module 3 Flashcards
what is a clinical prediction rule
a tool that contains the best combo if individual components of history, physical examination, and or lab test that can help a clinicion make a diagnosis and prognosis or select a approciate treatment
what do CPRs do
help a clinician make a diagnosis and prognosis or select an appropriate treatment
what is a diagnostic CPR
ID patient with a disorder
example of a diagnostic CPR
Ottawa ankle rules
treatment classification CPR
ID pt that would respond most favorable to a treatment
EX: lumbar spine manipulations
what are the steps n developing a CPR
- derivation
- validation
- impact analysis
what is derivation
ID predictors and create a rule
what is Validation
apply the rule in seperate independent samples
what is the impact analysis
assess how the rule impacts clinical outcomes and behavior
and or reduce costs
what is a narrow validation
application of the rule in a smilar clinical setting and population as the derivation stage
what is a broad valdiation
application of the rule in multiple clincial settings with varying prevelence and outcomes of the disease
approiate study designs for CPRs - prospective cohort studies
diagnostic/prognostic CPRs
treatment based CPRs in the derivation stage
what kind of study is also appropiate for diagnostic CPRs
cross sectional studies
approiate study designs for CPRs - RCT
treatment based CPR in the validation stage
what does operationally defined mean - target population
clearly articulated the inclusion and exclusion criteria
what can you assume if if a CPR survies validations
CPR is valid in predicting the outcome of interest
impact of the CRP clinician decision making
how to choose predictor factors for a CPR
ID varible that predict and outcome by itself
ID a verible that when grouped with other outcomes increases the likelihood of the outcome together
what do you use to see if predictor values work well together
multivariable regression analysis
how do you know if the sample size was large enough
look for the sample size and or power in the discussion of the paper
how should the outcome of interest be applied for CPRs
the tester should be trained on how to preform it
have objective criteria on positive and negative results
what is attrition
the loss of study units from a sample.
when a study loses a subject and they are therefore not included in the analysis
what level of attriction do you start to question the studies valdity
> 20%
in the valdiation stage what must the prediction rule be compared to
an established criterion (gold standard) or outcome of interest
during the validation stage what must be true of the subjects
must be a second in-pendent group of subjects
what is the relationship between the criterion and the CPR
they must be perform independently
what cuase a study to fail the validation stage
the rule was inadquatly developed
drastic difference in development population and validation pop
bad sample size or power to detect effects
what stage are most CPR for PT
derivation stage
many fail the validation stage
almost none have undergone a impact anaylsis
