1/80
Looks like no tags are added yet.
Name | Mastery | Learn | Test | Matching | Spaced |
---|
No study sessions yet.
Evidence Based Practice
Using the best current evidence conscientiously and judiciously in making decisions about patient care. It involves considering patient experience/values and what the literature says. It helps identify if something works (efficacy) and how well (efficiency). It is used for illumination, not just to support your ideas.
Boolean Operators
Words like AND, OR, NOT used with keywords in searches. AND makes your search more specific (narrows results). OR makes it more sensitive (broadens results). NOT excludes results.
Journal Impact Factor
A reference about a journal, related to how often articles in that journal are cited.
Bias
Threats to the credibility or usefulness of research. Bias is always present. You must recognize, acknowledge, and act accordingly. Methods like blinding help prevent it.
Disclosures/Conflicts of Interest
Information telling you if authors or funders might have influences that could affect the study findings or their publication.
Article Abstract
A very concise summary of the entire paper. Helps a reader quickly decide if they want to read the full article.
Correlation
Determining the association or relationship between scores on two or more variables, usually by observing data rather than manipulating things.
Causation
Examining if an intervention or factor causes a specific outcome. This is typically studied in experimental research like RCTs, where variables are manipulated.
Within-Groups
A study design where each subject receives every condition being tested. The independent variable has levels like different time points for each subject. Also called a cross-over design.
Between-Groups
A study design where each group of subjects is exposed to only one condition. The independent variable has different groups of subjects receiving different things. Also called a parallel design.
Null Hypothesis
The idea that there is no difference or no relationship between groups or variables.
Alternative Hypothesis
What the researchers hope to prove or disprove; their stated aim or purpose for the study.
Longitudinal
A study that collects data over a period of time.
Cross-Sectional
A study that collects data at a single point in time, like taking a snapshot.
Independent Variable
The factor the researcher manipulates or changes. It is presumed to be the cause of any change observed.
Dependent Variable
The factor the researcher measures or observes. It is presumed to be the effect caused by the independent variable.
Retrospective
Looking back at data or records already collected.
Prospective
Designing a study and then collecting new data going forward.
Inclusion Criteria
Rules stating who got into the study and what characteristics they must have had.
Exclusion Criteria
Rules stating what characteristics kept people out of the study.
Parametric
Statistical tests used when data meets certain assumptions, such as being normally distributed and measured on interval or ratio scales [35].
Non-Parametric
Statistical tests used when parametric assumptions are not met [36].
Random Allocation
Assigning subjects to study groups (like experimental or control) by chance [21, 37]. This helps distribute potential confounding factors evenly [13].
Concealed Allocation
Keeping the group assignment hidden from people involved (like researchers or subjects) until the assignment is made [37]. This helps prevent bias.
Effectiveness
The usefulness of a treatment under normal clinical conditions [38-40]. A pragmatic type of study design [39].
Efficacy
The biological effect of a treatment under carefully controlled conditions [38, 40, 41]. Asks, 'how well does it work?' [38].
Efficiency
Relates to the cost of delivering an intervention relative to the results obtained [4, 38].
Blinding/Masking
Preventing people (like subjects, researchers, or testers) from knowing which treatment group a subject is in to avoid influencing results [13, 14].
Type I Error
Also called Alpha error [24, 42]. Concluding there IS a difference between groups when, in reality, there IS NOT [24, 42]. A false positive [42].
Type II Error
Also called Beta error [25, 42]. Concluding there is NO difference between groups when, in reality, there IS [25, 42]. A false negative; failing to detect a real difference [25, 42].
Sensitivity
Refers to diagnostic accuracy. Means there are very few false negatives [43].
Specificity
Refers to diagnostic accuracy. Means there are very few false positives [43].
Internal Validity
Whether the observations and measurements in a study are truly representative of what's being measured [11]. Rigorous control of variables, like through randomization, improves this [19].
External Validity
How well the study findings can be generalized to other patients or settings outside of the study [39, 44]. Stringent selection criteria can limit this [13].
Paired T-test
A parametric test used to compare differences between two measurements taken from the same subjects or matched subjects. (In contrast to the independent t-test used for two independent groups) [36, 45].
Unpaired T-test
Also called the independent t-test [36]. A parametric test used to analyze differences between two independent groups [36, 45].
PEDro
The Physiotherapy Evidence Database [46]. An online search engine for physical therapy studies, including practice guidelines, systematic reviews, meta-analyses, and randomized controlled trials [46]. Provides a score evaluating study methodology [47].
Cochrane
An organization that produces and disseminates systematic reviews of healthcare interventions [48]. Provides concise summaries of evidence [48].
Strength of Recommendation Taxonomy (SORT)
A scale (A, B, C) used in some journals to grade recommendations based on the quality and consistency of evidence [49]. Grade A is the strongest, based on consistent, good evidence [49].
Clinical Guideline Recommendations
Graded levels of advice for clinical practice (e.g., A, B, C) based on the strength of evidence from research studies [49-51]. Strong evidence (e.g., Grade A) leads to strong recommendations [50, 51].
PICO
An acronym (Patient, Intervention, Comparator, Outcome) used to frame clinical questions. It's the first step in evidence-based practice. Adding T (Timeframe) makes it PICOT.
Five Steps of EBP
1. Frame the clinical question. 2. Retrieve the best evidence. 3. Appraise the evidence. 4. Apply the knowledge to the patient. 5. Evaluate the outcomes.
Evidence Hierarchy
A ranking of study designs based on their likelihood of bias. Systematic reviews and meta-analyses are at the top, followed by high-quality Randomized Controlled Trials (RCTs). Case reports and expert opinion are lower.
Randomized Controlled Trial (RCT)
An experimental study design where subjects are randomly assigned to different groups (like treatment vs. control). Considered a powerful design for examining cause-and-effect due to control of variables. High-quality RCTs are high on the evidence hierarchy.
Cohort Study
A non-experimental study that follows a group of subjects over time. Can look forward (prospective) or back (retrospective).
Case Control Study
A non-experimental study that compares a group with a specific condition (cases) to a group without it (controls).
Sample Size
The number of subjects included in a study. Needs to be large enough for the study to have adequate power to detect a real difference and avoid a Type II error.
Case Report
A systematic description of the care provided to a single patient. Often used to share new ideas or test theories. Lower on the evidence hierarchy.
Case Series
A systematic description reflecting outcomes from multiple subjects with similar conditions. Lower on the evidence hierarchy.
Probability Sampling
Selecting a subgroup of subjects from a population so that the sample represents the larger group. Includes methods like random sampling.
Non-probability Sampling
Selecting a subgroup of subjects from a population based on methods other than random chance. Includes convenience sampling (using easily available subjects) or purposive sampling (selecting for a specific reason).
Systematic Review
A type of study that synthesizes all existing research around a specific clinical question. High on the evidence hierarchy.
Meta-Analysis
A statistical technique used to combine results from multiple studies, often done as part of a systematic review.
Quantitative Research
Focuses on studying groups and manipulating variables. It is a formal, objective, systematic process, often using numbers and statistical analysis. Common in physical therapy research. Aims to examine causality and prove effectiveness.
Qualitative Research
Focuses on broad description to understand experiences or phenomena without manipulating variables. Uses language to define data and content. Data collection stops when saturation is reached. More common to use purposive sampling.
Frequency Distributions
Ways to summarize or portray data, showing how often different values or groups occur. Examples of shapes include symmetrical, uniform, bimodal, or skewed.
Measurement Scales
Rules for assigning numbers to represent quantities or attributes.
Measurement Scales
Rules for assigning numbers to represent quantities or attributes [75]. Examples include nominal (classification only, no rank) [75].
Descriptive Research
A type of research that aims to summarize information about a topic using data [76].
Predictive Research
A type of research that tries to determine if a score on one variable can predict a score on another variable [18, 76].
Causal Research
A type of research (often experimental) that examines cause-and-effect relationships [20, 76].
Reliability
The extent to which a measurement is consistent or reproducible time after time [77, 78]. Refers to the consistency of the measurement [78, 79]. Validity usually assumes this is present [79].
Validity
The accuracy or credibility of a study or measurement [11]. A measurement is valid if it is measuring what it is intended to measure [79]. Validity usually assumes reliability is present [79].
Patient Reported Outcome Measures (PROMs)
Tools where the patient reports on their own status, symptoms, or function [80, 81]. Examples include pain levels or functional questionnaires [80, 81].
Construct Validity
The validity of the underlying concept or idea being measured, important if the concept is not clearly defined [79, 82]. For example, how do you define and measure 'strength'? [82].
Content Validity
Whether a measure fully represents the concept you are interested in, often relevant for questionnaires or tests [82, 83]. Asks if the measure covers everything it should [83].
Criterion Validity
The extent to which a measure is related to other measures or outcomes [83]. Can be concurrent or predictive [83].
Concurrent Validity
A type of criterion validity where a new measure is compared against a known gold standard measurement at the same time [83, 84].
Predictive Validity
A type of criterion validity where a measure taken at one point in time can predict future status or outcomes [83].
Mean
A descriptive statistic used to characterize central tendency [73]. It is the average value in a dataset [85].
Median
A descriptive statistic used to characterize central tendency [73]. It is the middle value in an ordered dataset.
Mode
A descriptive statistic used to characterize central tendency [73]. It is the most frequent value in a dataset.
Standard Deviation
A descriptive statistic used to characterize variability [73]. It measures the spread of data around the mean [86].
Foreground Information
More specific knowledge needed for a complex clinical question, assuming you have some general background knowledge [87]. Requires critically reviewing studies from the literature [87].
Background Information
General, foundational information needed when you don't know much about a topic (e.g., What is it? How do you diagnose it?) [88]. Can often be found in textbooks or review articles [87].
Correlation Coefficient
A statistical measure showing the strength and direction of the association or relationship between two variables [86].
Database
A collection of stored articles [89, 90]. Examples include Medline and CINAHL [89].
Search Engine
A tool used to search within a database to find articles based on keywords [6, 90]. Examples include PubMed and Ovid [6].
Reporting Guidelines
Standards (like CONSORT or STARD) that authors follow to ensure their research papers include all necessary information about the study design and methods [91].
Incidence
The rate of new cases of a specific health outcome over a period of time [92]. Like the water coming into a bathtub [92].
Prevalence
The total number of existing cases of a health outcome at a specific point in time [92]. Like the water already in a bathtub [92].