocial risk factors. Methods: The research was carried out in 10 prisons with a total of 1087 offenders. In the study 74.2% of the offenders are male. Based on the literature review and focus group discussions socioeconomic status, alcohol, substance abuse, mental health problems, risk of harm and personality traits are identified as five risk areas. All of the questions for each subscale were created and the questionnaire is named as ARDEF. Different interviewers have done second interviews with the 173 offenders. SCL-90, Levenson Psychopathy Scale, Borderline Personality Inventory, Novelty Seeking Scale, the AUDIT and the CAGE scale was applied to the offenders. Results: Cronbach's alpha coefficient of the subscales are between 0.71 and 0.89. Subscale and total scores were obtained from the only factor in the factor analysis, and 57.62% of total variance was explained. Inter-rater correlations of the subscales ranged from 0.88 to 0.70. Correlation was observed between the scores of screening tools; harm subscale score and the SCL-90 anger subscale scores, the risk of mental health problems subscale and SCL-90 total score, risky personality traits subscale score and Levenson Psychopathy Scale total score and Borderline Personality Inventory total score, the risk of alcohol and substance use subscale score and CAGE and the AUDIT. Conclusion: Results of this study show that, ARDEF was found to be a reliable and valid measure of risk of adults who have committed crimes. However, it should be noted that this scale should be used as a complementary tool to assess for the benefit of prisoners.