Reasoning About Design & Execution of Research Flashcards
The Scientific Method
[8]
- Generate a testable question
- Gather data and resources
- Form a hypothesis
- Collect new data
- Analyze the data
- Interpret the data and existing hypothesis
- Publish
- Verify results
FINER Method of Evaluating A Research Question
- Feasible?
- Interesting?
- Novel?
- Ethical?
- Relevant?
FINER Method of Evaluating A Research Question
- Feasible?
- Interesting?
- Novel?
- Ethical?
- Relevant?
What is a control (standard)?
Conditions that can be applied to act as a method of verifying results
What is a control (standard)?
Conditions that can be applied to act as a method of verifying results; used to correct for any influences of an intervention that are not part of an experimental model
Positive Control
Ensure that a change in the dependent variable occurs when expected
Negative Control
Ensure that no change in the dependent variable occurs when none is expected
Negative Control
Ensure that no change in the dependent variable occurs when none is expected
Placebo Effect
An observed or reported change when an individual is given a sham intervention
Placebo Effect
An observed or reported change when an individual is given a sham intervention
Independent Variable
The part of the experiment that is being purposefully manipulated; plotted on the x-axis
Dependent Variable
The part of the experiment that is being observed; plotted on the y-axis
Casual Relationship
[If-then Relationship]
When there is a theoretical or known mechanism that links the independent and dependent variables
Explain a causal relationship.
If the change in the independent variable always precedes the change in the dependent variable AND the change in the dependent variable does not occur in the absence of the experimental intervention.
Explain a causal relationship.
If the change in the independent variable always precedes the change in the dependent variable AND the change in the dependent variable does not occur in the absence of the experimental intervention.
Accuracy (Validity)
Quality of approximating the true value
Precision (Reliability)
Quality of being consistent in approximation
Why are causal relationships harder to prove in human subject research?
Causal conclusions are harder to determine because circumstances are harder to control
Why are causal relationships harder to prove in human subject research?
Causal conclusions are harder to determine because circumstances are harder to control
Types of Human Subject Research
Experimental and Observational Studies
Types of Human Subject Research
Experimental and Observational Studies
Randomization
Used to control for differences between subject groups; controls whether a subject is placed into a control group or treatment group
Randomization
Used to control for differences between subject groups; controls whether a subject is placed into a control group or treatment group
Blinding
Subject and/or Investigators do not have information about which group the subject is a part of in order to remove bias
Blinding
Subject and/or Investigators do not have information about which group the subject is a part of in order to remove bias
Single-Blind Experiments
Only the patient or the assessor is blinded
Double-Blind Experiments
The investigator, subject and assessor all do not know the subject’s group
Double-Blind Experiments
The investigator, subject and assessor all do not know the subject’s group
Types of Observational Studies
Cohort
Cross-Sectional
Case-Control
Types of Observational Studies
Cohort
Cross-Sectional
Case-Control
[Look for connections between exposures and outcomes]
Cohort Studies
Subjects are sorted into two groups based on differences in risk factors
Cohort Studies
Subjects are sorted into two groups based on differences in risk factors (exposure) and then assessed at various intervals to determine how many subjects in each group had a certain outcome
Cohort Studies
Subjects are sorted into two groups based on differences in risk factors (exposure) and then assessed at various intervals to determine how many subjects in each group had a certain outcome
Cross-Sectional Studies
Attempt to categorize patients in different groups at a single point in time
Case-Control Studies
Identifies the number of subjects with or without a particular outcome and then look backwards to assess how many subjects in each group had exposure to a particular risk factor
Hill’s Criteria
The components of an observed relationship that increase the likelihood of causality in the relationship;
The more criteria that are satisfied, the likelier it is that the relationship is causal
List of Hill’s Criteria
- Temporality
- Strength
- Dose-Response Relationship
- Consistency
- Plausibility
- Consideration of Alternate Explanations
- Experiment
- Specificity
- Coherence
List of Hill’s Criteria
- Temporality
- Strength
- Dose-Response Relationship
- Consistency
- Plausibility
- Consideration of Alternate Explanations
- Experiment
- Specificity
- Coherence
Temporality
The exposure (independent variable) must occur before the outcome (dependent variable)
Strength
As more variability in the outcome variable is explained by variability in the study variable, the relationship is more likely to be causal
Strength
As more variability in the outcome variable is explained by variability in the study variable, the relationship is more likely to be causal
Dose-Response Relationship
As the study or independent variable increases there is a proportional increase in the response. The more consistent this relationship, the more likely it is to be causal
Dose-Response Relationship
As the study or independent variable increases there is a proportional increase in the response. The more consistent this relationship, the more likely it is to be causal
Consistency
The relationship is found to be similar in multiple settings
Plausibility
There is a reasonable mechanism for the independent variable to impact the dependent variable supported by existing literature
Plausibility
There is a reasonable mechanism for the independent variable to impact the dependent variable supported by existing literature
Consideration of Alternate Explanations
If all other plausible explanations have been eliminated, the remaining explanation is more likely
Experiment (Hill’s)
If an experiment can be performed, a causal relationship can be determined conclusively
Experiment (Hill’s)
If an experiment can be performed, a causal relationship can be determined conclusively
Specifcity
The change in the outcome variable is only produced by an associated change in the independent variable
Specifcity
The change in the outcome variable is only produced by an associated change in the independent variable
Coherence
The new data and hypothesis are consistent with the current state of scientific knowledge
Coherence
The new data and hypothesis are consistent with the current state of scientific knowledge
Sources of Error
Selection Bias
Detection Bias
Observation Bias
Confounding Variables
Bias
A result of flaws in the data collection phase of an experimental or observational study
Confounding
Error during analysis
Selection Bias
Subjects used for the study are not representative of the target population; must be measured and assessed before any intervention
Selection Bias
Subjects used for the study are not representative of the target population; must be measured and assessed before any intervention
Detection Bias
Results from educated professionals using their knowledge in an inconsistent way; prior studies that have indicated correlation between two variables, finding one of them increases the likelihood that the research will search for the second
Detection Bias
Results from educated professionals using their knowledge in an inconsistent way; prior studies that have indicated correlation between two variables, finding one of them increases the likelihood that the research will search for the second
Observation Bias
[Hawthorne Effect]
Posits that behavior of study participants is altered simply because they recognize that they are being studied
Confounding Variables
A data analysis error in which an incorrect relationship is characterized; for example, researching the effectiveness on drug (X) on a chance of recovery (Y) is affected by gender (Z)
Z confounds the relation between X and Y since Z causes both X and Y
Confounding Variables
A data analysis error in which an incorrect relationship is characterized; for example, researching the effectiveness on drug (X) on a chance of recovery (Y) is affected by gender (Z)
Z confounds the relation between X and Y since Z causes both X and Y
4 Core Ethical Tenets
Beneficence
Nonmaleficence
Autonomy
Justice
Beneficence
Obligation to act in patients best interest
Nonmaleficence
Obligation to avoid treatments or interventions in which the potential for harm outweighs the potential for benefit
Nonmaleficence
Obligation to avoid treatments or interventions in which the potential for harm outweighs the potential for benefit
Autonomy
Responsibility to respect patients decisions and choices about their own healthcare
Autonomy
Responsibility to respect patients decisions and choices about their own healthcare
Justice
Responsibility to treat similar patients with similar care and to distribute healthcare resources fairly
Justice
Responsibility to treat similar patients with similar care and to distribute healthcare resources fairly
Respect for Persons
Need for honesty between the subject and the research and generally prohibits deception
Informed Consent
A patient must be adequately counseled on the procedures, risks and benefits and the goals of a study
Informed Consent
A patient must be adequately counseled on the procedures, risks and benefits and the goals of a study
Morally Relevant Differences
Differences between individuals that are considered an appropriate reason to treat them differently
Equipoise
Lack of knowledge about which arm of the research study is better for the subject
Equipoise
Lack of knowledge about which arm of the research study is better for the subject; treatments have to be equal and not superior to the other
Equipoise
Lack of knowledge about which arm of the research study is better for the subject; treatments have to be equal and neither can be superior to the other
Monetary Compensation
Does not impact the decision to participate in the study
Coercive Influence
The subject loses autonomy to make the decision to participate
Coercive Influence
The subject loses autonomy to make the decision to participate
Population
Complete group of every individual that satisfies the attributes of interest
Population
Complete group of every individual that satisfies the attributes of interest
Parameter
Information that is calculated using every person in a population
Sample
Any group taken from a population that does not include all individuals from the population
Statistic
Information about a sample
Internal Validity
1
External Validity
2
External Validity (Generalizability)
2
Internal Validity
Identification of causality in a study between independent and dependent variables
External Validity (Generalizability)
Ability of a study to be generalized to the population that it describes
External Validity (Generalizability)
Ability of a study to be generalized to the population that it describes
Statistical Significance
The low likelihood of the experimental findings being due to chance
Clinical Significance
The usefulness or importance of experimental findings to patient care or patient outcomes
