Principles of Research Flashcards
Research ethics.
Reasonable uncertainty in context of clinical trials about whether the experimental treatment is better than standard treatment
Equipoise
Research ethics.
Tenet that benefits of treatment must outweigh harms
Beneficence
Research ethics.
Process of obtaining agreement to participate in research, while ensuring they are fully informed about consequences and implications of a study
Informed consent
Research ethics.
A committee that reviews research proposals to ensure their ethicality
Institutional review board
Information gathered subjectively to determine if patterns or themes are present
Best used for complex processes, like human social behavior
Examples: Ethnography, focus group
Qualitative
Data collected objectively and analyzed statistically
Best used for determining relationships between variables
Examples: Randomized controlled trial, correlation studies
Quantitative
Observational
Data is gathered from the same subjects repeatedly over time
Measure how body composition changes as we age
Longitudinal
Observational
Measures both outcome and exposure at the same point in time
Finding the prevalence of breast cancer in a population at a single point in time
Cross-sectional
Observational
Measures outcome status and then relates it to history of exposure
Studying whether lung cancer patients have greater smoking history than healthy controls
Case control
Observational
Detailed information about a small group of individuals
Detailed report of a unique and very rare medical case
Case series
Observational
Longitudinal study that correlates exposure with rate of an outcome in a cohort over time
Studying whether smokers have higher rates of lung cancer by following them over time
Cohort study
Experimental
Random allocation of subjects to different treatments and comparing them based on a measured outcome. Gold standard for clinical trials
Clinical trial randomizing between a new drug and placebo
Randomized control
Review
Data from multiple studies are statistically analyzed and combined into a single review
Review of multiple clinical trials demonstrating that aspirin reduces stroke risk
Meta-analysis
Research principles:
Internal and external validity
Internal validity: The degree to which a study demonstrates causality and controls for confounders
External validity: The degree to which results of a study can be generalized to other populations
Research principles:
Construct and criterion validity
Construct validity: Ability of a study to measure what it intends to measure
Criterion validity: Ability of a study to correlate with another test, usually an accepted standard