Studies Flashcards
Therapy
- RCT design
Diagnosis
- cross sectional design
Harm
- CC design
Prognosis
- cohort
Sensitivity
- the probability of an true positive testing positive
Specificity
The probability if a true negative actually testing negative
Best measure of central tendency in a skewed distribution
- median
Masking aka blinding
- reduces measurement bias
Random allocation
- reduces selection bias
Absolute risk
An estimation of future occurrence based on past incidents
Following a group of people based on exposure to a potential risk factor
- cohort
- Study is based on development of the disease given exposure
Study based Where in the unit of analysis is a group of people
- ecological
- birthweight in cities with high pollution
Power of the study
Determined by the number of people in the study i.e. the more people study the greater the power statistical significance or not
90,95,99%
1.65,2,2.58 resp
Alpha
Type I error
- rejecting null hypothesis when it’s true
Beta
Type II error
- accepting the null hypothesis when it’s false
OR
- BC/AD
ARR
- {CER-EER}
- Preserves baseline risk but does not discriminate between small and large values
RRR
CER-EER/CER
- having a career reduces your relative risk in life
- shows the different between risk categories but losses base line (may be a relatively big difference but small when compared to baseline)
Predictive value and prevalence
- as prevalence decreases PPV goes down parabolically
