Study Designs
This type of study selects a defined group of people and follows them over time to monitor a specific outcome
Cohort Study
This numerical value is the standard cut-off; results equal to or less than this are considered statistically significant.
0.05 (P-value)
Defined as the likelihood of an abnormal test result in persons known to have the disease (True Positives / All with disease)
Sensitivity
The Women's Health Initiative RCT was stopped early after finding that Estrogen plus Progestin increased risks for breast cancer, stroke, and this primary outcome
Coronary Heart Disease
A retrospective cohort study of over 500,000 children in Denmark provided strong evidence against the link between autism and this vaccine
MMR (Measles, Mumps, and Rubella) vaccine
This design compares individuals who already have a diagnosis to a matched group of individuals who do not have the diagnosis.
Case-Control Study
A 95% Confidence Interval (CI) is generally considered not statistically significant if it crosses this number.
1
This is the likelihood of a normal test result in persons known to be disease-free.
Specificity
While increasing cardiovascular risks, the Women's Health Initiative found that hormone therapy actually decreased the risk of this
hip fractures
This antiviral medication was shown in an RCT to significantly reduce the transmission of genital herpes to susceptible partners
Valacyclovir
This statistical analysis merges data from several smaller studies to yield a more robust result in a larger population
Meta-Analysis
This metric represents the number of patients a clinician must treat for one person to benefit.
NNT (Number Needed to Treat)
This term describes the likelihood of an abnormal result in a person known not to have the disease, calculated as "1 - specificity."
False-Positive rate
Meta-analyses of passive smoking showed causality was bolstered by this finding, where higher levels of exposure correlated with higher risk.
dose-response relationship
According to a record review of large health systems, this advanced imaging test saw a massive 57% annual increase between 1996 and 2010.
PET scan
This "gold standard" study design involves randomization and a control group, ideally with both patients and researchers "blinded.
Randomized Controlled Trial (RCT)
If a p-value is above 0.05, the results are technically insignificant, but the sources state they may suggest this.
trend
Laboratory data can never replace these two primary clinical tools and must be interpreted in light of the clinical setting.
history and physical examination
In studies of smoking, factors like socioeconomic group or diet might mask the true relationship; these are known as this
confounding factors
To prevent one case of HSV-2 infection in a susceptible partner, one would need to treat this many people for a year.
38
What observational studies are cused when an RCT is not funded, practical, or ethical.
cohort or case-control designs
This risk measure provides a percentage of improvement (e.g., a 50% reduction in risk) rather than the raw number of lives saved
Relative Risk
Choosing a test cutoff that yields high sensitivity will typically result in this level of specificity.
low specificity
This specific type of study was used to evaluate passive smoking in the workplace or among spouses of smokers.
Cohort Study
This country was the site of a massive retrospective cohort study of over 500,000 children regarding the MMR vaccine
Denmark