Critical numbers Flashcards
Incidence
number of new cases occurring in the pre-defined period divided by the number of people at risk
What was you take into account with incidence?
Those with genetic conditions predisposing to disease may be excluded from the case group and
population at risk depending on what is being investigated.
Incidence rate
no of new cases in a period at risk in a population
Prevalence
no of people who have the disease amongst a population at a specific time
Case fatality
no. of people who die from disease in a period/ no. of people with disease
Mortality rate
no. of people who die from the disease in period/ no. of people who die in
period
Risk= probability of disease.
1=certain to happen, 0=certain not to happen
Relative risk= risk for the treated group/risk of the control group x 100
Risk ratio
used to find risk factors of a disease
cohort studies
measure of relative risk
RR=1
Exposed risk same and unexposed
RR
protective effect of exposure
RR>1
exposure is associated with harm
Relative risk reduction =
100- relative risk (ie. The difference the new treatment makes to the
condition)
odds
probability of event/ probability that event does not occur. (or odds= no. of times the
event occurs/no. of times it does not occur)
odd ratio
odds for disease of exposed group/ odds of disease of unexposed group
Can be used when a randomised controlled trial is dichotomous
Relative risk reduction
means the reduction in risk in the treated group compared to the untreated group
Categorical variables
Can only be assigned to a number of distinct categories. subdivided into ordinal and nominal
ordinal
categorical variables which can be ordered
Bar charts
for categorical or discrete data
Pie charts
categorical or discrete numerical
Histograms
frequency distribution of continuous variables
area of bar = frequency
Normally distributed?
uses mean and standard deviation
Non-normally distributed?
median and IQR
So it doesn’t effect the outliers
Appropriate test for
Numerical vs numerical?
correlatiom
Numerical vs categorical independent groups
t-test
Mann-Whitey test
Numerical vs categorical non independent groups
Paired t-test
Sign test
Why do larger studies have smaller CI?
larger data groups reduce spread and increase accuracy
Standard error
measure of precision
used in calculation of CI
= standard deviation/root of the sample size
positive predictive value
probability that someone has the condition if they test positive=number who have disease/no test positive
Negative predictive value
Probability someone doesn’t have the condition if they test negative-number who don’t have the disease/number who test negative
Validity
measures accurately what is meant to be measured
Responsiveness
can detect real changes when they occur
Inductive hypothesis?
proposed after analysis
Deductive hypothesis?
proposed before analysis
Referencing an article?
JOURNAL ARTICLE (print): family name INS. Title of article. Journal. Year (and date) ; vol(issue) : pages
