ch 7 - experimental study Flashcards
what is an experimental study? defining feature of study? other names?
- investigator assigns study participants to two groups; one group receives the exposure of interest, and the other does not. Then, the investigator follows the groups over time for the incidence of disease.
- investigator assigns study subjects
- intervention studies, clinical trials, and randmized clinical trials
what is the format of an experimental study?
- Enrollment based on eligibility criteria
- Investigator assigns exposure to the members of the source population
- Time
- Compare the incidence of the outcome in treatment and comparison groups
what is an experimental study trying to replicate?
- lab experiment that is controlled for everything except for the test chemical
what does the research question involve in experimental studies? key difference from observational?
- involves prevention or treatment. Feasible and ethical, small effect expected
- investigator intervenes
what is the research question in observational studies? key feature?
- nvolves prevention treatment or causal factor, an experimental study is not feasible or ethical. Moderate or large effect expected.
- he investigator watches
equipoise? when is a study appropriate?
- a state in which the investigator truly do not know whether one treatment is better than the other but truly believes that withholding treatment from study participants will not harm them
- An experimental study is only appropriate and ethical when equipoise is present
Prevention trial
- agent is given to healthy or high-risk individuals to prevent disease occurrence.
Therapeutic (clinical) trial-
agent given to diseased individuals to treat or cure disease
Individual trial?
- treatment allocated to individuals
Community trial
treatment allocated to the entire community
parallel trial
individuals in each group simultaneously will receive one study treatment
Crossover trial
two or more study treatments are administered one after another to each group
Factorial trial
- two or more treatments are combined in a factorial trial
Simple trial
Each group receives a treatment consisting of one component
what are the considerations need for a study pop?
- People with a high likelihood of compliance with treatment likely to be followed for the total study period
- We may need to exclude certain people, such as those with conditions for which the drug under study is contraindicated
Generalizability
- he extent to which the results from a study can be generalized (or extended) to people who did not participate in the study
why is the selection of the study population important? what happens if there are too many inclusions or exclusions?
- it directly impacts whom the results generalize once the study is over
- study loses generalizability.
what is a reference population?
- : a group whose results are applicable/generalizable
- examples: All humans
All men/women
Urban children at risk of lead poisoning
what is the purpose of sample size
- to determine a true difference between treatment and comparison groups, enrollment of an adequate number of study subjects is necessary
what is statistical power in terms of sample size?
= ability to demonstrate an association if one exists. Power ranges from 0-100%, good power is >80%
what sizes do trials require? what is needed when the sample difference is small?
- Most trials require a large number of participants
- smaller the anticipated difference between the treatment and comparison groups, the larger the necessary sample size
when are null results common?
- Null results are more common in sample sizes that are too small, and power is too low to detect differences
How do investigators decide how many subjects to include in a trial?
- Annotated differences between the treatment and comparison groups
- The background rate of the outcome
- The proablity of making statistical errors known as alpha and beta errors
what does the Institutional review board (IRB) do? what do investigators need to do?
- must approve study protocol before initiating these procedures
- submits detailed protocol including consent, recruitment, and exposure assignment.
what msut all eligible inidvuals do to be enrolled in a trial?
give consent that includes a clear explanation of research goals and methods, risks and benefits of participating, assurance of confidentiality, right to withdraw.
randomization? what can it minimize?
Each study participant has the same probability of receiving treatment and is meant to prevent biased group assignment
- inaccurate results caused by confounding
Confounders? how are risk factors distribiuted?
- a factor that distorts the true association between exposure and outcome
- unequally across groups
what is the ultimate goal of randomization balancing confounders?
- treatment is the only difference between study groups
what does randomization mean by creating two groups that are the same?
- At baseline, treatment, and comparison group characteristics are balanced
- With randomization, this occurs for both known and unknown confounders
when does randomization work?
- The study is large enough
- The investigator does not influence the treatment assignment
what is masking?
method of ensuring that the participants and or study investigators do not know whether a study participant has been assigned to the treatment or comparison group - not always possible
what is single, double and triple blind?
Single-blind- study participant does not know
Double-blind - study participant does not know, and the study investigator administering the treatment.
Triple-blind - study participant does not know and study investigator administering the treatment and investigator analyzing effects
Why use a placebo?
Makes exposed and unexposed groups’ experiences as comparable as possible (this goal harkens back to lab experiments)
(method of blinindg)
when can placebos not be used?
Interventions such as surgery
It is not always ethical; often use current standards of care instead of a placebo
power of suggestion
Placebo effect participants assigned to the placebo group improve because they are told that they will, which stems from the
Hawthorne effect -
participants in an experimental study either consciously or unconsciously change their behavior because they are being studied
compliance? goal?
- following the study protocol exactly as required throughout the course of the trail
- have groups be as alike as possible on other important characteristics (randomization)
- have groups be as different as possible on exposure
what effect does non-compliance have?
- tends to make treatment and comparison groups more alike
- reduces the ability to detect a difference between groups (when groups become to similar)
what does noncompliance do?
nce tends to bias toward the null
what are the measures of frequency?
Set up a 2x2 table
CI or IR in treatments and comparison
RD or RR comparing treatment and comparison groups
Intention to treat analysis groups? good and bad?
- (treatment vs comparison group)
- Maintains randomization and - good: Gives a real-life picture
- bad: But maybe biased towards null due to non-compliance
Efficacy analysis
group? good and bad?
- Compliers from the treatment group vs compliers from the non-compliers group
- Positives: evaluate the effect of treatment in people treated (compliant)
- Negatives: loss of benefits of randomization and does not give a real-life picture
why cant Experimental studies always be used?
Practical issues: cost, compliance, follow-up
Ethical issues: equipoise
Many methods we implement in other epidemiology study designs are designed to mimic RCTse
