descriptive epidemiology Flashcards
literature review
has two main approaches
narrative
and systematic
narrative review
brings together published literature into a single article to allow readers to understand issues
aka literature review,scoping review or non systematic review
systematic review
highly structured approach to search and sift and summarising literature
narrative reviews stregnths and weakness
easier and faster to write
more up to date
useful when looking at areas of limited research
can be usefuk when work from different disciplines is being bought together in an easily answerable research question
disadvantage
subject to bias as authors select work to support their opinion
no search is specified
systematic reviews
aim to collate all evidence that relates to a highly focused research question
highly specified protocol
includes evidence based on pre specified data (inclusion criteria)
can take many months to design the search
disadvantages
a less comprehensive search strategy may mean you miss out on evidence
only as good as indices used
only as good as evidence incorprated
very quickly out of date due to delay in publishing,look at search date not publication date
published every 2-10 years
method of systematic review
research quesrion
structured search
indices
screening
reporting
writing
submitting
structured search
combine searches together
indices
medlline
mbase
psychinfo
need to explain which indices you will use
based on published research
registeries
registration of research yet to be completed
important to avoid duplication or ommision
screening
PRISMA diagram
shows how many articles have been found or removed due to indices being the same
how many articles will be involved
grey information
many implementations and programmes have been reported but with varying degrees of accuracy
eg on google scholar
meta analysis
a quantitative formal epidemiological study desgin used to systematically assess previous research studies to derive conclusions about that body of research
combines quatitatve findings from separate studies into a pooled estimate
forest plot
used in meta analysis
contains
study
intervention and control
relative risk
diamond is a combination of all the relative risks
if confidence interval crosses 0 we cant reject the null hypothesis
heterogeneity
variation in a studies outcome
clinical-within patients and selection criteria
methodological- study designs blinding etc
statistical;reporting differences
weight
usually proportional to study size
fixed and random effects models may give you different weights
funnel plot
shows balance of evidence between studies assuming an overall side effect
more publications on one side of the line may mean that the studies havent been reported
implications of meta analysis
limited by quality of studies used
will need a systematic review
may need to go back to original authors to ask for more data
two types of efficacy endpoints
Primary endpoint – this is the endpoint for which the study has been powered; that is to say that the number of trial participants (sample size) will have been recruited on the basis of the pre-specified power and difference.
Secondary endpoint – it is common that a study will want to examine a slightly different endpoint in addition to the primary endpoint. For example, while a study seeks to examine survival (i.e. alive or dead) another – often ‘softer’ - measure such as recurrence of disease or hospital admission might also be measured. If the secondary endpoint is proven but the primary endpoint is not, then the findings of the study may still contribute to the understanding of disease.
safety endpoints
Safety endpoints are quite intuitive. On the one hand it could be anaphylaxis or direct mortality associated with the therapy. Such major issues should usually be detected early in the trial process (before its rolled out to large numbers of patients). But more commonly the safety endpoints will be more nuanced: potentially measuring commonly observed adverse events (AEs) and grading them into a hierarchy of significance. A large proportion of patients reporting AEs will require investigation.
composite endpoint
can describe any endpoint
multiole potential endpoints have been added together
surival analysis
use kaplan meier plots
bradford hill cirteria
strength-a stronger association increases the confidence that an exposure causes an outcome
consistency-consistent findings rule out errors that might occur in studies
specificity-a disease arising among specific workers is valuable in supporting argument for casuality but concedes lack of specificty doesnt invalidate casual relationship
temporality-an exposure must precede an outcome
biological gradient-theres a dose response relationship between exposure and outcome
plausibility-relationship shouldnt be implausible
coherance-ensures that association is in line with existing science
experiment-not ethical
analogy-rests on exsisting knowledge
internal validity
an association exists in an experiment
the extent to which findings accurately describe the relationship between exposure and outcome in the context of the study
external validity
findings from an experiment can be generalised to other people
selection bias
when a person has been exposed or had the outcome
eg healthy worker effect/non response bias
how to avoid selection bias
minimise non response
control representative of target population
compare respondents with non respondants
information bias
misclassification of exposure or disease status or both
includes recall,response,interviewer and diagnostic bias
interviewer bias
more likely to ask patient questions is they have the outcome about the expposure
recall bias
pt may forget past exposures
non differential missclassification
when exposure status is misclassified between everyone
odds ratio will always be towards null (differntia can be towards either)
confounding
the effect of the extraneous variable that wholly or partially accounts for the apparent effect of the study,exposure or that masks an underlying true association
criteria for confounding
1.knowledge about subject matter
2.3 conditions-its assocaited with the exposure in the source population,associated with the outcome in the absence of the exposure,not a consequence of exposure
3.stratification:look at difference of apparent affect within different population strata
4.compare crude and adjusted estimates
effect modification
when the strength of an association varies over different levels of a third variable eg gender
natural phenomenon
present stratified results eg by gender
test for effect modification
breslow day test
q test
interaction terms in regression models
synergism and antagonism
synergism is where the effect modifier potentiates the effect of the exposure
antagonism is where the effect modifier dimishes the effect of the exposure
crude vs adjusted model
Crude model (which is the univariate analysis of exposure vs outcome). This is simply looking at the impact of the exposure on the outcome – with no consideration of anything else.
Adjusted model (which is the multivariate analysis of a range of exposures vs. outcome). A multivariate analysis means that multiple potential exposures have been included. The inference is that the outputs of these analyses mean that holding all other adjusted variables equal, X is the association between exposure and outcome.
