Chronic insomnia and bed partner actigraphy data

The files contain seven nights of continuous actigraphy measurements of 40 subjects with chronic insomnia and their 40 bed partners. Wrist actigraphy was used, collected for one week using Respironics Actiwatch Spectrum Pro and Actiware software (Respironics, Bend, OR, USA), with movement counts were sampled in 60-second epochs. All recruited subjects wore the devices at all times during day and night. All subjects were free to move and were not prohibited from doing any activities of daily living. The actigraphy data files are first processed to extract features followed by machine learning model, which used these features to classify subjects in chronic insomnia subjects and normal sleepers.

Methods
The data were collected using Respironics Actiwatch Spectrum Pro and Actiware software (Respironics, Bend, OR, USA), with movement counts were sampled in 60-second epochs. The actigraphy data were recorded from cohabiting couples with one subject seeking treatment for their insomnia. Subjects were recruited through online advertisements, radio, print media, and via referrals from a psychologist, sleep clinics or general practitioner (GP) in Melbourne, Australia. The subjects were recruited as part of a larger ongoing clinical trial (Project REST; Australian New Zealand Clinical Trials Registry Registration: ACTRN12616000586415), to study partner associated behavioural interventions in insomnia. The study was approved by the human research ethics committee of Monash University, approval number CF16/276 - 2016000125.