Stats- key points Flashcards
(23 cards)
LR +
sens / (1-spec)
how many times more likely it is that someone has disease if they test positive
LR-
1-sens / (spec)
how many times more likely it is that someone does not have the disease if they test negative
LR interpretation
an LR+ greater than 10 and an LR- less than 0.1 show that a test reliably discriminates between people who do and do not have disease
Experimental event rate
EER = number diseased / total number in experimental group
Relative Risk
RR = risk in exposed (EER) / risk in unexposed (CER)
Control event rate
CER= number disease / total number in control group
Odds ratio
Odds ratio = odd in exposed / odds in unexposed
OR
with with x without without / with without x without with
Odds
Odds = number diseases / number not disease
Odds v Risk
Odds:
disease/not disease (within an exposed/not exposed group)
Risk:
disease/all in group
what does R² = 0.64
means that the independent variables (like age and comorbidities) explain 64% of the variation in hospital stay length
Which parametric test can be used for >2 samples?
ANOVA, to see if means come from same population
Which parametric test is used for 2 samples
T/Student’s T, test that the samples come from a population with the same mean
Which parametric test is used for 1 sample
chi squared- compares improvement with two treatments, gives p value
What is the Mann Whitney U test used for?
Non parametric, compare means between 2 groups and give p value to see if significant
What is ARR?
difference in event rate in intervention v control
100/NNT
80% improve in intervention, 60% improve in control. 80-60= 20%
ARR= subtrAction
What is RRR
Proportion by which intervention reduces event rate
40% in placebo and 20% in control
=50% RRR
R=peRcentage
What is NNH?
100 / (% with nausea in intervention - % nausea in control)
eg 100 / (6-1) = 100/5 = 20
1 in 20 will get nausea
How do we interpret r as a value?
Correlation coefficient
0-0.2= meaningless
0.4-0.6= reasonable
0.6-0.8= high
0.8-1= suspiciously high
Relative risk
risk in exposed (a/a+b) / risk in unexposed (c/c+d)
Case control
AO-OA
Odds Ratio
Retrospective
Cohort
COhOrt=PrOspective
Relative Risk
Cohort of pirates say aRRRRR
What tests can we use to assess correlation coefficient?
Pearsons= normal
Spearmans= not normal
spearmans= special
Regression v correlation
correlation= linear association
regression= how one changes when another changes
Correlation quantifies the strength and direction of a relationship, while regression models the relationship to make predictions
