Saturday, April 2, 2005
Mayflower I & II (Hilton Cincinnati Netherland Plaza)
Session: 1190, Life Span, 1:00 PM

A New Approach to Synthesis of Lifespan Research

Judith Floyd, PhD, RN, FAAN, Professor1, James Janisse, PhD, Assistant Professor2, Elizabeth Jenuwine, PhD, MLIS, Research Assistant1, and Joel Ager, PhD, Biostatistician2. (1) College of Nursing, Wayne State University, 5557 Cass Avenue, Detroit, MI 48202, (2) Center for Healthcare Effectiveness Research, Wayne State University, Shiffman Medical Library, Detroit, MI 48202

Dependable knowledge about the magnitude of sleep change with age became increasingly more important over the past decade as sleep clinicians including nurses began to question the reliability of previously accepted norms for adult sleep. Purpose. The purpose of this study was to describe linear and non-linear changes in sleep over the adult lifespan. Theoretical Framework. A mid-range theory of lifespan development underpinned this research. Sample. 587 studies of sleep and age were included in the research synthesis. Intercoder reliability exceeded the cutoff for excellent (Kappa=.75). Methods. A new research synthesis method developed by the investigators was used to estimate both linear and non-linear components of aging-related sleep change over the adult lifespan. The method consisted of generating scatterplots of hypothetical data points based on the means and standard deviations reported for groups of similarly-aged healthy subjects. Once scatterplots were generated, Cubic B smoothing splines were fitted to the data. They were inspected to determine the kind of function that best fit the data. Results. Functions could be identified as linear, quadratic, or linear-quadratic for 14 sleep variables. Slopes were used to describe age-related change in sleep variables with linear functions, minimum points were identified for quadratic functions, and ages at which the rate of change in sleep accelerated were identified using segmented regression for linear-quadratic functions. Differences in sleep patterns based on sex of subjects could be shown for only 4 sleep variables because most commonly studied sleep variables remain under-studied in women. Conclusions. This new approach to research synthesis of aging-related change was effective for describing sleep change over the adult lifespan and for identifying gaps in the sleep and aging research literature. Funding. This study was funded by the National Institute of Nursing Research, grant R01 NR003880, “Aging-Related Sleep Changes: A Meta-Analysis,” 2000-2004.

Session #1190 - Life Span

The 29th Annual MNRS Research Conference (April 1-4, 2005)