Abstract
Background
This study investigated the role of a large range psychological, attitudinal and health related variables as predictors of depression trajectories amongst older adults over a 4-year time period.
Methods
Data from three consecutive waves of the TILDA survey of older community dwelling adults aged 50+ in Ireland were combined for analysis. Depression symptom scores were assessed using the Center for Epidemiological Studies– Depression scale (CES-D). Changes in depression scores over three time points were modelled as distinct trajectory classes using group-based trajectory modelling, whilst simultaneously controlling for demographic, attitudinal and health related predictors of these trajectory classes using multinomial regression.
Results
Four distinct depression trajectories were identified as (1) a stable low symptom level group (79%), (2) a moderate but deteriorating symptoms group (7.6%), (3) a moderate but improving group (10.1%) and (4) a vulnerable group with consistently high symptoms (3.1%). Multinomial logistic regression indicated that limiting pain, mobility impairments, perceived stress and loneliness predicted membership of the moderate and higher depressive symptom classes. Retirement status and higher reported levels of worry were associated with a greater likelihood of membership of the moderate symptom classes only. Limitations Use of the CES-D is open to bias due to subjective nature of respondent reporting.
Conclusions
Results concur with previous studies on the development of depression among older people and highlight the key health related and psychological variables that may inform interventions aimed at mitigating risks of developing depression among older adults.
This study investigated the role of a large range psychological, attitudinal and health related variables as predictors of depression trajectories amongst older adults over a 4-year time period.
Methods
Data from three consecutive waves of the TILDA survey of older community dwelling adults aged 50+ in Ireland were combined for analysis. Depression symptom scores were assessed using the Center for Epidemiological Studies– Depression scale (CES-D). Changes in depression scores over three time points were modelled as distinct trajectory classes using group-based trajectory modelling, whilst simultaneously controlling for demographic, attitudinal and health related predictors of these trajectory classes using multinomial regression.
Results
Four distinct depression trajectories were identified as (1) a stable low symptom level group (79%), (2) a moderate but deteriorating symptoms group (7.6%), (3) a moderate but improving group (10.1%) and (4) a vulnerable group with consistently high symptoms (3.1%). Multinomial logistic regression indicated that limiting pain, mobility impairments, perceived stress and loneliness predicted membership of the moderate and higher depressive symptom classes. Retirement status and higher reported levels of worry were associated with a greater likelihood of membership of the moderate symptom classes only. Limitations Use of the CES-D is open to bias due to subjective nature of respondent reporting.
Conclusions
Results concur with previous studies on the development of depression among older people and highlight the key health related and psychological variables that may inform interventions aimed at mitigating risks of developing depression among older adults.
| Original language | English |
|---|---|
| Pages (from-to) | 345-354 |
| Number of pages | 10 |
| Journal | Journal of Affective Disorders |
| Volume | 298 |
| Issue number | Pt A |
| Early online date | 29 Oct 2021 |
| DOIs | |
| Publication status | Published (in print/issue) - 1 Feb 2022 |
Bibliographical note
Funding Information:We thank The Irish Longitudinal Study of Ageing and also the Irish Social Sciences Data Archive- www.ucd.ie/issda for providing the data utilized to conduct this research study. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Publisher Copyright:
© 2021 Elsevier B.V.
Keywords
- Depression
- Trajectory Class Modelling
- Older Adults
