Abstract

Background: Identifying leading determinants for disease-free status may provide evidence for action priorities, which is imperative for public health with an expanding aged population worldwide. This study aimed to identify leading determinants, especially modifiable factors for disease-free status using machine learning methods.Methods: We included 52,036 participants aged 45–64 years from the 45 and Up Study who were free of 13 predefined chronic conditions at baseline (2006–2009). Disease-free status was defined as participants aging from 45–64 years at baseline to 55–75 years at the end of the follow-up (December 31, 2016) without developing any of the 13 chronic conditions. We used machine learning methods to evaluate the importance of 40 potential predictors and analyzed the association between the number of leading modifiable healthy factors and disease-free status.Results: Disease-free status was found in about half of both men and women during a mean 9-year follow-up. The five most common leading predictors were body mass index (6.4–9.5% of total variance), self-rated health (5.2–8.2%), self-rated quality of life (4.1–6.8%), red meat intake (4.5–6.5%), and chicken intake (4.5–5.9%) in both genders. Modifiable behavioral factors including body mass index, diets, smoking, alcohol consumption, and physical activity, contributed to 37.2–40.3% of total variance. Participants having six or more modifiable health factors were 1.63–8.76 times more likely to remain disease-free status and had 0.60–2.49 more disease-free years (out of 9-year follow-up) than those having two or fewer. Non-behavioral factors including low levels of education and income and high relative socioeconomic disadvantage, were leading risk factors for disease-free status.Conclusions: Body mass index, diets, smoking, alcohol consumption, and physical activity are key factors for disease-free status promotion. Individuals with low socioeconomic status are more in need of care.

Highlights

  • The global population is aging, and it is estimated that 16% of the total population will be 65 years or older by 2050 [1]

  • The importance of determinants in rank on disease-free status is less known [12, 14], determining the leading modifiable and non-modifiable predictors based on big data using prediction models especially machine learning considering its advantage in prediction performance is imperative for prioritizing public health actions [18]

  • We aimed to prospectively examine the association of lifestyle behaviors, family history of chronic disease, socioeconomic status, psychological and geographic factors with disease-free status and evaluated the importance of 40 potential predictors using machine learning methods based on a large cohort study and claims databases

Read more

Summary

Introduction

The global population is aging, and it is estimated that 16% of the total population will be 65 years or older by 2050 [1]. Physiological degeneration with aging is associated with numerous complications, including cardiometabolic disorders, cancer, mental disorders, dementia, Parkinson’s disease, musculoskeletal disorders, and asthma [4, 5]. These conditions account for a predominant proportion of global mortality with cardiovascular disease and cancer as the first two leading contributors [6]. Middle age represents an important period for chronic disease prevention, identifying the leading determinants for disease-free status during this period is essential [16]. Identifying leading determinants for disease-free status may provide evidence for action priorities, which is imperative for public health with an expanding aged population worldwide. This study aimed to identify leading determinants, especially modifiable factors for disease-free status using machine learning methods

Objectives
Methods
Results
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call