Hidden Markov models to classify depression and estimate transitions across health states
Open Access
- 1 January 2026
- book chapter
- Published by Firenze University Press
- p. 63-68
Abstract
Depression is one of the most common mental health disorders. It is characterised by a great heterogeneity of symptoms and leads to a significant deterioration in quality of life. Population ageing, combined with the increasing prevalence of depression in older age groups, has made this disorder an increasingly important public health problem. The aim of this work is to study the dynamics and the evolution of depression over time among elderly Italians. To this aim, we analyse data from the Survey of Health, Ageing and Retirement in Europe (SHARE). The final sample is composed of 1322 old Italian people, interviewed in the fifth, sixth, eighth and ninth waves of SHARE (2013, 2015, 2019 and 2022). Based on the latent nature of the variable of interest, the data are analysed with Hidden Markov Models (HMMs): they are able to identify different behavioural profiles that are not directly observed, allowing to identify groups of individuals with similar behavioural patterns over time from a set of observed variables. The best fitting model defines four latent classes with a first-order Markov chain and response probabilities constant over time; gender and age are active covariates for the initial state and transitions. These classes identify the elderly group with “no risk of depression”, “mild depression”, “moderate depression” and “severe depression”, respectively. As expected, we observed gender differences in levels and some trajectories. In particular, the transition from severe depression to a better state was more difficult after the Covid-19 pandemic. Recognising structural changes in the mental health of the elderly in the post-pandemic period may be useful to support targeted interventions for the population most at risk, optimising prevention or treatment.Keywords
This publication has 11 references indexed in Scilit:
- Prevalence and risk factors of depression among community dwelling elderlyArchives of Gerontology and Geriatrics, 2014
- Introduction to Probability Theory2014
- Data Resource Profile: The Survey of Health, Ageing and Retirement in Europe (SHARE)International Journal of Epidemiology, 2013
- Risk factors of major depression in the elderlyEuropean Psychiatry, 2005
- Risk Factors for Depression Among Elderly Community Subjects: A Systematic Review and Meta-AnalysisAmerican Journal of Psychiatry, 2003
- Depression in Late Life: Review and CommentaryThe Journals of Gerontology, Series A: Biological Sciences and Medical Sciences, 2003
- Epidemiology of women and depressionJournal of Affective Disorders, 2003
- Development of the EURO–D scale – a European Union initiative to compare symptoms of depression in 14 European centresThe British Journal of Psychiatry, 1999
- Review of community prevalence of depression in later lifeThe British Journal of Psychiatry, 1999
- Psychiatric disorders and 15-month mortality in a community sample of older adults.American Journal of Public Health, 1989