Leveraging automated machine learning to benchmark, deconstruct, and compare frailty indices for predicting adverse spinal surgery outcomes

Frailty indices, composite summary scores of diagnostic codes, labs, or vitals, are widely used to assess patient vulnerability. However, their value in predictive modeling remains uncertain. In a cohort of 2,912 elective lumbar spinal fusions, we retrospectively evaluated two established indices, the Hospital Frailty Risk Score (HFRS) and the electronic Frailty Index (eFI), against a curated set of 63 clinical features from electronic medical records. Using automated machine learning optimized logistic regression models, we compared predictive performances and feature importances of frailty indices, and their individual components, against curated clinical features. The curated clinical feature set consistently outperformed (measured by AUC) both frailty indices and their respective components across three outcomes: discharge disposition, length of stay, and 90-day readmission. While individual frailty components improved performance over their respective summary scores, they underperformed compared to clinical variables. However, integrating select frailty-based components with curated clinical features modestly improved predictive performance. Interaction analysis uncovered nonlinear relationships, such as between hemoglobin and smoking, that frailty indices alone did not capture. These findings highlight that while frailty indices aid broad risk stratification, their composite nature limits utility for precise, patient-specific predictive modeling, supporting hybrid strategies centered on curated variables enhanced with selected frailty elements.

• Publication year

  2026

• Venue

  Scientific Reports

• Publication date

  2026-01-10

• Fields of study

  Medicine, Computer Science

• Identifiers
  DOI 10.1038/s41598-025-31453-9PMID 41513871PMCID 12804805
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar, PubMed

• No claims are published for this paper.

• No concepts are published for this paper.

• Comparison of three frailty evaluation tools in predicting postoperative adverse events in older patients undergoing lumbar fusion surgery: a prospective cohort study of 240 patients

  2025cited by this paper
• Pre-Operative Anemia is an Unsuspecting Driver of Machine Learning Prediction of Adverse Outcomes after Lumbar Spinal Fusion.

  2025cited by this paper
• Optimizing predictive model performance in adult spinal deformity surgery: a comparative head-to-head analysis of learning models for perioperative complications.

  2025cited by this paper
• Identification of testicular cancer with T2-weighted MRI-based radiomics and automatic machine learning

  2025cited by this paper
• Explainable coronary artery disease prediction model based on AutoGluon from AutoML framework

  2024cited by this paper
• Baseline Frailty Measured by the Risk Analysis Index and 30-Day Mortality After Surgery for Spinal Malignancy: Analysis of a Prospective Registry (2011–2020)

  2024cited by this paper
• The Evolution of Risk Assessment in Spine Surgery: A Narrative Review.

  2024cited by this paper
• Integration of chronological age does not improve the performance of a mixed-effect model using comorbidity burden and frailty to predict 90-day readmission after surgery for degenerative scoliosis.

  2024cited by this paper
• Integrating Machine Learning and the SHapley Additive exPlanations (SHAP) Framework to Predict Lymph Node Metastasis in Gastric Cancer Patients Based on Inflammation Indices and Peripheral Lymphocyte Subpopulations

  2024cited by this paper
• Current Applications and Future Implications of Artificial Intelligence in Spine Surgery and Research: A Narrative Review and Commentary

  2024cited by this paper
• Assessing the limitations of relief-based algorithms in detecting higher-order interactions

  2024cited by this paper
• An improved hyperparameter optimization framework for AutoML systems using evolutionary algorithms

  2023cited by this paper
• Encoding categorical data: Is there yet anything 'hotter' than one-hot encoding?

  2023cited by this paper
• Machine Learning in Spine Surgery: A Narrative Review

  2023cited by this paper
• AutoML accurately predicts endovascular mechanical thrombectomy in acute large vessel ischemic stroke

  2023cited by this paper
• Developing Mixed-effects Models to Optimize Prediction of Postoperative Outcomes in a Modern Sample of Over 450,000 Patients Undergoing Elective Cervical Spine Fusion Surgery

  2023cited by this paper
• Application of explainable artificial intelligence in medical health: A systematic review of interpretability methods

  2023cited by this paper
• Comparing the Hospital Frailty Risk Score and the Clinical Frailty Scale Among Older Adults With Chronic Obstructive Pulmonary Disease Exacerbation

  2023cited by this paper
• Explainable Artificial Intelligence for Predictive Modeling in Healthcare

  2022cited by this paper
• Genetic Algorithm Based Hyper-Parameter Tuning to Improve the Performance of Machine Learning Models

  2022cited by this paper
• The impact of frailty on postoperative complications in geriatric patients undergoing multi-level lumbar fusion surgery

  2022cited by this paper
• Machine Learning Applications of Surgical Imaging for the Diagnosis and Treatment of Spine Disorders: Current State of the Art

  2022cited by this paper
• The inclusion of frailty improves predictive modeling for postoperative outcomes in the surgical management of primary and secondary lumbar spine tumors.

  2021cited by this paper
• Frailty adversely affects outcomes of patients undergoing spine surgery: A systematic review.

  2021cited by this paper
• Hyperparameter Optimization: Comparing Genetic Algorithm against Grid Search and Bayesian Optimization

  2021cited by this paper
• Congress of Neurological Surgeons Systematic Review and Evidence-Based Guidelines for Perioperative Spine: Preoperative Surgical Risk Assessment

  2021cited by this paper
• Central Sarcopenia, Frailty and Comorbidity as Predictor of Surgical Outcome in Elderly Patients with Degenerative Spine Disease

  2021cited by this paper
• The performance of frailty in predictive modeling of short-term outcomes in the surgical management of metastatic tumors to the spine.

  2021cited by this paper
• Comparison of frailty metrics and the Charlson Comorbidity Index for predicting adverse outcomes in patients undergoing surgery for spine metastases.

  2021cited by this paper
• CatBoost for big data: an interdisciplinary review

  2020cited by this paper
• Survey on categorical data for neural networks

  2020cited by this paper
• Validation of clinical prediction models: what does the "calibration slope" really measure?

  2020cited by this paper
• A tutorial on calibration measurements and calibration models for clinical prediction models

  2020cited by this paper
• Frailty Screening Using the Electronic Health Record within a Medicare Accountable Care Organization.

  2019influential reference
• The Impact of Regularization on High-dimensional Logistic Regression

  2019cited by this paper
• Calibration: the Achilles heel of predictive analytics

  2019cited by this paper
• Development and validation of a Hospital Frailty Risk Score focusing on older people in acute care settings using electronic hospital records: an observational study

  2018cited by this paper
• External validation of the adult spinal deformity (ASD) frailty index (ASD-FI)

  2018cited by this paper
• External Validation of the Adult Spinal Deformity (ASD) Frailty Index (ASD-FI) in the Scoli-RISK-1 Patient Database

  2018cited by this paper
• A Unified Approach to Interpreting Model Predictions

  2017influential reference
• Benchmarking Relief-Based Feature Selection Methods

  2017influential reference
• Automating Biomedical Data Science Through Tree-Based Pipeline Optimization

  2016cited by this paper
• Development and validation of an electronic frailty index using routine primary care electronic health record data

  2016cited by this paper
• TPOT: A Tree-based Pipeline Optimization Tool for Automating Machine Learning

  2016cited by this paper
• Extended length of stay after surgery: complications, inefficient practice, or sick patients?

  2014cited by this paper
• Multiple Threshold Spatially Uniform ReliefF for the Genetic Analysis of Complex Human Diseases

  2013cited by this paper
• MICE: Multivariate Imputation by Chained Equations in R

  2011cited by this paper
• Scikit-learn: Machine Learning in Python

  2011cited by this paper
• Multiple imputation by chained equations: what is it and how does it work?

  2011cited by this paper
• Estimation of the Youden Index and its Associated Cutoff Point

  2005cited by this paper
• Comparison of Machine Learning Techniques with Classical Statistical Models in Predicting Health Outcomes

  2004cited by this paper
• Chapter 53 The shapley value

  2002cited by this paper
• Multifactor-dimensionality reduction reveals high-order interactions among estrogen-metabolism genes in sporadic breast cancer.

  2001influential reference
• Genetic programming as a means for programming computers by natural selection

  1994cited by this paper
• Index for rating diagnostic tests

  1950cited by this paper

• No citing papers are available for this paper.