Lectures 1&2 - The importance of evidence in the practice of medicine Flashcards Preview

YR1 Epidemiology in Practice > Lectures 1&2 - The importance of evidence in the practice of medicine > Flashcards

Flashcards in Lectures 1&2 - The importance of evidence in the practice of medicine Deck (19)
Loading flashcards...

what is evidence based medicine?

the use of critically appraised info to determine strength of evidence for the use of treatments and medicine in clinical practice


outline the different stages of clinical medicine where EBM can be of use

CLINICAL FINDINGS- use of patient history and physical exam
AETIOLOGY - identifying cause of disease
CLINICAL MANIFESTATIONS OF DISEASE - how often & when a disease causes clinical problems
DIFFERENTIAL DIAGNOSIS - selecting likely, serious and treatment-responsive causes of a problem
DIAGNOSTIC TESTS - selecting and interpreting tests to confirm/exclude diagnoses
PROGNOSIS - estimating a patient's clinical course and anticipating complications
THERAPY - selecting treatments that are cost effective and where good > harm
PREVENTION - reducing chance of disease by identifying and modifying risk factors and use of screening in early diagnosis


what are the criticisms of EBM?

- not enough time for doctors to critically appraise info
- EBM has been inappropriately used by the gov't to justify decisions that clinicians disagree with


what is the importance of EBM?

- better service to patients
- better patient care & safety
- increased medical knowledge
- revalidation - clinicians are up to date with evidence
- professionalism
- better interpersonal & communication skills


list the order of the hierarchy of studies

1) meta-analyses & systematic reviews
2) RCTs
3) cohort studies
4) case-control studies
5) cross sectional studies
6) animal trials & in-vitro studies
7) case reports, opinion papers, letters


describe systematic reviews & meta-analyses

- systematic reviews answer a defined research question by collecting and summarising evidence
- meta-analysis refers to use of statistical techniques to integrate the results of studies that match the eligibility criteria
- cheap
- avoids issue of large sample size as data for the study is pooled from many smaller studies


describe RCTs

- strong inclusion/exclusion criteria
- requires a large sample size
- expensive


describe cohort studies

- involves a group of people before they develop a condition
- exposures and risk factors are observed
- group is followed up over a period of time to see who ends up suffering from disease
- more effective with common diseases
- less prone to bias
- can be prospective or retrospective


describe case-controlled studies

- involves people suffering from disease and people not suffering from disease (control)
- more useful for rare conditions
- quick and cost-efficient
- can investigate many exposures simultaneously
- selection bias
- recall (of information) bias
- poor for rare exposures


describe ecological studies

- descriptive/observational study
- used to measure prevalence and incidence of disease in different populations, particularly when disease is rare


describe cross sectional studies

- routinely collected data that helps to describe the status of individuals with respect to absence/presence of both exposure and disease assessed at the same point in time
- hard to establish causal effect


describe case reports

- description of single case
- not used to support clinical practice
- may be useful in picking out new syndromes/conditions


what is association?

the statistical dependence between 2 variables - the degree to which the rate of disease in persons with a specific exposure is higher/lower than the rate of disease without the exposure


what 4 things can be used to evaluate statistical association?

Confounding factors
Causal effect


how is chance used to evaluate statistical association?

- how big is the sample size?
- use of power calculations - minimum sample size required so that you can likely detect an effect of a given size
- p values - p < 0.05 = low p that it's due to chance = significant
confidence intervals - range within which the 'true' value is expected to lie with a given degree of certainty


how is bias used to evaluate statistical association?

bias is a consequence of defects in design/execution (systemic error)
- selection bias - selection of participants
- measurement bias - measurements/classifications of disease/exposure


how are confounding factors used to evaluate statistical association?

factors that distort risk of disease but do not alone cause disease
- age
- sex
- socio-economic
- geography


how is causal effect used to evaluate statistical significance?

judgement of a cause-effect relationship based on logic -> observed association between exposure and disease -> sufficient evidence


outline the factors to consider in the Bradford-Hill Criteria for causation

ANALOGY - there is a source of more elaborate hypotheses about the association
STRENGTH - of association, measured by magnitude of relative risk
SPECIFICITY - if an exposure increases the risk of 1 disease but not many diseases
EXPERIMENTAL EVIDENCE - evidence on humans/animals
TEMPORAL RELATIONSHIP(!) - risk factor causes disease if it precedes the disease
CONSISTENCY - similar results in different populations using different study designs
DOSE-RESPONSE RELATIONSHIP - increased exposure = increased risk of disease
PLAUSIBILITY - association is more likely to be causal if consistent with other knowledge
COHERENCE - information is not conflicting

All Students Study Epidemiology To Continue Doctors' Practices Coherently