Non-randomised trial, 1997
Is this a study about PMS or PMDD?
No. This is a statistics paper about how to analyse missing data in a type of drug trial design, with no connection to PMS, PMDD, or the menstrual cycle.
What they did
The authors worked out a mathematical method for analysing a specific clinical trial design (a two-sequence three-period cross-over design) when some data points are missing, such as when participants drop out. They illustrate the method with one worked example.
What they found
The abstract reports a statistical technique, not a health finding. It shows the method gives the same results as standard methods when no data are missing, and still gives valid confidence intervals when some data are missing, as long as dropout is unrelated to the measurement errors.
What it means for you
This paper is not relevant to anyone trying to understand PMS or PMDD treatments. It is a technical tool for statisticians analysing drug comparison trials, with no symptom, treatment, or outcome data about PMS or PMDD.
About the study
- Original title
- Statistical methods for two-sequence three-period cross-over designs with incomplete data
- Authors
- Chow SC, Shao J
- Journal
- Statistics in medicine
- Year
- 1997
- Study design
- Non-randomised trial
- What was tested
- None. No treatment for PMS or PMDD is tested; the paper proposes a statistical analysis method for existing trial data.