Related Experiment Video
Updated: Jul 13, 2026
04:20
Practical Considerations for the Design, Execution, and Interpretation of Studies Involving Whole-Bone Bending Tests of Rodent Bones
Published on: September 1, 2023
Sample size requirements for bone density precision assessments and effect on patient categorization: a Monte Carlo
Alireza Moayyeri1, Mohsen Sadatsafavi, William D Leslie
1Department of Public Health and Primary Care, Institute of Public Health, University of Cambridge, Cambridge, UK.
Bone
|August 21, 2007
Summary
A small sample size in bone mineral density (BMD) precision studies can lead to inaccurate assessments of patient change. Larger sample sizes are crucial for reliable BMD monitoring and to minimize categorization errors.
Area of Science:
- Medical Imaging
- Biostatistics
- Osteoporosis Research
Background:
- Bone mineral density (BMD) monitoring is essential for managing osteoporosis and assessing treatment efficacy.
- Current precision studies may use insufficient sample sizes (degrees of freedom, df), potentially compromising the reliability of BMD change detection.
Purpose of the Study:
- To evaluate the impact of precision study sample size on the accuracy of categorizing bone mineral density (BMD) changes in clinical patients.
- To determine the optimal sample size (df) required to minimize errors in BMD change classification.
Main Methods:
- Monte Carlo simulations were employed to assess the effect of varying sample sizes (5-500 df) on BMD change detection.
- Least Significant Change (LSC) was calculated from 198 spine and 193 hip scan-pairs and compared against a reference change fraction (RCF).
- Bootstrap sampling was used to estimate confidence limits (80% and 95%) for LSC values.
Main Results:
- A sample size of 30 df led to significant overdetection (up to 12.5%) and underdetection (up to 10.0%) of BMD changes at 95% confidence limits.
- Approximately 140 df are needed to avoid a 5% overdetection of spine change, and 150 df to avoid a 5% underdetection, with 95% confidence.
- Current recommendations for precision study sample sizes may neglect the impact on change classification accuracy.
Conclusions:
- Sample sizes larger than the commonly used 30 df are necessary for achieving low levels of categorization error in BMD monitoring.
- Adequate sample size in precision studies is critical for reliable clinical patient assessment and effective osteoporosis management.
- Future recommendations for BMD precision studies should incorporate sample size requirements that ensure accurate change classification.
