Related Experiment Video
Updated: Feb 18, 2026
07:15
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
7.6K
Symptom Clusters Using the Edmonton Symptom Assessment System in Patients With Bone Metastases: A Reanalysis
Luluel Khan1, Gemma Cramarossa1, Emily Chen1
1Rapid Response Radiotherapy Program, Odette Cancer Centre, Sunnybrook Health Sciences Centre, University of Toronto, Canada.
World Journal of Oncology
|November 18, 2017
Summary
Statistical methods significantly impact symptom cluster identification in bone metastasis patients. Different methods yield varying results, highlighting the need for standardized analysis in future research.
Area of Science:
- Oncology
- Biostatistics
- Palliative Care
Background:
- Bone metastases present complex symptom profiles.
- Understanding symptom clusters is crucial for effective palliative care.
- Previous research utilized Principal Component Analysis (PCA) for symptom cluster extraction.
Purpose of the Study:
- To compare symptom cluster composition using three statistical methods: PCA, Hierarchical Cluster Analysis (HCA), and Exploratory Factor Analysis (EFA).
- To examine temporal changes in symptom clusters.
- To compare symptom clusters between responders and non-responders to palliative radiation treatment.
Main Methods:
- Utilized a dataset of 518 bone metastasis patients who completed the Edmonton Symptom Assessment System (ESAS).
- Applied PCA, HCA, and EFA to extract symptom clusters at baseline and at 1, 2, 4, 8, and 12 weeks post-radiation.
- Analyzed clusters within subgroups of radiation responders and non-responders.
Main Results:
- No complete consensus was reached among PCA, HCA, and EFA regarding the number and composition of symptom clusters.
- Little correlation was observed between clusters derived from the three methods using the same dataset.
- Distinct symptom clusters emerged in responders versus non-responders across all methods and time points.
- Clusters varied over time within each subgroup, though depression and anxiety consistently clustered together.
Conclusions:
- The choice of statistical method significantly influences the identification and composition of symptom clusters in bone metastasis patients.
- Variability in cluster analysis necessitates the adoption of a common analytical approach for consistency and comparability in future research.
