Lasso regression demonstrated the best predictive performance for quality of life in MS patients (R2 = 0.86), with disability level (EDSS) emerging as the strongest negative predictor, followed by fatigue severity, sleep quality, and impaired functional mobility.
Key Findings
Results
Among six machine learning models tested, Lasso regression demonstrated the best predictive performance for quality of life in MS patients.
Lasso regression achieved RMSE = 5.02, MAE = 4.04, and R2 = 0.86
Other models tested included Linear Regression, Elastic Net, Support Vector Machines, Random Forest, and XGBoost
Five-times repeated five-fold cross-validation was applied for internal validation
Model performance was evaluated using root mean squared error (RMSE), mean absolute error (MAE), and coefficient of determination (R2)
Results
Disability level (EDSS) emerged as the strongest negative predictor of quality of life in MS patients.
EDSS was identified as the strongest negative predictor in the multivariable machine learning model
Quality of life was assessed using the MS Quality of Life-54 (MSQoL-54)
Disability was measured using the Expanded Disability Status Scale (EDSS)
The study included 101 individuals diagnosed with MS in a cross-sectional design
Results
Fatigue severity, sleep quality, and impaired functional mobility were identified as significant negative predictors of quality of life in MS patients.
Fatigue severity was the second strongest predictor after EDSS
Sleep quality was assessed using the Pittsburgh Sleep Quality Index
Functional mobility was assessed using the Timed Up and Go test
These variables contributed independently to the multivariable predictive model
Results
Depression showed a strong bivariate association with quality of life but had a relatively lower weight in the final multivariable predictive model.
Depression was measured using the Beck Depression Inventory
The discrepancy suggests depression's effect may be partially mediated or shared with other predictors in the multivariable context
This finding highlights that bivariate associations do not necessarily translate to strong independent predictors in multivariable models
Results
Disease duration was not significantly correlated with quality of life in univariate analyses but showed a positive contribution in the multivariable model.
Disease duration demonstrated no significant correlation with quality of life when examined in isolation
In the multivariable Lasso regression model, disease duration showed a positive contribution
This finding suggests that disease duration's relationship with quality of life may only be apparent when accounting for other clinical variables simultaneously
Methods
The study sample consisted of 101 individuals diagnosed with MS evaluated across demographic, clinical, and functional domains.
The study used a cross-sectional design
Variables assessed included demographic variables, clinical characteristics, disability level (EDSS), fatigue severity, sleep quality (Pittsburgh Sleep Quality Index), depression level (Beck Depression Inventory), and functional mobility (Timed Up and Go test)
Quality of life was assessed using the MS Quality of Life-54 (MSQoL-54)
Sample size of 101 MS patients was used to train and validate multiple machine learning regression models
What This Means
This research suggests that the quality of life for people living with multiple sclerosis (MS) is most strongly shaped by their level of physical disability, fatigue, sleep quality, and ability to move around independently. Researchers studied 101 people with MS and used several advanced statistical approaches called machine learning models to determine which factors best predict overall quality of life. The most accurate model — a technique called Lasso regression — was able to explain 86% of the variation in quality of life scores, suggesting it captured most of the key influences on how well people with MS feel day-to-day.
One interesting finding was that depression had a strong relationship with quality of life when looked at alone, but its independent contribution became smaller once all other factors were considered together. This suggests that depression may be closely intertwined with disability, fatigue, and sleep problems rather than acting as a fully separate influence. Similarly, how long someone had been living with MS did not appear related to quality of life on its own, but it did show a positive association when considered alongside the other factors, which may reflect complex interactions between disease experience and other clinical features.
This research suggests that rehabilitation programs and clinical care for MS patients could benefit from addressing not just physical disability, but also fatigue, sleep disturbances, and mobility limitations as interconnected targets for improving quality of life. The use of machine learning methods to analyze these multidimensional factors provides a framework that could help clinicians identify which patients are most at risk for poor quality of life and tailor interventions accordingly.
Check Your Own Numbers
Upload your bloodwork. We'll cross-reference your results against this study and 4,700 others.
Yetiş A, Canli M, Canli &, Kocaman H, Yildirim H, Yildiz N, et al.. (2026). Machine learning-based identification of multidimensional predictors of quality of life in individuals with multiple sclerosis.. Medicine. https://doi.org/10.1097/MD.0000000000049025