Sample Sizes for Cancer Trials Focusing on Quality of Life
Author Information
Author(s): Julious S A, Campbell M J, Walker S J, George S L, Machin D
Primary Institution: SmithKline Beecham, University of Sheffield, University of Southampton
Hypothesis
Can the assumption of Normal distribution in HRQoL data lead to incorrect sample size calculations in cancer trials?
Conclusion
Assuming Normal distribution for HRQoL data can result in unrealistically sized studies, and a more appropriate method should be used.
Supporting Evidence
- HRQoL instruments are important in evaluating health care in cancer trials.
- Calculating sample size is essential for detecting differences in trials.
- Many researchers incorrectly assume Normal distribution for HRQoL data.
Takeaway
When researchers plan cancer studies, they need to figure out how many people to include. If they think the data is shaped like a bell curve when it’s not, they might end up with too many or too few people in the study.
Methodology
The study investigated the distribution of scores from HRQoL instruments completed by lung cancer patients.
Potential Biases
Potential bias in assuming Normal distribution for HRQoL data.
Limitations
The study may not cover all types of HRQoL instruments or cancer types.
Participant Demographics
Lung cancer patients.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website