Related Experiment Videos
Outcome measures in split mouth caries trials and their statistical evaluation
1Dental Services, Dental School, University of Western Australia, Perth.
Community Dentistry and Oral Epidemiology
|June 1, 1994
Summary
Split mouth study designs in dental trials can be improved with better statistical analysis. Researchers recommend using relative risk and confidence intervals for more accurate outcome measures.
Area of Science:
- Dental research
- Clinical trials
- Biostatistics
Background:
- Split mouth study designs control for confounding variables in dental trials.
- Traditional outcome measures like material retention and effectiveness lack robust statistical evaluation in many studies.
- McNemar's X2 is often used but cannot directly assess effectiveness and net gain.
Purpose of the Study:
- To evaluate statistical outcome measures in split mouth trials.
- To suggest methods for calculating confidence intervals for effectiveness and net gain.
- To recommend relative risk as a preferred outcome measure.
Main Methods:
- Literature survey of statistical evaluations in split mouth studies.
- Analysis of traditional outcome measures (effectiveness, net gain).
- Demonstration of calculating confidence intervals and relationship to relative risk.
Main Results:
- Many split mouth studies lack statistical evaluation of outcomes.
- McNemar's X2 is common but limited in assessing effectiveness and net gain.
- Relative risk is closely related to effectiveness and recommended for split mouth trials.
Conclusions:
- Statistical evaluation, including confidence intervals, is crucial for split mouth trials.
- Relative risk should be considered the preferred outcome measure for its statistical robustness.
- Improved statistical methods enhance the reliability of split mouth study findings.