General Methodological Concepts of Research Flashcards
Name the 2 types of Quantitative Studies
Interventional and Observational
Describe Interventional Study Design characteristics
aka. “experimental” have forced allocation study groups and have the potential to be less ethical
Describe Observational Study Design characteristics
observe subject elements occurring naturally or elements that subjects choose to be a part of
there is no researcher-forced group allocation
Describe the relationship observational studies have with proving causation
observational studies cannot prove causation
List Interventional study designs in increasing level of evidence
phase 0, phase 1, phase 2, etc
List Observational study designs in increasing level of evidence
Case reports
Ecological
Cross-sectional
case-control
cohort
Define the null hypothesis in a study
a research perspective which states there will be no (true) difference between the groups being studied
researchers either reject or dont reject this perspective based on the results obtained via data analysis
What are the 3 various statistical-perspectives that can be taken by the researcher for the null hypothesis?
Superiority
Noninferiority
Equivalency
Explain the Superiority null hypothesis perspective. What is a common comparison it is used for?
the null hypothesis is “im not superior”
asks the question “I wonder if im better”
usually used when comparing to placebo
Explain the Noninferiority null hypothesis perspective. What is a common comparison it is used for?
The null hypothesis is “im not worse”
asks the question “I wonder if im not worse”
Usually used to compare a new treatment to something that is already accepted to be effective
Explain the Equivalency null hypothesis perspective. What is a common comparison it is used for?
The null hypothesis is “Im not equivalent”
asks the question “I wonder if im equivalent”
True or False: Studies that yield a rejected null hypothesis of superiority are said to be less effective than what they were being compared to.
False
After a superiority null hypthesis is rejected, the relationship could still be either “noninferiority” and “equivalency”
Compare probability and non probability samples
probability sample: every element in the population has a known probability of being included in the sample
Nonprobability sample: every element in the population does NOT have a known probability of being included in the sample
In terms of probability sampling schemes, define Simple Random Sampling
assigning random numbers, then taking randomly selected numbers to acquire the desired sample size
(or assign random #’s and then take a sequential order of the randomly given numbers)
In terms of probability sampling schemes, define Systematic Random sampling
assigning random numbers, sort these numbers, and then select subjects via a predetermined sampling interval (take every Nth numbers) to get the desired sample size
In terms of probability sampling schemes, define multi-stage random sampling
uses simple random sampling at “multiple stages” towards patient selection
Ex. Take simple random samples out of one country, then out of one zip code, then out of one household, then out of one person
In terms of probability sampling schemes, define clustered multi-stage random sampling
basically multi stage random sampling, however ALL elements at each stage are included in the study
Ex. ALL clinics in a zip code and then ALL households in a community
In terms of probability sampling schemes, define Convenience (Quasi-systematic) sampling
aren’t really completely random. The researcher decides what fraction of the population is to be sampled and how they will be sampled.
Ex. All persons with last names C-D, all members that work for the same business, all MWF visitors of a clinic
In terms of probability sampling schemes, define stratified random sampling
collecting subjects in different strata based on a desired characteristic (may be a known confounder) and then using simple random sampling to get the desired sample size
In terms of probability sampling schemes, define stratified disproportionate random sampling
Collecting subjects in a stratified manner that helps achieve a desired proportional representation of a underrepresented portion of the population.
over sampling is an example of this, however over sampling requires a “weighted” return sample population back to the baseline population
Describe the difference between patient-oriented outcomes and disease-oriented outcomes.
pt-oriented outcomes: avoiding a MI (things that most patients care about)
Disease-oriented outcomes: cholesterol number (most patients dont care about their cholesterol number if it isnt actively affecting them)
Describe internal validity as it applies to methodology of study
methods inside of the study affect the internal validity of the study.
high internal validity means that the study is scientifically rigorous/standardized and the results are accurate/reproducible.
True or False
3 of the 4 principles of bioethics must be met by a study in order to be ethical.
False
all 4 principles must be met
List and describe the 4 principles of bioethics.
Autonomy: participants fully understand all of the risks/benefits that could occur in the study, and willingly volunteer for it
Beneficence: the researchers must have good reason to believe that the treatment will benefit the patient NOT society
Justice: equal and fair treatment of all patients must occur
Nonmaleficence: researchers must do no harm to subjects (no lying/witholding to subjects)
List the purpose and 3 guiding principles of the Belmont Report
it’s purpose is to protect the human subjects involved in research
Respect for Persons: subjects must be autonomous and voluntarily involved
Beneficence: the risks of the research must be justified by the potential benefits of the treatment
Justice: risks and benefits of the research must be equally distributed to subjects
Describe the difference between consent and assent
consent: mentally capable people over the legal consenting age (18) agree to be a subject after being completely informed
Assent: the same as above, except the subject is NOT able to give legal consent (too young and caregivers/guardians must give consent)
What does IRB stand for, what is another name for them, and what do they do?
Institutional Review Board
Ethics committee
they protect human subjects from undue risk. Their decisions are regulated by law “they have teeth”
What must the IRB do prior to a study beginning?
review all human subject studies
What is the OHRP?
Office of Human Research Protections
an agency that administers and enforces regulations on research studies (basically enforces IRB decisions)
List and describe the 3 levels of IRB reviews in terms of the risk to subjects, pt identifiers, and the amount of time it usually takes to be approved.
Full Board: used for all interventional trials with any risk to patients (takes the longest to be approved)
Expedited: minimal risk and/or no patient identifiers (middle ground in terms of approval time it requires)
Exempt: no pt. identifiers, low/no risk, and uses already gathered datasets. (gets approved very quickly)
What is DSMB? when do they come into play?
Data Safety and Monitoring Board: semi-independent committee not involved int he conduct of the study
they review the study data at predetermined intervals “as the study progresses” to asses for undue risk/benefit between groups.
Compare and contrast Systematic Review and Meta-analysis in terms of their location on the research evidence pyramid.
Systematic reviews: “collective book report” of several studies written up as one big study
Meta-Analysis: uses the data from several other studies and calculates figures as if they were all part of one big study
Both are at the same level, at the top of the pyramid (highest strength of evidence)
Where are interventional (exploratory) and Pragmatic (explanatory) trials at on the research evidence pyramid?
they are at the same level, 2nd closest to the top.
Define Equipoise
genuine confidence that an intervention may be worthwhile (risk vs. benefit) in order to use it in humans
Give a pro and con perspective on the subject of equipoise
Pro perspective: this ensures that the subjects participation in the study will be worthwhile because they may gain the benefits of the experimental treatment
Con perspective: some believe placebo groups in a study to be unethical because the placebo group may not be receiving the benefits of the experimental treatment.
What are study designs based on? (7 but really 6 things)
Perspective of the research question (hypothesis)
Group allocation / randomization
Ethics
Efficiency / practicality / cost
Internal validity
External validity / generalizability