New Data Analysis Methods for Actigraphy in Sleep Medicine (MASM)
Clinical Trials URL:
Study Type: Epidemiology Study
Prepared on November 30, 2012
Last Updated on November 30, 2012
Study Dates: July 2009 - November 2011
Consent: Unrestricted Consent
Commercial Use Restrictions: No
NHLBI Division: DLD
Collection Type: Open BioLINCC Study - See bottom of this webpage for request information
Define an object oriented data model for large activity datasets and patient level data, and apply existing and new advanced statistical and visualization methods for activity data.
An Actical™ is a watch-like device attached to the wrist that uses an accelerometer to measure movement nearly continuously over several days. Actigraphy data is recorded densely, such as every minute or every 15 seconds. The general approach to analysis of this activity data is to reduce the time series of measurements to summary statistics such as sleep/wake ratios, sleep time, wake after sleep onset, and ratios of nighttime activity to daytime or total activity. While summary measures allow for hypothesis testing using classic statistical methods, large amounts of information are lost and problems may arise from masking the circadian patterns. Novel statistical methods are needed to analyze this complex data in a more comprehensive manner.
585 non-pregnant adult patients recruited from the clinic at the Washington University in St. Louis Sleep Medicine Center. Clinic patients with a suspected diagnosis of obstructive sleep apnea (OSA), insomnia, or restless legs syndrome (RLS) were invited to participate. Clinic patients working an evening or overnight shift were excluded. Participants wore the actical device on the non-dominant wrist for a period of 7 days. Activity measurements were collected on 420 participants with a repeated series of measurements on 76 participants 2 months to 1 year later.
Publicly available functional data analysis tools utilizing the r package were developed to analyze the activity data. Early analyses have shown that patients with high apnea have statistically lower activity patterns during the day, and BMI did not impact the circadian patterns. The use of functional data analysis has the potential to reposition analysis of actigraphy data from general sleep assessment to circadian activity rhythms. (Journal of Circadian Rhythms, 2011 9:11)