Exam 2 Flashcards
A
-Base line
- Take away the intervention
B
- Intervention
C
- New intervention
What does CPG include?
- search evidence
- clinical expertise
- patient perspectives
Delphi
- formal consensus techniques
- a panel or a group of experts coming up with consensus
- very systematic
Were patients’ views and preferences sought?
- Formal consultations with patients/ public to determine priority topics
- Participation of patients on the guideline development group
- External review by patients on draft documents
Was the guideline externally reviewed by experts?
- Reviewers should not have been involved in the guideline development group
- Reviewers should include experts in the clinical area as well as some methodological experts, Target population (patients, public) representatives may also be included
CPG is updated when either
- important new research is published or
- every 5 years
CPG should provide
- criteria that can be used to monitor or audit clinicians’ adherence to guideline recommendations
Outcome measures
- Questionnaire-based measures
- Performance-based measures
Internal consistency
- multiple items within an outcome measure reflect the same construct
- Cronbach alpha scores close to 0, no correlation
- Cronbach alpha scores close to 1, high and positive correlation, high internal consistency
Test-retest reliability
- outcome measure produces the same result
- repeatedly applied to a patient who has not experienced change in the characteristic being measured
Intra-rater reliability
- same therapist produce the same score on the same patient
Inter-rater reliability
- different raters produce the same score for the same patient
Is the time between assessments appropriate?
- time too short, raters may be influenced by the initial test
- time too long, more risk that the participants will change on the outcome of interest
Interclass correlation coefficient (ICC)
Continuous
Spearman’s rho
Ordinal
Kappa
Nominal
Content validity
- Establish that an outcome measure includes all the characteristics that it purports to measure
- Delphi: a panel of experts subjectively establishes content validity
Face Validity
- not organized
- based on informal evaluation
- a measure appears to measure what it is intended to measure
Concurrent validity
- outcome measure has a high correlation with a criterion measure taken at the same point in time
Predictive validity
- outcome measure has a high correlation with a criterion measured in the future
Known-group validity
- different score for groups with known differences
Convergent validity
- high Coefficient
- similar characteristic or concept
- positive correlation
Discriminant validity
- Low Coefficient
- Distinctly different characteristic or concept
- negative correlation
Spearman’s rho (r)
Value of zero = no association between variables
Value of 1 = Perfect association between variables
Study result that is statistically significant difference
- p less than 0.05 when the alpha level is set at 0.5
- support the establishment of construct validity
Floor and ceiling effects
- lack of sufficient range in the measure to fully characterize a group of patients
Minimal detectable change
- minimum amount of change required on an outcome measure to exceed anticipated error
Responsiveness
- outcome measure’s ability to detect change over time
Minimal clinically important different
- minimum amount of change on an outcome measure that patients perceive as beneficial
Consecutive Recruitment
- minimize bias
- strongest design
- everyone will have the same opportunity
Single- Subject research
- go indepth on one subject at a time
Two consecutive data points have to be what to be statistically significant?
- above or below the 2SD bands in the treatment phase
Mixed methods research
combines qualitative and quantitative data
Why does this correlative exist?
Qualitative research
Phenomenology
- how and why participants behave a certain way
Ethnography
- systematic description of a culture that is based on direct observation of a particular group
- involves the researcher having an extensive immersion in the participants’ environment
Grounded theory
- a theory that explains the phenomena of interest
- how and why an event occurs or how and hwy people might behave a certain way
