One of the aims of the Ageing Trajectories of Health: Longitudinal Opportunities and Synergies (ATHLOS) project is to develop a single metric of healthy ageing by integrating data across multiple functioning domains. Based on a set of self-reported health questions and measured tests, which can vary across the different waves and surveys included in ATHLOS, a common metric of health will be created over the harmonised data set generated at a first stage.
Researchers participating in ATHLOS have conducted a preliminary analysis over the different waves of the English Longitudinal Study of Ageing (ELSA), one of the longitudinal surveys included in ATHLOS. The manuscript entitled “Advanced analytical methodologies for measuring healthy ageing and its determinants, using factor analysis and machine learning techniques: the ATHLOS project” has been published in Scientific Reports and co-authored by researchers from Universidad Autónoma de Madrid, Parc Sanitari Sant Joan de Déu, Spain, Harokopio University of Athens, Spring Techno, and the World Health Organization.
Different statistical techniques (including Factor Analysis, Bayesian multilevel Item Response Theory and Machine Learning methods) have been employed in the manuscript to achieve the aim of creating a common metric of health which could be compared across the different waves of a longitudinal survey.
The metric of health created showed a good performance in terms of predictive ability and compared with a simple sum of chronic conditions. The procedure described in the article can be extended and applied to the general aims of ATHLOS, considering the presence of items that vary across studies together with items available in all waves and items that vary across waves. The metric could be then employed to compare trajectories on health across the participants from the different epidemiological cohorts included in ATHLOS.