The pathophysiology of SIDS remains unknown. Although a multifactorial cause appears plausible on the basis of available data, new data are needed to determine which components of this multifactorial hypothesis are most important and whether other factors need to be added. We need to better understand control of breathing in the newborn infant and the manner in which maturation of cardiorespiratory control progresses during infancy. The unique period of vulnerability for SIDS, in which risk is less in the neonate than at 2 to 6 months of age, remains unexplained. Is there a worsening in some aspect of cardiorespiratory control in infants destined to die of SIDS? An improved understanding of the increased risk in black infants, preterm infants, and infants with intrauterine drug exposure, although only a small percentage of all SIDS deaths, should contribute substantially to our understanding. Appropriately designed and well-controlled prospective studies are needed in asymptomatic infants at risk, to determine the true contemporary nonintervention rate of SIDS and the extent to which any assessment or intervention lowers this rate. Prospective pneumogram screening studies have demonstrated significant group differences in respiratory patterns in normal infants compared with later SIDS victims, but have failed to achieve sufficient sensitivity and specificity to be useful for populationwide prospective screening. To assess aspects of brainstem cardiorespiratory control in addition to those assessed by a conventional pneumogram, future studies will need to be based on an expanded or modified technology. On the basis of both physiologic considerations and available technology, addition of an oxygen saturation channel offers the most promise for providing a more comprehensive assessment of cardiorespiratory control. If there is an underlying deficiency in asphyxic arousal responsiveness, for example, with or without other respiratory control deficits, continuous monitoring of oxygen saturation as part of a second-generation pneumogram system currently has the greatest promise for providing a modified pneumogram assessment of greater clinical use. The use of continuous oxygen saturation as a home monitoring technique should also be investigated. Power spectrum analysis of cardiorespiratory variability also appears to have potential advantages over conventional pneumogram analyses, and needs to be evaluated in prospective studies. The following statements summarize our current knowledge regarding SIDS, apnea, pneumograms, and home monitors: The cause(s) of SIDS remains unknown.(ABSTRACT TRUNCATED AT 400 WORDS)