RESUMO
Our primary objective during this pilot work has been to determine if the computerized RAB can be used in place of the paper-and-pencil version with an IDU population. As stated, reliability estimates in the form of correlations of risk scores between the two versions ranged from .82 to .94, and percentages of exact agreement of the scoreable items ranged from 88% to 100%. These correlations indicate that the C-RAB is comparable to the RAB in its psychometric properties. Further, subjects' positive ratings of the computer program and perceptions that it would better protect confidentiality indicate that this population would participate in assessments involving computer interactions. While these self-reports may be biased on clients' desire to be cooperative and the novelty of the assessment approach, the high level of approval ratings suggest that a majority of participants do prefer completing assessments on the computer. Similarly, participants' demonstrated ability to use the computer and their rating of the program as easy to use suggest that the C-RAB has potential to be a reliable and efficient tool in assessing risk behaviors among IDUs enrolled in HIV vaccine trials.
