Why evidence based medicine

Question 1

why does EBM matter to clinicians

Answer 1

Revalidation
Medical knowledge
Practice-based learning and improvement
Interpersonal and communication skills
Professionalism

Question 2

what are possible explanations for observed associations

Answer 2

chance, bias, confounding, causation

Question 3

what is chance (coincidence)

Answer 3

Make inference from samples rather than whole populations , therefore we are estimating. Sample size. Power calculations. Probability (P) values and statistical significance (p=1/20. The role of chance can be assessed by appropriate statistical significance tests or a confidence interval (rage of values in which true value lies)

Question 4

Use of EBM 

Answer 4

Clinical findings. Aetiology- causes of disease. Clinical manifestations of disease. Differential diagnosis- and selecting which are most likely. Diagnostic tests. Prognosis- likely clinical course of events after diagnosis. Therapy- select most effective treatments. Prevention- identifying and modifying risk factors, and how to diagnose disease early by screening

Question 5

what is the heirachy of studies

Answer 5

Systematic reviews and meta-analysis. Randomised controlled trials. Cohort studies. Case-control studies. Ecological studies. Descriptive/cross-sectional studies. Case report/series

Question 6

define association

Answer 6

refers to the statistical dependence between two variables. Consider chance, bias, confounding, cause (CBCC)

Question 7

what is bias

Answer 7

Systematic error. Consequence of defects in the design or execution of the study. Selection bias e.g. Researcher choosing preferable patients, loss of follow up, non-response
Measurement bias e.g. Faulty equipment, recall. Observer bias – Prevented by Double Blinding. Responder bias – Prevented by Single Blinding

Question 8

define confounding

Answer 8

mixing of effects between exposure, the disease and a third factor.Account for confounding using matching, randomisation, stratification and multivariate analysis

Question 9

what is a potential confounder

Answer 9

any factor which is believed to have a real effect on the risk of the disease under investigation and is also related to the risk factor under investigation

Question 10

what are common confounders

Answer 10

age, sex, socio-economic status and geography

Question 11

what is causal-effect

Answer 11

judgement of a cause-effect relationship based on a chain of logic that addresses two main areas: Observed association between an exposure and a disease is valid. Totality of evidence taken from a number of sources supports a judgement of causality. Must eliminate chance, bias and confounding before considering causal effect

Question 12

what are cohort studies

Answer 12

carried out over a period of time - prospective

Question 13

what are case control studies

Answer 13

compares cases to controls. retrospective. information on past exposure to possible risk factors is obtained for cases and controls, and exposure in the cases is compared with that in the controls.

Question 14

what is a case report

Answer 14

follow up of individual patient

Question 15

what is an Ecological study

Answer 15

observationalstudydefinedby the level at which data are analysed, namely at the population or group level, rather than individual level. Ecological studiesare often used to measure prevalence and incidence of disease, particularly when disease is rare.

Question 16

what are descriptive/cross sectional studies

Answer 16

Observationalstudythat analyses data collected from a population, or a representative subset, at a specific point in time

Question 17

list the Bradford-Hill criteria for establishing causation

Answer 17

strength of association. consistency. specificity. temporal relationship. dose-response relationship. plausibility. coherence. experimental evidence. analogy

Question 18

what is strength (Bradford hill criteria)

Answer 18

Measured by the magnitude of relative risk. Strong association is more likely to be causal and a weak association could be more easily the result of confounding or bias(but doesn’t rule out a causal relationship). E.g passive smoking and lung cancer

Question 19

what is consistency(Bradford hill criteria)

Answer 19

Similar results found in different studies since it is unlikely that they were all subject to the same type of errors. However, lack of consistency does not exclude a causal association since different exposure levels, different populations using different designs may reduce the impact of the causal factor

Question 20

what is specificity (Bradford hill criteria)

Answer 20

A particular exposure increasing the risk of a certain disease but not the risk of other diseases shows strong evidence in favour of a cause-effect relationship e.g. Mesothelioma. However, one-to-one relationships between exposure and disease are rare. Lack of specificity should not be used to refute a causal relationship; e.g cigarette smoking causes many diseases.

Question 21

what is temporal relationship (Bradford hill criteria)

Answer 21

Essential criterion. For a putative risk factor to be the cause of a disease, it has to precede the disease. Generally easier to establish from cohort studies but rather difficult from case-control or cross-sectional studies when measurements of the cause and the effect are made at the same time.

Question 22

what is dose response relationship (Bradford hill criteria)

Answer 22

Increasing levels of exposure lead to increasing risks of disease shows evidence of a causal effect. Some however show a jump rather than a monotonic trend.

Question 23

what is plausibility(Bradford hill criteria)

Answer 23

Association is more likely if consistent with other knowledge e.g animal experiments, biological mechanisms. Should not be taken too seriously because lack of plausibility may just reflect lack of scientific knowledge.

Question 24

what is coherence (Bradford hill criteria)

Answer 24

Cause and effect interpretation does not conflict with what is known of the natural history. However, absence of coherent information (e.g. conflicting information) should not be taken as evidence against an association being causal.

Question 25

what is experimental evidence (Bradford hill criteria)

Answer 25

On humans (rarely available) or animals

Question 26

what is analogy (Bradford hill criteria)

Answer 26

Provides a source of more elaborate hypotheses about the association in question. Absence only reflects lack of imagination or experience.