lay definitions Flashcards
Sensitivity
Proportion of individuals with disease that have positive test result e.g. sensitivity 66% 2:3 of the individuals who have the disease will be identified by testing; 1:3 with disease will be testing negative and missed
False negative
Test result is negative in the presence of the disease
Specificity
Proportion of individuals without disease that have negative test result
e. g. specificity 90% 9:10 of the individuals who don?t have the disease will test negative;
1: 10 without disease will be incorrectly labelled with disease
False positive
Test result is positive in the absence of the disease
Positive predictive value
Proportion of individuals with positive test who truly have the disease
e.g. PPV 90% 9:10 of those who tested positive have the disease; 1:10 test positive but don?t have the disease
Confounder
A factor (exposure) which can explain (entirely or partially) the observed association between ?exposure? and ?outcome?
A confounder is a factor that distorts the association between ?exposure? and ?outcome?
Interaction
The association between ?exposure? and ?outcome? is of different strength in different strata of another factor e.g. the association between smoking and lung cancer is stronger in older people than in younger people. In this association, age is the effect modifier
Crude estimate
Estimate derived without accounting for the effect of confounders (such as age, sex)
Adjusted
estimate
Controlled for the effect of confounders
Taking into account ?factors? (such as age, sex, deprivation, etc) that distort the association between ?exposure? and ?outcome?
Regression analysis
Statistical method that accounts simultaneously for multiple confounding factors
Standardisation
A way of controlling for age (or other potential confounders) so that rates of disease or death in populations with different age structures (or other potential confounding factors) can be compared
DSR
A method used to account for the difference in the age structure (or in other confounders) of populations in order to make a valid comparison of rates of disease/death
SMR
A method used to account for the difference in the age structure (or in other confounders) of populations in order to make valid comparison of death rates
p-value
The probability of having the observed estimate due to chance e.g. if P < 0.05 the difference observed occurs by chance alone less than 5 times in 100
95% Confidence interval
95% confident that the true effect size lies within this range, uncertainty range
Power
Indicates how good a study is in identifying an effect (or difference in intervention) if in reality an effect (or difference) exists
Precision
Refers to the width of the
confidence interval
Prevalence
Proportion of existing cases
Incidence risk
Proportion of new cases developed during a given time
Incidence rate
Number of new cases over the sum of the different times each individual was at risk
Risk
Probability that an event will occur
Odds
Probability that an event will occur
Rate
Measure of effect (e.g. death) changing with time
Hazard
Rate
Risk ratio
Probability of having the outcome among exposed as compared to ?unexposed?
Odds ratio
e.g. OR of lung cancer with smoking of 1.3
30% increase in the likelihood of developing lung cancer among smokers as compared to non-smokers
Attributable fraction
Proportion of cases that would be avoided if the exposure is removed
Population
attributable
fraction
How much of the disease burden could be prevented by eliminating exposure
NNT
Number of patients that need to receive an intervention over a certain period to prevent one outcome
Mean
Average
Mode
Most common observation
Median
Middle observation
Quintiles /
Deciles
Data broken down into 5 / 10 sections, each having 20% / 10% of the values. e.g. first quintile is the point with 20% of the data below it and 80% above it.
Outlier
A value that is outside the expected range
Bias
Distorted truth due to error in selecting the study population or in measuring outcome/exposure
Validity
Test measuring what is intended to measure
Reliability
Consistency of the test
Effectiveness
Effect in real life condition
Efficacy
Effect in ideal conditions
Efficiency
Maximum output with minimum cost
Systematic
review
Systematically identifying, selecting and critically appraising of literature on topic of interest
Meta-analysis
A statistical method used to combine information from studies addressing the same research question to give an overall summary estimate
Blinding
Participants and/ or investigators don?t know to which intervention they have been allocated
Randomisation
Random allocation of participants of a study to the intervention or the control group
Intention to treat analysis
Analysis according to the groups that participants were allocated to, whether or not they actually received the intervention