Poster Presentations

Polysomnography (PSG) in the home has advantages over in-laboratory PSG, but one important disadvantage is the inability of current devices to record lights off (Loff) and on times and thus important indices such as sleep onset latency and sleep efficiency cannot be determined. This study evaluates the characteristics of a prototype light sensor (Compumedics) used with a portable PSG device (SomtePSG, Compumedics), and its utility in the home where light conditions are uncontrolled and it is impractical to calibrate the light sensor to the conditions in each individual home. \ud\udMethods: Three examples of the light sensor were exposed to incandescent light at a range of controlled light levels to determine their signal characteristics. Twenty-four home PSGs were analysed to explore the characteristics of the light sensor signal in the home. \ud\udResults: The table below shows the results for the sensor signal characteristics. The light sensor allowed a discernable Loff to be identified in 19 of 24 home PSGs, and in these 19, the mean difference between patient reported and light sensor Loff was 1.2 min (SD 16.6, range -23 to +50). \ud\udDiscussion: The light sensor signals were found to have good sensitivity and linearity, low drift and record a range of lux appropriate for the home setting. When used in home PSG, these light sensors were able to establish Loff in the majority of PSGs. The wide range of differences between patient reported and sensor Loff demonstrates the importance of objective determination of Loff in home PSG, particularly for studies where accurate measurements of Loff dependant indices such as sleep onset latency are needed