EBP

What is the aim of medicines information?

To support the safe, effective and efficient use of medicines by the provision of evidence-based information and advice on their therapeutic use

What are the 5 roles of MI centres?

1. An enquiry answering service

2. Proactive provision of information and advice

3. Education and training

4. Planned introduction of new medicines

5. Quality assurance

What are the stages of answering an MI enquiry?

1. Understand the question

2. Carry out your research

3. Prepare your answer

4. Feedback your answer

Should you document every stage of an MI enquiry?

Yes

Define point prevalence

The proportion of people who have a disease at a specific point in time

Define period prevalence

The number of people who have the disease in a time period divided by the population at the mid-point of the period

What numbers will prevalence be between? What is the significance of each number?

• Will always be between 0 and 1

• Can be given as a percentage

• The closer to one the larger the proportion of cases there are within the population

Why use the mid-point population when calculating period prevalence?

• To account for fluctuations in population over the time period

• Birth, death, migration

What are the two main categories to measuring disease frequency?

• Incidence

• Prevalence

Why may prevalence of a disease change overtime?

• Changes in incidence

• Changes in disease duration

• Intervention programmes

• Changing classifications

What are the two main factors that prevalence depends on?

• Disease incidence

• Duration

How is the prevalence of a disease found?

from cross-sectional surveys or cohort studies

How can the risk of a disease be expressed?

As a probability outcome, hazard of outcome or incidence of outcome

What are the ways to compare the effect of diseases?

• Absolute measures

• Relative measures

What is the difference between the absolute measures and relative measures?

Absolute risk gives the actual risk/ number of cases whereas the relative risk compares this between two groups, usually to give a ratio

What is another name for the relative risk?

Risk ratio

What is the risk ratio?

Probability that an individual will be diagnosed with outcome over the follow up period

incidence exposed/ incidence unexposed

What does NNT stand for?

Number needed to treat

Define NNT

Number of patients who must be treated to achieve one more success than an old treatment

How do you calculate NNT?

1 divided by absolute risk difference

always round up to the nearest whole number

How do you know what NNT means?

1 - new treatment is always effective

Increasing NNT indicates less effective treatment

What does it mean if NNT is less than 1?

the new treatment is worse than the old treatment

could potentially be harmful

What does NNH stand for?

Number needed to harm

the number of people who when treated results in one additional harmful outcome

How do you calculate NNH?

1/ absolute risk difference

always round down to nearest whole number

When you should you use an absolute risk over a relative risk?

• If the risk of an adverse event is very small, expressing it as a relative risk could make it appear more serious and bias the perceived risk upward

Why use hazard?

A way to quantify differences in survival

May change over time - ratio of hazards will remain constant

What confidence intervals should results have?

a 95% confidence interval

Define 95% confidence interval

If you repeated the study 100 times, on 95 of those times, the point estimate would be within the confidence interval

Outline the reasons to use clinical trials

• Designed to determine efficacy

• Collect efficacy and safety data on drugs and devices

• Are the first time new drugs are tested in patients

• Are limited by a critical element

Describe phase I clinical trials

• first application of new treatment in humans

• generates preliminary PK/PD data

• usually has 20-50 participants

• usually single dose or only a few weeks of treatment

Who are phase I clinical trials usually conducted in?

• healthy subjects when expected toxicity is minor

• extremely ill patients if therapy is toxic

Describe phase II clinical trials

• first time patients are exposed to the drugs

• small study of efficacy

• provides experience with treatment and administration

• sample size less than 100 people

• outcome is a measure of treatment efficacy

• aims to find out more about PK/PD and common adverse reactions

• determines daily dosage and treatment regimen to be tested more rigorously in phase III

Describe phase III clinical trials

• to get a definitive assessment of intervention against a control

• evaluates an intervention’s toxicity and efficacy

• between 500 and 3000 patients exposed

• interventions can include new drugs, devices and treatment methods

What are the advantages of a double blind trial?

Protects against information bias and neither patient or physician/ researcher knows the treatment being taken

Why use randomisation?

• comparable groups with respect to measured and unmeasured risk factors

• any association is either casual or the result of chance

What do the inclusion and exclusion criteria?

can affect the extent to which the results of the trial can be generalised

Which patient groups are there few trials?

• Women - pregnant women

• Children

• Older people

• People with heart disease

• Alcoholics

• Drug abusers

• People with learning disabilities

• Epileptics

How do you summarise results for a systematic review?

Descriptively

How do you summarise results for a meta-analysis?

Quantitatively - together with checks for similarity

What are the sources for a literature search?

• Pubmed

• Cochrane

• Web of science

• Trial registries

• Pubmed MESH database

What does PICO stand for?

P - population

I - intervention

C - comparison any non-drug intervention, placebo, support or behavioural intervention

O - outcome

What do papers need to include for a literature review?

• RCTs or non-randomised comparative trials

• Specific age range

• Correct criteria at the start of the intervention

What is meta analysis?

statistical methods for combining the results of a number of studies

How can results be combined for a meta analysis paper?

• test heterogenity

• if there is heterogenity then use a random effect meta analysis

• heterogenity is often tested with Cochran’s Q test which assumes the true effect doesn’t vary between studies

• Q finds if there is a significant difference between each study’s result and the fixed effect result

• Q increases with increasing numbers of studies and precision

What is the random effects model with regards to meta analysis?

Adds an extra term into the weight to account for between study variability

Random effects weights are smaller and more similar to each other so smaller studies have a greater relative weight and wider confidence intervals

What are the difficulties with literature studies?

• Are all relevant studies included

• What about information that didn’t get published

• Information has been published but not enough details

• Selective reporting of one major result instead of all results

What are the advantages/ disadvantages of observational studies?

Advantages:

• Non-experimental no risk to patients

• Can select subjects included

Disadvantages:

• No control over exposures assigned

• Confounding will occur - results may be confused

What is the difference between a fixed and a dynamic cohort?

Fixed is when members of a cohort can only join at the start whereas a dynamic cohort they can join during the time period

What is incidence rate?

The number of new cases of the disease within a specific time period

Why must exposure be clearly defined?

misclassification may lead to an under- or over- estimation between exposure and outcome

What are the two types of misclassification?

• Non-differential: to the same degree for all groups (under-estimate)

• Differential: different between groups (under or over)

What might influence the results of a cohort study?

• Age

• Distribution of gender included

• Other co-morbidities

• Other medications

• Smoking

• BMI

• Socioeconomic status

etc…

What should you implement at design stage of a cohort study to reduce confounding?

• Randomisation

• Restriction

• Matching - match characteristics in the exposed and non-exposed groups so that there’s a more even distribution

• Stratified/ multivariate analysis

What are the advantages/ limitations of cohort studies?

Advantages:

• Efficient for rare exposures

• Multiple effects can be studied

• Less prone to bias

• Can calculate incidence rates and relative risks

• Temporal relationship easier to establish

Limitations:

• Can have difficulties with loss to follow-up

• May have problems with bias

Compare and contrast case-control studies and cohort studies

In case-control studies, the starting population already has the outcome disease and this is compared to a control group who don’t but are exposed whereas in a cohort study, the population starts with the risk factor/ exposure

Case-controlled studies compare cases vs controls (disease vs no disease) whereas cohort studies compare exposed vs non-exposed

Case-control studies look backwards to find exposure whereas cohort studies look forward to see outcomes

Case-control studies are better for rare diseases whereas cohort studies are better for rare exposures

Both have some degree of potential bias

What are important criteria for the control population in a case-control study?

• have to not have the disease but are otherwise comparable to the cases

• must be at risk of outcome but also not have the outcome under study

What are the types of bias?

• Information bias: interviewer bias, misclassification of exposure, recall bias, measurement bias

• Selection bias

What kind of error does bias cause in a study?

Systematic - can result in over and under-estimates

What analysis method do you use for a case-control study?

The ODDS ratio

OR = (odds of a person in group 1 having the event) / (odds of person in group 2 having the event)

= (a/c)/(b/d)

<1 = reduced risk in group 1

>1 = increases risk in group 1

=1 = no difference in groups

What are the advantages and disadvantages of case-control studys?

Advantages:

• Results are available quickly and usually fairly cheaply

• Good for rare diseases or those with long latent periods

• Look at multiple aetiologic factors for a single disease

Disadvantages:

• Inefficient for rare exposures - need to have some chance of exposure in the case and controls

• Can’t calculate incidence or prevalence of disease

• Difficult to establish temporal relationships

• More prone to bias than cohort studies

Define confounding

The distortion of a risk estimate due to the mixture of the people in the study population

What are Hill’s aspects of association?

• Strength of association

• Consistency - if there is a casual association then the same association should be found in all investigations

• Specificty - disease should only occur in people exposed to the suspected agent (rarely occurs more likely to have multi-causation)

• Temporality - exposure must preced onset of disease

• Biological gradient

• Biological plausibility - should be consistent with the known biological activity of the suspected agent

• Coherence - causes and effects for an association does not conflict with known natural history and biology of the disease

• Experimental evidence - reproducible in animal models? Challenge/ re-challenge?

• Analogy - is the association similar to other well-known associations?