A Dyadic Latent Profile Analysis of Older Couples’ Psychological, Relational, and Physical Health

Abstract The present study explored the heterogeneity of older couples’ psychological, relational, and physical health using a sample of 535 couples above the age of 62. A dyadic latent profile analysis was conducted to identify and predict unique clusters of couples’ relative psychological (depressive symptoms and daily hassles), relational (problematic affective communication and marital satisfaction), and physical health (number of health problems and self-reported health satisfaction). Predictors of class membership included relationship length, age, income, and hours worked outside the home. Results revealed 4 distinct classes: Happy & Healthy Together (63.5%), Individually & Relationally Strained (14.7%), Relationally Happy with Strained Wives (12.3%), and Relationally Happy with Strained Husbands (9.3%). Typology descriptions and predictors of class membership will be discussed. These findings highlight that health promotion efforts should be tailored to the specific psychological, relational, and physical health concerns of both partners rather than a one-size-fits-all approach.

Download Full-text

Mindfulness, self-reported health, and cortisol: A latent profile analysis

Journal of Health Psychology ◽

10.1177/1359105320931184 ◽

2020 ◽

pp. 135910532093118

Author(s):

Stephen M Leach ◽

Amanda M Mitchell ◽

Paul Salmon ◽

Sandra E Sephton

Keyword(s):

Physical Health ◽

Undergraduate Students ◽

Latent Profile Analysis ◽

Profile Analysis ◽

Cortisol Response ◽

Future Directions ◽

Reported Health ◽

The Relationship ◽

Mental And Physical Health ◽

Latent Profile

This study utilized a latent profile analysis approach to examine the relationship between mindfulness profiles and self-reported mental and physical health, as well as salivary cortisol levels in a sample of 85 undergraduate students. Consistent with theory, the Judgmentally Observing (high monitoring, low acceptance) reported poorer mental health and exhibited flatter diurnal cortisol slopes than the Unobservant Accepting (low monitoring, high acceptance) and Average Mindfulness profiles. No differences in self-reported physical health, cortisol response to awakening, or diurnal mean cortisol were observed among the profiles. Future directions are discussed.

Download Full-text

Toward a Multidimensional Understanding of Later Life Disability: A Latent Profile Analysis

Innovation in Aging ◽

10.1093/geroni/igab046.778 ◽

2021 ◽

Vol 5 (Supplement_1) ◽

pp. 203-204

Author(s):

Natasha Peterson ◽

Jeongeun Lee ◽

Eva Kahana

Keyword(s):

Older Adults ◽

Latent Profile Analysis ◽

Profile Analysis ◽

Multinomial Logistic Regression ◽

Community Dwelling ◽

Physical Impairment ◽

Sensory Impairment ◽

Class Membership ◽

Participation Restrictions ◽

Latent Profile

Abstract Disability is difficult to define succinctly. Current literature on disability has primarily focused on physical functional limitations. However, relying on a single dimension or index cannot accurately represent disability as the experience of disability is nuanced and complex. To address these gaps, this study aims to understand the multidimensional nature of disability among retired, community-dwelling older adults. Using a sample of 414 older adults between the ages of 72 and 106 years (M=84.84, SD=4.56), latent profile analysis was employed to identify classes based on five indicators of disability across three domains. The five indicators of disability included difficulties with activities of daily living (ADLs), cognitive impairment, physical impairment, sensory impairment, and participation restrictions. Three classes were found to represent the data best. The most favorable and highly functioning group comprised the highest number of participants (n=242, 59.5%). The next group, class 2 (n=157, 37.9%), was characterized by high physical impairment and ADL-difficulty. The smallest group, class 3 (n=15, 3.6%), had the highest ADL-difficulty and participation restrictions but drastically lower cognitive and sensory impairment. Multinomial logistic regression revealed that class membership was related to sociodemographic characteristics. Finally, class membership predicted several mental health outcomes such as depressive symptoms, positive affect, and life satisfaction in the expected direction. If supported by future work, these findings could inform practitioners in developing more specific interventions relevant to older adults based on their disability profiles. Understanding various combinations of disablement has potential implications for services and interventions to be tailored to individuals’ distinct disability-related needs.

Download Full-text

Influence of baseline user characteristics and usage patterns within the first 24 hours on long-term engagement and effectiveness of a web-based weight loss randomized controlled trial: a latent profile analysis (Preprint)

10.2196/preprints.26421 ◽

2020 ◽

Author(s):

Andre Q Andrade ◽

Alline Beleigoli ◽

Maria De Fatima Diniz ◽

Antonio Luiz Ribeiro

Keyword(s):

Weight Loss ◽

Latent Profile Analysis ◽

Profile Analysis ◽

Controlled Trial ◽

User Characteristics ◽

Class Membership ◽

Usage Patterns ◽

The Impact ◽

Latent Profile

BACKGROUND Adherence to online behaviour change interventions is one of the main challenges impacting long-term efficacy. Better understanding of baseline user characteristics can improve design and fit. OBJECTIVE We aim to understand the impact of users’ characteristics and the first 24h usage patterns of a web-platform for weight loss on user engagement and weight loss in the long-term (6 months). METHODS Data from participants of the POEmaS randomised controlled trial, which compared a weight loss platform, platform plus coach and control, were analysed. Data included baseline behaviour and usage logs from initial 24h after platform access. Latent profile analysis (LPA) was used to identify classes and Kruskal-Wallis was used to test whether class membership was associated with long-term (24 weeks) adherence and weight loss. RESULTS Among 828 participants assigned to intervention arms, three classes were identified through LPA: Motivated Healthy (better baseline health habits, high 24h platform use), Indifferent Majority (balanced), Unhealthy Quitters (worse habits and low 24h platform use). Class membership was associated with long-term adherence (p<0.001), and Unhealthy Quitters had the lowest adherence. Weight loss was not associated with class membership (p=0.49), regardless of the intervention arm (platform or platform plus coach). However, Indifferent Majority users assigned to platform plus coach lost more weight than those assigned to platform only (p=0.02). CONCLUSIONS Baseline questionnaires and usage data from the first 24h after login allowed distinguishing classes, which were associated with long term adherence. This suggests that this classification might be a useful guide to improve engagement and select interventions to individual users. CLINICALTRIAL ClinicalTrials.gov NCT03435445; https://clinicaltrials.gov/ct2/show/NCT03435445.

Download Full-text