Advanced analytical methodologies for measuring healthy ageing and its determinants, using factor analysis and machine learning techniques: the ATHLOS project

Francisco Félix Caballero,Worrawat Engchuan,Josep Maria Haro,Albert Sánchez-Niubó,Holger Arndt,George Soulis,José Luis Ayuso-Mateos,Somnath Chatterji,Demosthenes B Panagiotakos

doi:10.1038/srep43955

Abstract

A most challenging task for scientists that are involved in the study of ageing is the development of a measure to quantify health status across populations and over time. In the present study, a Bayesian multilevel Item Response Theory approach is used to create a health score that can be compared across different waves in a longitudinal study, using anchor items and items that vary across waves. The same approach can be applied to compare health scores across different longitudinal studies, using items that vary across studies. Data from the English Longitudinal Study of Ageing (ELSA) are employed. Mixed-effects multilevel regression and Machine Learning methods were used to identify relationships between socio-demographics and the health score created. The metric of health was created for 17,886 subjects (54.6% of women) participating in at least one of the first six ELSA waves and correlated well with already known conditions that affect health. Future efforts will implement this approach in a harmonised data set comprising several longitudinal studies of ageing. This will enable valid comparisons between clinical and community dwelling populations and help to generate norms that could be useful in day-to-day clinical practice.

Highlights

Mortality or other outcomes[6,7], including high health care expenditures[8]
Using a Bayesian multilevel Item Response Theory (IRT) approach, items that vary across studies will be considered together with anchor items and items that vary across waves in order to create a metric of health which allows for comparison across studies and waves of those longitudinal studies
A two-step factor analysis approach was conducted to assess the unidimensionality of the self-reported health questions and measured tests found at ELSA baseline

Summary

Introduction

Mortality or other outcomes[6,7], including high health care expenditures[8]. just counting diseases could fail to capture the true impact of health conditions on health status. There is a need to come up with a metric that can quantify the health status of the individuals This metric could be aggregated to the population level, allowing for comparisons across populations and over time, and being sensitive to change. The aims of the present study, and under the context of the Ageing Trajectories of Health: Longitudinal Opportunities and Synergies (ATHLOS) project (EU HORIZON2020–PHC-635316, http://athlosproject.eu/), were to create a metric of health using an Item Response Theory (IRT) approach, which can be used for comparison of the population across the first six waves of the English Longitudinal Study of Ageing[11] (ELSA) conducted between 2002 and 2012, and to explore determinants of this health score based on socio-demographics. Advanced analytical methodologies in pattern recognition and computational learning, as Machine Learning approaches, can be employed to explore factors associated with the metric of health

Methods

Results

Conclusion