Predicting mortality and incident immobility in older Belgian men by characteristics related to sarcopenia and frailty
There is an increasing awareness of sarcopenia in older people. We applied machine learning principles to predict mortality and incident immobility in older Belgian men through sarcopenia and frailty characteristics. Mortality could be predicted with good accuracy. Serum 25-hydroxyvitamin D and bone mineral density scores were the most important predictors.
Machine learning principles were used to predict 5-year mortality and 3-year incident severe immobility in a population of older men by frailty and sarcopenia characteristics.
Using prospective data from 1997 on 264 older Belgian men (n = 152 predictors), 29 statistical models were developed and tuned on 75% of data points then validated on the remaining 25%. The model with the highest test area under the curve (AUC) was chosen as the best. From these, ranked predictor importance was extracted.
Five-year mortality could be predicted with good accuracy (test AUC of .85 [.73; .97], sensitivity 78%, specificity 89% at a probability cut-off of 22.3%) using a Bayesian generalized linear model. Three-year incident severe immobility could be predicted with fair accuracy (test AUC .74 [.57; .91], sensitivity 67%, specificity 78% at a probability cut-off of 14.2%) using a multivariate adaptive regression splines model. Serum 25-hydroxyvitamin D levels and hip bone mineral density scores were the most important predictors of mortality, while biochemical androgen markers and Short-Form 36 Physical Domain questions were the most important predictors of immobility. Sarcopenia assessed by lean mass estimates was relevant to mortality prediction but not immobility prediction.
Using advanced statistical models and a machine learning approach 5-year mortality can be predicted with good accuracy using a Bayesian generalized linear model and 3-year incident severe immobility with fair accuracy using a multivariate adaptive regression splines model.
KeywordsSarcopenia Osteoporosis Machine learning Big data Prediction
This work was supported by the Obel Family Foundation of Aalborg, Denmark, which is funding the 3-year PhD fellowship of the corresponding author in full. The work was also supported by the Department of Clinical Medicine of Aalborg University, Denmark, which co-funds the activities inherent in the PhD fellowship of the corresponding author. Grant numbers are not maintained in Denmark.
Compliance with ethical standards
Conflicts of interest
CK has received travel grants from Eli Lilly, Otsuka Pharmaceutical and is a speaker for Novartis and Otsuka Pharmaceutical. PE is an advisory board member with Amgen, MSD, and Eli Lilly and at the speaker’s bureau with Amgen and Eli Lilly, stocks from Novo Nordisk A/S. PV has received unrestricted grants from MSD and Servier, and travel grants from Amgen, Eli Lilly, Novartis, Sanofi-Aventis, and Servier.
- 5.De Buyser SL, Petrovic M, Taes YE, et al (2016) Validation of the FNIH sarcopenia criteria and SOF frailty index as predictors of long-term mortality in ambulatory older men. Age AgeingGoogle Scholar
- 6.Cawthon PM, Blackwell TL, Cauley J, Kado DM, Barrett-Connor E, Lee CG, Hoffman AR, Nevitt M, Stefanick ML, Lane NE, Ensrud KE, Cummings SR, Orwoll ES (2015) Evaluation of the usefulness of consensus definitions of sarcopenia in older men: results from the observational osteoporotic fractures in men cohort study. J Am Geriatr Soc 63(11):2247–2259CrossRefPubMedPubMedCentralGoogle Scholar
- 10.Lapauw B, Goemaere S, Zmierczak H, van Pottelbergh I, Mahmoud A, Taes Y, de Bacquer D, Vansteelandt S, Kaufman JM (2008) The decline of serum testosterone levels in community-dwelling men over 70 years of age: descriptive data and predictors of longitudinal changes. Eur J Endocrinol 159(4):459–468CrossRefPubMedGoogle Scholar
- 15.Kuhn M (2008) Building Predictive Models in R Using the caret Package. J Statist Softw 28(5):1–26Google Scholar