The Assessment of a New Technology for Evaluating Respiratory Abnormalities in Sleep

This study assessed the diagnostic utility of two technologies for evaluating respiratory abnormalities in sleep. We compared conventional polysomnography with a new, ambulatory microprocessor technology. Two key features of the comparison involved: (a) the use of multiple, skilled interpreters of each system; and (b) inter-rater agreement within the systems as a crucial test for the diagnostic utility of each. Results indicated that general categories of respiratory abnormalities could be judged more reliably than more specific categories, regardless of technology. For certain categories of respiratory abnormalities, however, inter-rater agreement with the newer technology was extremely low (e.g., reliability coefficients of .12, .09, and -0.6). Factors contributing to these low diagnostic reliabilities are discussed. Our data indicate that any assessment of new technology cannot be made apart from the clinical judgments to be rendered with that new technology. This approach may be generalizable to the assessment of other diagnostic technologies.