Exam 1 - Research Designs & Methods Flashcards
Validity
The accuracy of results
Extent to which an instrument measures what it is intended to measure
(If it isn’t accurate, it isn’t valid)
Internal validity
Degree to which study outcomes can be explained by the differences in the assigned intervention
To strengthen: add a control group
External validity
Extent to which the results of a study can be generalized to other settings
(Is it valid outside the study?)
Causality
The outcome is the result of the treatment
(Think cause and effect)
Analytical studies
Provide casual interpretation of an existing phenomenon
Used to test a hypothesis
(Casual anal)
Descriptive studies
Describes a phenomenon
Describes/summarizes info about diseases/events without making any casual inferences
Casual inferences = who, what, when, where, why
Prospective studies
Data is collected after the study as individuals are followed
Time and cost intensive
Researchers determine variables
Interventional/experimental studies
Researcher controls the treatment
Usually involved randomization
Retrospective studies
Researchers go backward in time to determine relationship between cause and outcome that has already occurred
Require minimal resources; researchers have NO control over variables
(Retro = past)
Quasi-experimental studies
Experimental designs WITHOUT randomization
Kind of like observational studies
What is the gold standard in evaluating the safety and efficacy of an intervention
Randomized controlled trials
2 essential elements of RCTs
Randomization of participants to interventional and control groups
Prospective - patients in study group are followed after the intervention to evaluate changes in clinical outcome
(Random and prospective)
Do RCTs increase or decrease internal validity
Increase
What is the key element of observational designs
Non-randomization of the independent variable
Observational designs
Researcher observes the relationship between independent (intervention/exposure) and dependent (outcome/disease) variables in a natural setting
(Observing variables)
Types of observational studies
Case report
Cross sectional
Case control
Cohort studies
(4 Cs or Really Sad Cassie Studies)
Case report
Study of a single case of a new disease/manifestation
Cross-sectional
The exposure and outcome are measured at the same point in time
(The exposure and the outcome cross)
Case control
Involve comparison of exposure status among individuals w/ disease or outcome of interest (cases) and those without (controls)
(To compare, you have to control the case)
Cohort studies
Followed exposed and unexposed groups over a period of time until the development of the outcome of interest (RR)
Research methodology
Focuses on data collection and measurement techniques
(Collection and techniques are methods)
Primary methods (research methodology)
Collect data specifically for the research question under consideration
Ex. Surveys and observations
Pros/cons of primary methods (research methodology)
PROS: researcher can collect data to fit needs of study
CONS: resource intensive (cost and time)
(Takes more time and money to have something personalized)
Pros/cons of secondary methods (research methodology)
PROS: easier to conduct and less costly
CONS: may not include certain variables needed for the purposes of the study and may be difficult to interpret
(You save time and money if you don’t order something personalized, but it may not be exactly what you want)
Secondary methods (research methodology)
Use of data that was collected for a different purpose
Ex. Medical charts and medical claims
Reliability
Consistency and reproducibility of results
RCTs are in which phase of the drug development process
Phase III
Bias
Systematic errors that can occur during the implementation of a study
What can be done to reduce selection bias
Randomization
Investigator bias
Errors in the the:
- study design
- implementation
- analysis by the investigator
Simple randomization
Use of a random number generator to allocate participants to study groups
(Random number generator sounds like the easiest [simplest] option)
Hawthorne effect
Subjects modify their behavior because of the fact they’re being studied or observed
(Hawthorne was a man, and men be changing their behavior like this)
Selection bias
Preferential enrollment of specific patients into one treatment group over another
Randomization
Assigned patients to a treatment or control group by chance
Block randomization
Ensures treatment groups have an equal number of patients by dividing into “blocks”
Stratified randomization
Ensures certain baseline characteristics are equal between groups of a study
Group/cluster randomization
Subjects in the same cluster would be randomized and receive the same treatment
Blinding
People involved in the trial are unaware of what treatment the patients are receiving
(Blind to their treatment)
Double blind
Two sets of individuals are unaware of what the patients are receiving
(Double = two sets)
Single blind
Only one set of individuals is unaware of what the patients are receiving
(Single = one set)
Sample size
Number of participants to be enrolled in each treatment group in a study
Effect size (sample size)
Statistical elimination of the magnitude of effect due to treatment or the association between 2+ variables that is likely to occur
(How big is the effect)
Power (sample size)
Measures the capacity to detect a difference in the study groups if a true difference exists (a larger study group indicates greater power)
(Think power difference)
Clinical research protocol
Standardized document that provides instructions to the investigators on all aspects of carrying out the study
(Think of it as the instruction manual)
Inclusion criteria
The specific characteristics that the investigator is most interested in studying
(You’re only included if the researcher is interested in you)
Exclusion criteria
The factors that would confound or impair the ability to interpret the study results or eliminate patients that should not be receiving the intervention
Relative risk
Ratio between the rate of the outcome in the treatment and control groups
Placebo
Consists of an inert substance (ex. Lactose powder) that is identical in appearance (shape, form, taste, color) to the active treatment
Relative risk difference
How far away the relative risk is from the “no difference mark” of 1
Number needed to treat
The number of patients that must receive the treatment in order for one patient to experience a desired outcome
(Number needed for treatment to work in one person. Like 1 in X)
Number needed to harm
The number of patients that must receive the treatment in order for one patient to experience an adverse outcome (should be high to be considered a safe medication)_
(Think out of # people, only one was harmed. Ex. You have a 1 in 1 million chance)
