EBP Midterm Flashcards
PICO
P = patient/population/problem I = intervention (frequency/duration) C = comparison, if any O = outcomes
Stating the PICO question
What is the evidence…
for a patient P (with a specific problem or diagnosis)…
that for a specific intervention I (specific treatment) …
C compared to another intervention or no intervention …
produces a specific O outcome?
Treatment vs no treatment
Go treatment every time (as opposed to a group receiving nothing)
Steps in critical reflection on a treatment intervention
Understand basic science and logic of the disorder AND of the treatment
Determine if the basic science and logic of the treatment will address the basic science and logic of the disorder
Before investigating research on effectiveness of a treatment, critically reflect on the
treatment
When breaking a pico question into search concepts, you place each concept into
a different search block and separate them by AND
Within a search block, when you are trying to search all synonyms and similar concepts, you use
OR between words
On Scopus, you must use ______ around concepts with multiple words or it will search each word individually
quotations
Controlled vocabulary
standardized words or phrases used to express a specific concept
Conflicting evidence
when doing a search there are articles that support treatment of interest, and articles that do not
Publication bias
Positive findings more likely to be published than negative findings
Level I evidence
- Systematic review (meta-analysis) of RCTs
* Individual RCT (well-designed RCT of appropriate size, e.g., n>30)
Level II evidence
- Systematic review of cohort studies
* Individual cohort study (including low quality RCT)
Level III evidence
Evidence from well-designed trials without randomization, single group pre-post, cohort, time series, matched case-controlled studies
Level IV evidence
Descriptive, non-experimental studies that include analysis of outcomes (single subject design, case series, cross-sectional studies and poor quality cohort and case controlled studies)
Level V evidence
Opinions of respected authorities, based on clinical evidence, descriptive studies, or reports of expert committees
Factors influencing bias
sample size, homogenous group, blinded intervention, blinded assessment, randomly assigned, proper outcome measure, comparison group, same dose
Most/least bias in levels of evidence
Most: V
Least: I
Systematic reviews
- Different than regular literature review
- comprehensive, less likely to be biased
- follow a rigorous methodology; gather all articles on a topic and critique the data quality (qualitative)
Meta-analysis
- quantitative summary of literature
- more powerful than systematic review
- can only use if you have combinable data
Primary variance
- only applies to treatment intervention (independent variable)
- Duration/length of treatment, frequency of treatment, administered by health care professional or trained individual, administered correctly
- You want to increase primary variance
Secondary variance
- does not apply to the treatment intervention
- Good things decrease secondary variance
- Bad things increase secondary variance
Good examples for secondary variance
randomization, control group, comparable patient characteristics, dose matched between the experimental and control, blinded intervention, blind observers
Bad examples for secondary variance
Not randomized, receiving other treatments, parent raters, no or limited restrictions on age and diagnosis, comorbidities, medications, patients dropping out
Error variance
- only applies to the measure (dependent variable) or statistics
- Good things decrease error variance
- Bad things increase error variance
Good examples for error variance
standardized instrument
Bad examples for error variance
uncalibrated instrument, non-standardized instrument, small sample size (less than 30 subjects), incorrect statistics
Research design – schematic for independent variable
X
Research design – schematic for dependent variable
O
Research design – schematic for randomized or matched
R, M, or nothing
Other parts of schematics
- Determine how many groups (treatment vs control/comparison group)
- Determine how often dependent variables are measured for each group (how many O’s and where they go in the schematic)
Categorical independent variable
Distinct groups of the independent variable
Ex) control group and experimental
Measurable independent variable
Not distinct groups
Ex) age, weight
Categorical dependent variable
Number of individuals .. (that have falls, showing abnormal sensory profile)
Measurable dependent variable
ROM, strength, scores/measure on questionnaire
Data is normal or skewed
