Bias The introduction of a

systematic difference
between groups being
studied OR between those
selected for the study and
those who were not
Selection bias
o Systematic differences in the characteristics of the
groups being studied
o Systematic differences between selection and
control groups = decreased internal validity
o Systematic differences between those selected
for the study and those who were not =
decreased external validity
Measurement bias
o Systematic differences between groups in the way
data is collected
o Misclassification is a source of measurement bias
ONLY if the misclassification unevenly affects the
groups
Covariate Any variable that is associated with the outcome. These are often distinguished from
confounders only by the context (i.e. when a variable is placed in a multiple regression
model it is called a covariate, but when a variable is purposefully distributed evenly
between treatment and control so as to eliminate its effect on the outcome, it is a
confounder)
Confounder A variable that is
associated with the
exposure and also
influences the outcome
(but independently of the
causal connection between
exposure and outcome)

(Point) Prevalence Number of existing cases

Incidence Number of new cases in a
population, in a given time
interval

per time interval

Incidence Rate Ratio Incidence in population 1 :
incidence in population 2
(

) (

) per time interval

Absolute/attributable
risk
What is the difference in
risk between treatment
and control?

Relative risk How many times less or
more likely is it that the
primary outcome will occur
to a person in the
treatment arm, compared
to someone who is in the
control?

Number Needed to
Treat
How many people would I
need to treat with the
intervention, in order to
prevent one instance of the
primary outcome in a given
time interval?

Hazard A hazard rate is not
measured on its own,
because it is an
instantaneous incidence
rate for a small period of a
study thus, each hazard
rate is just one in a series.

in a given time interval (e.g. week1,
week2)
Exposure
Confounder
Outcome
causes
Hazard Ratio The ratio of hazard rate in
treatment versus control,
over the whole period of
the study (it is not simply
the ratio of the hazard
rates at one particular
point)
=how many times more/less probable the outcome is in the
treatment versus control (regardless of what point in the study we
look at)
Number at risk The number of people for
whom the outcome can
still occur (i.e. everyone
who is still in the study and
to whom the outcome has
not yet occurred)

Odds The ratio of exposed to
unexposed in a case
control study

Odds ratio The ratio of odds for those
with a certain outcome
versus those without (case
control study)

where x is the condition of interest (e.g. breast
cancer)
Type 1 error () False positive showing
an affect when there
actually isnt one
Given by the p-value. Generally, a positive result is accepted only if
this is less than 0.05 (i.e. the probability of this positive result being
a false positive is less than 5%)
Type 2 error () False negative showing
that there is no affect,
when there actually is one
(in a study that has a power of 85%, there is a
1-0.85=15% chance that a negative result will be a false negative
p-value The probability that the
result obtained is not due
to type 1 error (it is a true
positive)
Usually, we have no evidence to reject the null hypothesis if this is
greater than 0.05 (i.e. there is a greater than 5% chance that what
we really have is a type 1 error)
Sample size
calculation
(comparison of two
means)

[

Causation Temporality - cause precedes effect. e.g. smoking precedes lung cancer
Strength of association e.g. risk of lung cancer in smokers is over 20 times that of non-
smokers
Dose-response relationship e.g. heavier smokers more likely to develop lung cancer
Reversibility of the relationship e.g. quitting smoking reduces likelihood of developing lung
cancer
Consistency e.g. several studies at different times, in different settings have shown the
same association
Biologic plausibility e.g. compounds in cigarettes carcinogenic in vitro
Internal validity The extent to which the
results of the study validly
apply to the sample of people
being studied
Depends on:
Study design (minimisation of bias via things like
randomisation and blinding)
Analysis (appropriate use of analysis, analysis that takes
into account confounding, meets criteria for statistical and
clinical significance, low number needed to treat etc.)
External validity The extent to which the
results of the study validly
apply to the range of patients
in practice
Depends on:
Whether the PICOT question answered by the study
reflects the PICOT question concerning a patient
OR, can my PICOT question be answered using a subset of
the study sample (e.g. the average age of the study was 64
and my patient is 55, but the study also has a table where
the results are separated out for different age brackets)
PICOT Population
Intervention/exposure
Control
Outcome
Timing

Specificity The percentage of people
who truly do not have the
disease and test negative

Sensitivity The percentage of people
who have the disease and test
positive

Positive predictive
value
% people with positive test
that truly have disease

Negative predictive
value
% people with negative test
that do not have disease

Positive likelihood
ratio
Likelihood that a positive
result would be present in a
patient with the disease,
compared to a patient
without the disease

Negative likelihood
ratio
Likelihood that a negative
result would be present in a
patient with the disease,
compared to a patient
without the disease