Test-retest reliability of a complex human immunodeficiency virus research questionnaire administered by an audio computer-assisted self-interviewing system

OBJECTIVES. To evaluate the test-retest reliability of a complex questionnaire administered by Audio Computer-assisted Self-interviewing to recently diagnosed human immunodeficiency virus-positive patients. METHODS. Thirty-seven English-speaking and 32 Spanish-speaking participants completed both test and retest interviews. Pearson correlation coefficients (r) and kappa (κ) and weighted kappa (κ) statistics were obtained for individual questions. From these, overall κ and Pearson correlation coefficients were calculated across all variables and for groups of questions. RESULTS. Overall measures of reliability were κ = 0.767, r = 0.728. Some variation in reliability existed for different response formats, question content groups, and languages of the participants. Differences in overall reliability by Spanish compared with English participants were small and not statistically significant. CONCLUSIONS. Audio Computer-assisted Self-interviewing provides reliable measures for items assessed in the Antiretroviral Treatment and Access Study baseline questionnaire. Some differences exist as a result of question content, interview language, and response format, requiring assessment in future studies and consideration in designing Audio Computer-assisted Self-interviewing systems and questionnaires.