Research Designs, Levels of Evidence and Reporting Appraisal Flashcards
What is an Intervention study used for?
When researchers want to test the effectiveness of a treatment on a measure of patient health, wellbeing or function, or any measureable outcome.
What are observational studies?
These are used to research how disease and death occur in populations and to identify risk factors for good and bad health.
What is a variable?
It is any quality, characteristic or feature that is observable and measurable, and can have more than one value or result.
What is an independent variable?
During experimental research, it is the treatment or the intervention, which is presumed to cause a treatment effect.
What is a dependent variable?
During research, this is generally the outcome, and what is presumed to be affected by the treatment.
What does co-vary mean?
It refers to when the variables on one value change, so too do values on the other
Summarise the effect of variables
Its the effect of treatment (independent variable) on the patients condition (dependent variable)
What is a case series study design?
This involves giving a treatment to a group of participants (or one person) and then measuring the outcome. e.g headache pill and testing the severity
What differences are in an experimental design?
This improves on a case series, by adding an EXTRA group that does not receive the intervention (usually nothing, or placebo)
What is a non-randomised trial?
This is a NHMRC level 3 evidence, and has two groups. Each group is assessed twice on the outcome; pre and post, with one group (treatment) receives the treatment while the other (control group) receives nothing/placebo
What is a randomised control trial?
Similar experiment design as a non-randomised, but participants are allocated to intervention or control randomly, allowing equal change to be in either.
What is an observational study?
A variant of quantitative study design where there is no deliverate intervention by the researchers, often passively observing real world changes or effects.
What are three types of epidemiological observational studies?
Cohort study, Case-control study, and Screening Studies.
Diagnosis and assessments must be accurate because?
It is essential that patients are getting the correct treatment.
What do prognostic studies provide?
They provide evidence about how diseases progress, again enabling clinicians to make better decisions.
What is bias?
Any threat to validity that could mislead interpretation of the results.
Is bias systematic? T/F
TRUE; generally operating in a consistent rather than random way.
What does the levels of evidence indicate?
The amount of risk or bias in the research/reporting. Lower number = lower risk/bias
What is level 1 evidence?
A systematic review of level 2 studies
What is level 2 evidence?
A randomised control trial
What is level 3-1 evidence?
A non-randomised control trial
What is level 3-2 evidence?
A comparative study with concurrent controls (non-randomised experimental trial, cohort study, case-control study and interrupted time series w/ control group)
What is level 3-3 evidence?
A comparative study without concurrent controls (historical control study, two or more single aim studies, interrupted time series…)
What is level 4 evidence?
Case series with either a post test or pre test/post test outcomes
Does the levels of evidence eliminate all bias? T?F
FALSE; its only one aspect of bias in research
What is critical appraisal?
The process of assessing and interpreting evidence, by systematically considering its validity, results and relevance to your own context.
Why are causal relationships important?
They help us understand what causes disorders, whether treatments improve health
What is an inference?
This is a decision or conclusion about truth based on evidence
What is a causal inference?
These are inferences about a causal relationship (concludes that one event causes another)
What are the three principles of causality?
Contiguity in time and space, Causes precede effects and constant conjunction.
Is cause and effect good proof?
No, as it cannot be proven because observation alone gives us no way to show that constant conjunction will always happen.
Explain necessity and sufficiency (mills method)
A causes B because B happens if and only if A happens.
Expand on method of agreement?
Sufficient! means its enough to A to happen for B also to happen
Expand on methods of difference
Necessary! means that A must happen for B to happen
When does a direct casual relationship happen?
When one event causes another without any intervening events
When does indirect causation happen
when a chain of events separates an initial cause from a final cause, often happens in development of disease or major accidents.
Explain a spurious causation
This occurs when one event appears to cause another event but really it doesnt. These relationships are a particular risk to cause inferences in healthcare
What is a moderator effect?
Moderator realtiionship occurs when the strength of the relationship between two events depends on another event.
What is a moderator effect?
Moderator realtiionship occurs when the strength of the relationship between two events depends on another event.
What are the principles for establishing causation from evidence?
Time, Strength of association/correlation, dose-response for clinical trials, replication of findings, plausability and When cause removed or not present, effect does not happen (as per mill).
What is aetiology?
The cause of disease
What does CONSORT stand for?
Consolidated Stands of Reporting Trials
