In recent years, wrist actigraphy has been scrutinized as a possible s
upplement to or even replacement for other methods traditionally used
to assess the sleep of clinical patients and research subjects. While
showing promise for some applications, its accuracy in measuring the n
octurnal sleep of insomniacs remains in question. A reexamination of t
he relevant data indicates that with respect to estimating the total s
leep time of insomnia patients, use of the actigraph does not yield si
gnificantly lower error than daily sleep logs and predicts only about
a third of the variance. However, the actigraph does show some potenti
al for assessing the night-to-night variability of a given individual'
s sleep, suggesting that it may be most useful for assessing longitudi
nal changes within a treatment program or experimental protocol.