INNOVATIVE METHODS TO CHARACTERIZE BEHAVIORAL SYMPTOMS OF PERSONS WITH DEMENTIA

2.50
Hdl Handle:
http://hdl.handle.net/10755/211567
Type:
Research Study
Title:
INNOVATIVE METHODS TO CHARACTERIZE BEHAVIORAL SYMPTOMS OF PERSONS WITH DEMENTIA
Abstract:
Purpose: This study purpose was to characterize complex patterns of BSD, and examine the association among different BSD patterns and cortisol profiles. This paper illustrates the application of two analytic approaches for examining patterns of behaviors over time: a) a non-parametric approach using THEME ™ a software that describes several dimensions related to clustering of behaviors, and b) a mixture model approach (group-based trajectory modeling) that identifies clusters of individuals following similar trajectories and characterizes each of these in terms of polynomial parameters. We illustrate the application of these approaches using BSD and cortisol profiles. Background: Commonly used methods for repeated measures typically estimate a pattern based on an average (magnitude or frequency) across participants for each observation point. While an important perspective, this approach may not fully characterize many dimensions of the patterning of behavioral symptoms of dementia (BSD) including within person variability and irregularity or clustering of behaviors. The limited ability to accurately detect and quantify complex BSD patterns has hampered the development and refinement of tailored interventions. Methods: Data from two studies reported elsewhere were pooled into a single data set for analysis (N = 99). From the pooled sample a secondary analysis was completed using data from two groups of nursing home (NH) residents with complete cortisol profiles (N = 28, N = 27) using THEME ™. BSD and cortisol data were collected for four consecutive days. Participants were categorized into two groups within each site: normal cortisol profiles (G1) and abnormal cortisol profiles (G2). Random effects model was used to characterize BSD trajectories over time. THEME™ was used to identify within person behavior patterns and complexity. Results: There was a significant difference in total BSD between those with a normal cortisol profile (G1) compared to those with an abnormal cortisol profile (G 2) at Time 3 (12:00PM–1:59PM) within each site  (p = 0.0456; p = 0.023) respectively. There was a significant difference in age, co-morbidities and anti-anxiety medication when G1 was compared with G2. THEME™ analysis showed a significant difference in BSD patterns (t = 1.94, p = 0.05) and complexity (t = 2.53, p = .014) when G1 was compared to G2. G1 exhibited more complex patterns of high intensity vocalization and restlessness than G2. Of those who exhibited 50-75% high intensity patterns, 80% in Site 1 had an abnormal cortisol rhythm, compared to 30% of those in Site 2. Group based trajectory methods showed no significantly different trajectories between groups. Conclusions: Higher complexity and patterns differed significantly between G1 and G2. While preliminary, results suggest that persons with higher variability of BSD exhibit complex patterns, in addition to deregulated circadian rhythms indexed by abnormal cortisol rhythms. These findings highlight the importance of using pattern analyses to identify complex behavioral manifestations and correspondence with biological markers. THEME ™ analysis was able to identify and characterize individual BSD patterns that drove the overall trajectory of behavior. Pattern analysis merits further examination to investigate the temporal clusters of BSD.
Keywords:
Dementia; Behavior patterns
Repository Posting Date:
20-Feb-2012
Date of Publication:
20-Feb-2012
Other Identifiers:
5495
Sponsors:
Western Institute of Nursing

Full metadata record

DC FieldValue Language
dc.typeResearch Studyen_GB
dc.titleINNOVATIVE METHODS TO CHARACTERIZE BEHAVIORAL SYMPTOMS OF PERSONS WITH DEMENTIAen_GB
dc.identifier.urihttp://hdl.handle.net/10755/211567-
dc.description.abstractPurpose: This study purpose was to characterize complex patterns of BSD, and examine the association among different BSD patterns and cortisol profiles. This paper illustrates the application of two analytic approaches for examining patterns of behaviors over time: a) a non-parametric approach using THEME ™ a software that describes several dimensions related to clustering of behaviors, and b) a mixture model approach (group-based trajectory modeling) that identifies clusters of individuals following similar trajectories and characterizes each of these in terms of polynomial parameters. We illustrate the application of these approaches using BSD and cortisol profiles. Background: Commonly used methods for repeated measures typically estimate a pattern based on an average (magnitude or frequency) across participants for each observation point. While an important perspective, this approach may not fully characterize many dimensions of the patterning of behavioral symptoms of dementia (BSD) including within person variability and irregularity or clustering of behaviors. The limited ability to accurately detect and quantify complex BSD patterns has hampered the development and refinement of tailored interventions. Methods: Data from two studies reported elsewhere were pooled into a single data set for analysis (N = 99). From the pooled sample a secondary analysis was completed using data from two groups of nursing home (NH) residents with complete cortisol profiles (N = 28, N = 27) using THEME ™. BSD and cortisol data were collected for four consecutive days. Participants were categorized into two groups within each site: normal cortisol profiles (G1) and abnormal cortisol profiles (G2). Random effects model was used to characterize BSD trajectories over time. THEME™ was used to identify within person behavior patterns and complexity. Results: There was a significant difference in total BSD between those with a normal cortisol profile (G1) compared to those with an abnormal cortisol profile (G 2) at Time 3 (12:00PM–1:59PM) within each site  (p = 0.0456; p = 0.023) respectively. There was a significant difference in age, co-morbidities and anti-anxiety medication when G1 was compared with G2. THEME™ analysis showed a significant difference in BSD patterns (t = 1.94, p = 0.05) and complexity (t = 2.53, p = .014) when G1 was compared to G2. G1 exhibited more complex patterns of high intensity vocalization and restlessness than G2. Of those who exhibited 50-75% high intensity patterns, 80% in Site 1 had an abnormal cortisol rhythm, compared to 30% of those in Site 2. Group based trajectory methods showed no significantly different trajectories between groups. Conclusions: Higher complexity and patterns differed significantly between G1 and G2. While preliminary, results suggest that persons with higher variability of BSD exhibit complex patterns, in addition to deregulated circadian rhythms indexed by abnormal cortisol rhythms. These findings highlight the importance of using pattern analyses to identify complex behavioral manifestations and correspondence with biological markers. THEME ™ analysis was able to identify and characterize individual BSD patterns that drove the overall trajectory of behavior. Pattern analysis merits further examination to investigate the temporal clusters of BSD.en_GB
dc.subjectDementiaen_GB
dc.subjectBehavior patternsen_GB
dc.date.available2012-02-20T12:02:43Z-
dc.date.issued2012-02-20T12:02:43Z-
dc.date.accessioned2012-02-20T12:02:43Z-
dc.description.sponsorshipWestern Institute of Nursingen_GB
All Items in this repository are protected by copyright, with all rights reserved, unless otherwise indicated.