Experimental study design Randomised Controlled Trials MSc Physician Associate Practice Master’s Phy Flashcards
Descriptive epidemiology tends to focus on?
generating hypotheses
Analytical epidemiology tends to focus on?
Testing hypotheses
What is descriptive epidemiology?
Descriptive Epidemiology is the study of the amount and distribution of
disease or other health-related characteristics in human populations by
person, place and time.
Descriptive epidemiology identifies who is affected,
when, and where the situation is occurring in the community or population
of interest. This is an important tool for health services planning and
programming.
What is an experimental study?
A study in which conditions are under the direct control of the investigator.
In epidemiology a study in which a population is selected for a planned trial
of a regimen whose effects are measured by comparing the outcome of the
regimen (e.g. new medication) in the experimental group with the outcome
of an alternative regimen (standard) in a control group.
To avoid bias in experimental studies what should take place?
To avoid bias, members of the experimental and control group should be
comparable in all variables except in the regimen that is offered them.
Allocation of individuals to experimental and control groups is ideally by
randomisation.
Most common experimental studies
- Randomised controlled trial (individuals)
- Community trials (communities, clusters)
Since the interventions are strictly determined by the study protocol, ethical
considerations are of paramount importance in the design of these studies. True or false?
True
What experimental study is the most reliable study design? Why?
Randomised controlled trial.
Most reliable study design due to randomisation and control of all other
variables except for the intervention (eliminates confounding)
• Regarded as the golden standard among primary study designs
Golden standard primary study design?
RCT
In RCT, groups are followed up to see how effective the experimental intervention was. Outcomes are measured at specific times and any difference in response between the groups is assessed statistically.
This method is also used to reduce ?
Bias
What does H0 tell?
Null hypothesis – almost always no difference between the two treatments
What does H1 tell?
Alternative hypothesis
What do all hypotheses tests use to weigh the strength of the
evidence (what the data is telling you about the population)?
P- value
What is the p value.
The p-value is a number between 0 and 1 and interpreted in the following
way:
– A small p-value (typically < 0.05) indicates strong evidence against the null
hypothesis, so you reject the null hypothesis. Thus you accept the alternative
hypothesis
– A large p-value (typically > 0.05) indicates weak evidence against the null
hypothesis, so you fail to reject the null hypothesis. Thus you accept the null
hypothesis
P values indicate the believability that the null hypothesis is true given the sample
data. True or false?
True
Why are P-values are an integral part of inferential statistics ?
P-values are an integral part of inferential statistics because they help you use your
sample to draw conclusions about a population.
Statistically significant means the difference is large or
scientifically / clinically important, true or false?
False
What type of question is statistical significance?
Mathematical
What type of question is a Clinical importance?
- Requires clinical judgment
– Must be considered & addressed before it is possible to ask about
statistical significance
Randomisation should equally distribute any
confounding variables between the two groups, although it is important to be
aware that differences in confounding variables may arise through?
chance
Randomisation ensures each individual (or each group in the case of cluster
randomisation) ?
Has the same chance of being allocated to either group
What prevents (conceal) advanced knowledge of what intervention
group people have been assigned to prior to the start of the RCT?
Concealment
Concealment should be done by?
Someone who is not responsible for recruiting people into the trial
This process prevents researchers from (unconsciously or otherwise)
influencing which people are assigned to a given group, intervention or
control
Concealment
Randomisation performance:
Coin toss: heads = control, and tails = intervention/treatment. Throwing a dice: three or under three = control, and
over three = intervention/treatment. A shuffled deck of cards: odds, and evens. Drawing allocations out of a hat.
Simple randomisation
Randomisation performance:
Maybe appropriate for smaller trials to ensure equal numbers in each group. With two treatment groups, block size
of either 4, 6 or 8
Blocked randomisation
Randomisation performance:
Ensures that a potential baseline confounding variable is equally distributed between the two groups is achieved by
generating a separate block for each combination of covariates, and subjects are assigned to the appropriate block
of covariates. Consider genetic variations as an example
Stratified randomisation
Randomisation performance:
(e.g.pharmacy) The clinician contacts this facility by telephone when he is ready to randomise the next patient;
the initials and study number of the patient are read to the person performing the randomisation, who records it
and then reads back the randomisation for that subject. Appropriate for large, multicentred trials
Use of remote randomisation facility
Analysis of the results should occur based on the initial randomisation, irrespective of what may subsequently have
happened to the subject (this is – intention to treat analysis). E.g. rescue mechanisms
True or false?
True
Advantages of RCT: minimised risk of bias
Participants are similar at the study entry
Protocols, Clinical Trial Registry, PICOS concept, strictly defined eligibility criteria
Matching control group: All baseline characteristics are the same (no statistical
differences of outcomes between the groups at the beginning of the study – baseline)
All variables are kept the same (except the treatment)
Groups differ only in terms of the treatment they receive (intervention & comparison)
which minimizes the confounding factors
Randomisation & concealment of allocation ensures equal distribution of any
confounding variables between the two groups. Once the patient has given consent to be
included in the trial, he/she is then irreversibly randomised by opening the next sealed
envelope containing his/her assignment.
Blinding or masking is a way to prevent researchers, patients (receiving the
treatment) and healthcare professionals (assessing outcomes) from knowing
which study group each patient is in so they cannot influence the results.
Type of blinding:
One of the parties is blinded to treatment e.g. patients
do not know which study group they are in (for example whether they are
taking the experimental drug or a placebo), or researchers
Single-blind study
Type of blinding:
neither the patients nor the
researchers/clinicians know which study group the patients are in.
Double-blind
Type of blinding:
the patients, clinicians and the people carrying out the statistical
analysis do not know which treatment patients had.
Triple blind study
Procedure where masking is not done – masking can not always be performed, but
use of objective outcomes (rather then subjective ones) can ensure that lack of
masking does not undermine validity – surgical interventions, for example
Open-label
Sources of bias in RCT domains:
Sequence generation
Selection bias
Sources of bias in RCT domains:
Allocation concealment
Selection bias
Sources of bias in RCT domains:
Blinding of participants and personnel
Performance bias
Sources of bias in RCT domains:
Blinding of outcome assessment
Detection bias
Sources of bias in RCT domains:
Incomplete outcome data
Attrition bias
Sources of bias in RCT domains:
Selective outcome reporting
Reporting bias
What are ethical issues?
Ethical issues are those involving actions and policies that are
right or wrong, fair or unfair, just or unjust
All epidemiological studies must be reviewed and approved by
ethical review committees. True or false?
True
The ethical principles that apply to epidemiological practice and research include:
- Informed consent
- Confidentiality
- Respect for human rights
- Scientific integrity
This is the most widely known and influential guideline in medical research worldwide. It is an official policy of the World Medical Association (WMA), which was adopted for the first time in 1964 and has since undergone a number of revisions. The Declaration can be regarded as the expression of the WMA’s effort in balancing the need to generate sound medical knowledge with the need to protect the health and interests of research
participants.
Declaration of Helsinki
This is a short document on moral principles that was published in 1978 by the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, in the
aftermath of scandals of research misconduct that were uncovered in the 1970s.
The Belmont Report
The Belmont Report is especially know for?
establishing a framework of basic moral principles—respect for persons, beneficence, and justice—which should guide the conduct of research
Ethical values in RCTs to be respected
Experiments should be socially valuable and scientifically valid, and people
have to be selected fairly, and respected.
The risks and benefits to
participants and the benefits to society need to be weighted against each
other, and there needs to be an independent outside review of the ethics of
the experiment
What was one of the most high profile examples of corporate meddling ?
One of the most high profile examples of corporate meddling was the revelation, 2015, that Coca-Cola had
paid scientists to push the message that exercise was a more effective weight loss tool than cutting down on
food and drink. The beverage giant went so far as to create a non-profit called the Global Energy Balance
Network to push the message that Americans spend too much time focusing on cutting calories, and not
enough time exercising. Following an investigation by the New York Times, the non-profit announced it was
disbanding. In a bid to be more transparent, Coca-Cola revealed it had spent $132.8m on scientific research
and partnerships between 2010 and 2015
RCTS advantage summary
- Comparative
- minimise bias
- minimises confounding factors
- statistical reliability
- sample size
- publishable
Disadvantages of RCTs
- logistics : Power calculation might demand vast samples size, which require more
resources from the investigators - external validity (generalisability) & applicability
- statistics
- ethical limitations
- some research cannot be ethically performed as an RCT
