The role of study design in addressing potential bias(es) Flashcards
Clinical studies require you to balance what three competing priorities
1) design + conduct a study that provides the correct answer
2) maximise the efficient use of resources
3) comply with ethical, legal, constitutional + professional regulations.
what are the 4 sources of potential bias
1) sampling/ selection bias
2) measurement bias
3) analytical bias
4) dissemination bias
what are the specific types of sampling bias
- external validity- non representative samples.
- confounding- selection influences exposure and outcome.
what are the specific types of measurement bias
- information bias- extent of information varies across participants.
- observer bias- influenced by prior knowledge/ belief.
- recall/ response/ prestige bias- influenced by prior knowledge or belief.
what are the specific types of analytical bias
- loss of follow up= specific participants
- omitted variable- imprecise adjustment for confounding.
- attributional bias- intereptation of causality
what are the specific types of dissemination bias
- publication bias- eventful results more likely to be published.
define a cross sectional study
- provides evidence of association within a sample.
- snapshot
- retrospective
define case control study
- provides evidence of association between sampled
- retrospective
- compares 2 groups one with and one without disease and works back to see what exposures they had
define cohort study
- provides evidence of directionality of associations
- prospective study
- analyses 2 group of people and see who develops disease and relates this to the exposure they were faced with.
define trial
-provides evidence of causality
define met analysis
- Provides evidence of reproducibility/generalisability
what is the hierarchy of analytical studies for evidence of effect.
- metanylasis (best)
- trial
- cohort
- case-control
- cross- sectional. (worse)
define deductive (analytical) reasoning
theory -> hypothesis -> testing -> observation -> reject/not reject.
define inductive (descriptive) reasoning
observation -> pattern -> tentative hypothesis -> theory
2 types of inductive studies
case study
cross- sectional