Stat's Flashcards
Cohort study
Longitudinal study where people followed up at random points and cross sections performed at random points
Individuals share common characteristics
Cross sectional study
Collect data from individuals at random point in time- will help estimate incidence at a certain time
Case control study
Looks at a cohort retrospectively to see what exposures may have led to a disease
Has control
Case series study
Looks back at a group over time with a disease with no control
Levels of evidence
1- meta analysis
Parametric vs non parametric tests
Parametric= normally distributed
Analysing parametric data
Paired vs unpaired t test
Analysing non-parametric data
Paired like before and after= wilcoxon signed
Mann- witney u= unpaired data
What is chi squared test used for
Categorical data comparisons (proportions/percentages)
Phases of clinical trial
Phase 1- a few people given to assess its safety and pharmacokinetics etc
Phase 2- 100 people given to assess therapeutic effectiveness and side effects
Phase 3- large cohort given it and assess difference to current treatment
Phase 4- post license where monitor long term effects
What shows publication bias in meta analysis
Funnel plot
What is outcome in cohort vs case control study
Cohort- RR
Case control- OR
How assess correlation in parametric vs non parametric data
Parametric- pearsons coefficient
Non-parametric- spearmans coefficient
Breslow thickness management
Remove and then based off pathology report determine if need further excision
0-1mm= 1cm
1-2mm= 1-2cm
2-4mm= 2-3
>4mm= 3cm
What is specificity and what is equation
% of people who test negative for a disease among group of people who do not have the disease