Evidence based medicine

Question 1

Define Evidence based medicine

Answer 1

The use of mathmatecal estimates of the risk of benefit and harm, derived from high quality research of population samples, to inform clinical descision-making in the diagnosis, investigation or managemnet of individual patients.

Question 2

Describe the differences between observational and experimental studies

Answer 2

Observational studies are also called analytical studies, they can show and association between one variable and another and aim to identify causality.

Where as experimental studies (interventional studies are where a new treatment or intervention is trialed.

Question 3

Name and describe the type of observational studies used

Answer 3

Cohort study- looks for causality
Case control stuyd- looks for causality
Cross-sectional/longitudinal studies- useful for looking at trends.

Question 4

Define what bias is

Answer 4

Any trend in the collection, analysis, interpretation, publication or review of data that can lead to conclusions that are systemically different from the truth

This is a systematic deviation from the truth in one direction.

Question 5

What is an error?

Answer 5

This is different from bias as errors occur randomly and bias is systematic

Question 6

WHat is a confounding factor?

Answer 6

A factor that is associated with both the exposure of interets and the outcome of interest.

Question 7

What is selection bias?

Answer 7

Sample does not represent population

Question 8

What is detection bias?

Answer 8

Observations in treatment group pursued more than control group

Question 9

What is observer bias?

Answer 9

Subjectivity of observer, variance in their decisions

Question 10

What is recall bias?

Answer 10

Patients know which group they are in and may be more likely to report symptoms

Question 11

What is response bias?

Answer 11

Patients enrolling themselves/self-selecting

Question 12

What is publication bias?

Answer 12

Positive trials are more likely to be published.

Question 13

Define incidence

Answer 13

The number of new cases of disease in a population, in a given time period, also known as the occurence rate. Usually reported as a percentage
Helps us to understand the risk of disease

Question 14

Define prevelence

Answer 14

The total number of cases of a disease in a population, either in a time period or at a specific point in time.

Prevelence helps us to understand the burden of disease

Question 15

What is a cohort study and when is it used?

Answer 15

A cohort of people are studied based on exposure and followed up to evaluate an outcome of interest.

The can be prospective or retrosepctive

Adv: incidence can be calculated in exposed and non-exposed individuals, multiple outcomes can be studied, bias less of an issue than in case-control studies

Disadv: requires large populations and take a long time (years) to do so are expensive

Question 16

Explain what relative risk is

Answer 16

(outcome measure in cohort studies)

Risk in exposed group/ risk in unexposed group= RR

Question 17

What is a case control study?

Answer 17

A comparision between individuals with a disease (outcome) of interest (cases) ad those without the disease of interest. The cases and controls are each assessed to determine if they have had exposure to the variable of interest.

Almost always retrospective

Adv: smaller sample sizes generally required, quicker reults, cheaper

Disadv: more prone to bias, can be difficult to prove causation, not possible to calcaulate incidence, selecting controls can be difficult, particulalry prone to recall bias

Question 18

Define what odds ratio is

Answer 18

The ratio of odds of exposure of those with the outcome to those without the outcome

If it is 1 this suggests there is no difference is odds of exposure between the 2 groups

Question 19

What is a cross-sectional study?

Answer 19

Looks at the outcome and exposure in a ppulation or an individual at a specific point in time- they look at a cross section of society.

If a cross sectional study is repeated multiple time in succession then this is a longitudinal study.

They both look at trends in a population. The CAN’T look for causality.

Question 20

Give examples of types of intervention trails

Answer 20

The randomised controlled trail (RCT)- this is the gold standard, these have little chance of bias but are time consuming and expensive

Question 21

What does single blind mean?

Answer 21

Patients don’t know which treatment they are taking (e.g. in an RCT)

Question 22

What does double blind mean?

Answer 22

Neither the patients or the clinical/scientists involved in alaysis know which treatments they are on.

Question 23

What is a systematic review?

Answer 23

A review of a clearly formulated question that used systematic methods to identify, select and critically appraise relevant research and to collect and analyse data from the studies that are included in the review.

Question 24

What is meta-analysis?

Answer 24

The data/statistical part of a systematic review. Pulls together data from individual studies and analyses this. The data is usually presented in a forest plot.

You can have a systematic review without a meta-analysis.

Question 25

What are the highest and lowest levels of evidence?

Answer 25

Highest: 1++ high quality meta-analyses, systematic review of RCTs or RCTs with very low risk of bias

Lowest: 4 expert opinion

Question 26

What does an A grade of recommendation mean?

Answer 26

At least one meta-anlysis, systematic review or RCT rated as 1 ++ and directly applicable to the guidelines target population or a body of evidence composed or studies rated as 1+ and with an overall consistency among them

Question 27

What does a B grade of recommendation mean?

Answer 27

A body of evidence composed of studies rated as 2++, directly applicable to the guideline’s target population and demonstrating overall conistiency among them or evidence extrapolated from studies rated as 1++ or 1+

Question 28

What does a grade C of recommendation mean?

Answer 28

A body of evidence composed of studies rated 2+ directly applicable to the guideline’s target population and demonstrating overall consistency among them, or evidence extrapolated from studies rated as 2++.

Question 29

What does grade D of recommendation mean?

Answer 29

Level of evidence 3 or 4 or evidence extrapolated from studies rated as 2+.

Question 30

What are the Bradford-Hill criteria for causation?

Answer 30

Strength of association- e.g. statistical significance
Consistency- mulitple different studies using different populations, locations and methods show the same association then causality is more likely.
Specificity- exposure causes only one disease
Temporality
Biological gradient
Plausibility
Coherence
Experiment
Analogy

Question 31

What does the criteris consistency mean?

Answer 31

Consistency- mulitple different studies using different populations, locations and methods show the same association then causality is more likely.

Question 32

What does the criteria strength of associated mean?

Answer 32

Strength- the larger the magnitude and if the study is statistically significant then causality is more likely.

Question 33

What does the criteria specificty mean?

Answer 33

Used to mean exposure causes only one disease but not is less relevant as we know rather than A causing B it tends to be a combination of A,C,D and E having a part in increasing the risk of B

Question 34

What does the criteria temporality mean?

Answer 34

Exposure must proced the onset of disease

Question 35

What does the criteria biological gradient mean?

Answer 35

Dose-response raltionship, this strongly suggests causality

Question 36

What does the criteria plausibilty mean?

Answer 36

Existsing modles explain how exposure causes disease.

Question 37

What does the criteria coherence mean?

Answer 37

similar to Plausibility

The association should make overall sense

Question 38

What does the criteria experment mean?

Answer 38

Manipulation of exposure should effect disease/outcome

Question 39

What does the criteria analogy mean?

Answer 39

When one causal agent is known, standards of eveidence are lowered for a second agent that is similar.

Question 40

What is the audit cycle?

Answer 40

1. preparing for audit
2. selecting criteria and standards
3. measure what is happening now and report on
4. make improvements
5. sustain improvement and re-aduit

Then go back to 1