SLEEP IN HEALTHY AGED PERSONS AND INHABITANTS OF A RESIDENTIAL HOME

It was the aim of this study to estimate the impact of exogenous influ ences on the sleep of elderly. To this end, ambulatory sleep recording s were made in 2 groups (7 Ss each) of elderly females: inhabitants of a residential home (mean age 85.9 yrs) and carefully selected healthy controls (mean age 91.7 yrs) living independently. A recording consis ted of a 48-h registration of the EEG, the EOG and EMG. The second 24- h period of each recording was analyzed: - according to conventional p olysomnographic criteria; and - by way of a FFT routine which calculat ed one averaged spectrum for every 4 minutes EEG. Between-group compar isons of the polysomnographically verified sleep times revealed 1. a p redominance of napping among the residents; 2. an earlier phase positi on of the main sleep of the residents, which was partly due to a longe r sleep latency for the controls; and 3. a relatively advanced REM dis tribution for the controls. The outcome of the spectral analyses of th e sleep EEG showed a clear dominance of the theta band for both groups . Moreover, detailed analysis gave evidence of a systematic variation of the spectra throughout the sleep period. Whereas the delta band is relatively best represented during the first half of the sleep period, the theta band is maximal during the second half. The spectra for the naps appeared to be most similar to the mean spectrum for the 4th 90 min period of main sleep. Finally, the accumulation of EEG energy in t he delta and theta bands over the first 7 hours of main sleep was high ly similar for the two groups. It is concluded that the observed group differences in the timing of main sleep, in the occurrence of napping , and in the phase of the REM distribution originate from differences in living conditions. Differences in the exposure to daylight may have played a major role in this. The sleep process itself, as manifested in the accumulation of EEG energy, appeared invariant under these diff erences.