Trial Development
This phase includes a small number of healthy volunteers or people with the studied disease state to test safety and dosage
What is Phase 1?
Considered to be the least reliable type of medical study
What is expert opinion?
Used to test for statistical significance when using continuous data with normal distribution with 3 or more samples
What is analysis of variance (ANOVA or F-test)?
States there is no statistically significant difference between groups (studies typically want to prove it is not true and reject it)
What is a null hypothesis (H0)?
A maximum permissible error margin and the threshold for rejecting the null hypothesis (commonly set to 0.05)
What is alpha?
This should be granted by the FDA for clinical studies to begin in human subjects
What is investigational new drug (IND) approval?
Retrospective or prospective comparisons of patients with an exposure to those without an exposure
What are cohort studies?
Test used when the patient serves as their own control in a before-and-after or pre/post-measurement study with continuous and normal distribution data
What is a paired t-test
The probability that a test will reject the null hypothesis correctly (ability to avoid a type II error)
What is power?
The probability (confidence) that the conclusion of the study is correct (or % chance that it is not)
What is the p-value?
This phase includes post-marketing surveillance for long-term and real-world efficacy/safety assessment
What is phase IV?
Preferred study type to determine cause and effect or superiority, but may not reflect real-life scenarios (depending on exclusion criteria)
What are randomized controlled trials (RCTs)
Test used to determine statistical significance between independent treatment groups with continuous data
What is an independent/unpaired student t-test?
What is a per protocol analysis?
Provides the same information about significance as the p-value, plus the precision of the result
What is the confidence interval (CI)?
What is phase II?
Good for looking at outcomes when intervention is unethical by looking for cases reported in medical records, but cause and effect cannot reliably be determined
What are case-control studies?
Test used to determine statistical significance between treatment groups with nominal or ordinal data
What is a chi-square test?
This needs to be increased to increase study power and decrease the risk of a type II error.
If the alternative hypothesis was accepted and the null hypothesis was rejected in error.
This bias occurs when study participants with specific characteristics (such as less severe disease) have a higher dropout rate, skewing the results of a study
What is attrition bias?
This is an example of what study type? Method: The authors wanted to define the appropriate treatment sequence after first-line treatment failure. No statistical tests reported. Conclusion: There are no predictive biomarkers that determine the best sequence of treatment. More studies needed.
What is a systematic review article?
Test commonly used for small sets of nominal data with 2 groups
What is Fisher's Exact test
The probability of a type II error
What is beta?
You can adjust for confounders by using multiple variables in this equation
What is multiple regression?