Sleep in community-dwelling older adults: issues of measurement

Suksatit, Benjamas

Download PDF

Request Version for Screen Reader

Last Modified

March 21, 2019

Creator

Suksatit, Benjamas
- Affiliation: School of Nursing

Abstract

A secondary data analysis was utilized to explore the first night effect (FNE) among community-dwelling older adults, age 70 years and older. The accuracy of actigraphy when compared with polysomnography (PSG) in measuring sleep and change in sleep in this sample was also explored. The data were derived from 63 community-dwelling older adults from two studies: PRISM and PTRACS. Two instruments were used, including 1) polysomnograph and 2) wrist actigraph. According to standard PSG, two sleep experts scored sleep states and the inter-rater agreement was acceptable across all records. For wrist actigraph, an actiwatch-light (AW-L) was employed in two original studies. Sleep data from AW-L were retrieved using Actiware-Sleep software v.3.3. This study was conducted under approval from the Institutional Review Board of the University of North Carolina at Chapel Hill. Data entry and analysis were performed using SAS software, version 9.3. All data were double entry and compared for any errors. FNE occurred in community-dwelling adults age 70 years and older, including more wake after sleep onset (WASO), more stage N1, more REM latency, less total sleep time (TST), less sleep efficiency (SE), and less stage N3. According to the accuracy of actigraphy against PSG, of four sensitivity settings, the high sensitivity setting of actigraphy provided fewer discrepancies. Using high sensitivity settings, actigraphy underestimated sleep onset latency (SOL) by 2 minutes, underestimated WASO by 21 minutes, and overestimated TST by 21 minutes as compared to PSG. For the accuracy of actigraphy when compared to PSG in measuring change in sleep, high sensitivity setting provided fewer discrepancies. Actigraphy with high sensitivity setting overestimated change in SOL by 21 minutes, overestimated change in WASO by 12 minutes, and underestimated change in TST by 41 minutes. Although PSG is the gold standard in measuring sleep, it has many disadvantages. FNE is one major issue. When FNE occurs, the second night of PSG is needed in order to capture subject's habitual sleep. Although actigraphy is a cost-alternative method in measuring sleep, it appears to underestimate wake and overestimate sleep in community-dwelling older adults, age 70 years and older when compared with PSG.

Date of publication