diagnostic
in the 2x2 table for diagnostic test analysis which letter is the true positives?
which letter is the true negatives
true positives= a (+disorder & + test result)
true negatives= d (- disorder & - test results)
what are important questions to ask about patient applicability
does the purpose answer our clinical question?
does our patient fit the inclusion and exclusion criteria?
what are the differences between inclusion and exclusion criteria?
inclusion criteria: study eligibility, broad enough to maximize sample size, narrow enough to minimize extraneous influence
exclusion criteria: study eligibility, specify extraneous factors that are anticipated to influence the outcome
what is purposive sampling
what is attrition and what is the solution to this
loss of subjects during a study
solution: replacement of subjects; document characteristics and reasons for withdrawal of subjects lost, statistical analysis
what is the definition of sensitivity and the equation for sensitivity through the 2x2 table?
If sensitivity is high and the test is negative do we rule in or out the condition?
sensitivity= true positives= a/(a+c). The proportion of patients with the condition of interest who test positive on the diagnostic of interest
Sn= sensitivity, N=negative, Out= rules out (SnNout); if sensitivity is high and the test is negative, can rule OUT the condition
why do we appraise credibility and quality of a research article
does the purpose answer our clinical question
does our patient fit the inclusion and exclusion criteria
what is the target population, accessible population, and a sample
target: total group of interest, group the researchers hope to get
accessible: potential subjects who are available for study, before inclusion and exclusion criteria are applied but could reasonably participate
sample/research subjects: individuals, organizations or other units of analysis who meet the inclusion and exclusion criteria and about whom information will be gathered for a research study
what is consecutive assignment
randomly assigned order of group assignment before subjects enrolled in the study, as subjects come into the study they are assigned according to this previously
what is compensatory equalization of treatments and the solution
individuals providing the interventions purposefully or inadvertently supplement the activities of the subjects in the control or comparison group
solution: mask the investigators or the people providing interventions so that they do not know group assignment; provide a clear and explicit protocol for intervention administration
what is the definition of specificity and what is the equation through the 2x2 diagnostic table?
if the specificity is high and test is positive do we rule in or out of the condition
specificity= true negatives= d/(b+d); the proportion of individuals without the condition of interest who test negative in the diagnostic test of interest
Sp= specificity, P= positive, In= rules in (SpPin); if specificity is high and the test is positive can rule in the condition
what is magnitude of the study?
what is the different between probabilistic and non-probabilistic sampling?
which sampling techniques fall for probability and nonprobability
Probability: randomization occurs at some point in the process, tends to result in more representative samples, and may reduce sampling error (simple random sampling, systematic sampling, stratified random sampling, cluster sampling)
nonprobability: no randomization, may result in over or under-representation of population subgroups within sample, may increase (convenience sampling, snowball sampling, purposive sampling)
simple random sampling: each member of population has = chance of being selected for the sample, the selection of subjects is independent of one another, each population member is assigned #, and a random # table or generator is used to select the sample
systematic: every "nth" subject on the sampling frame is selected, starting point (ex first subject) is chosen via random # table or generator
what is compensatory rivalry or resentful demoralization and the solution
subjects may acquire knowledge about different groups activities
solutions: keep subjects in each group separated to avoid communication between groups, mask subjects and investigators so they do not know to which group they belong, provide explicit instructions as to importance of adhering to specific protocol
What is pretest probability? What is post-test probability?
Why is posttest probability important?
pre-test: the odds that a patient has a condition based on their clinical presentation
post-test: the odds that a patient/client has a condition based on the result of a diagnostic test; this is important because it tells us how likely it is that the person has the condition of interest, whether further testing is needed or whether intervention can begin
what are 3 considerations when applying the results or a study to a patient?
1. can we perform diagnostic test, prognostic factor, intervention, and outcome measures given your knowledge and available resources
2. does the diagnostic test, prognostic factor, intervention, or outcome measure fit with patients beliefs, values, preferences
3. would the patient be willing to participate-perform the diagnostic test, prognostic factor, etc.
What is cluster sampling?
What about convenience sampling
cluster: used when naturally occurring pockets of population are geography dispersed, randomly select from the "pockets" ex: DPT programs across the USA, randomly choose DPT programs to participate
convenience: use readily available subjects, consecutive convenience sampling= used in a prospective trial
what are the following assignments:
by individual and block assignment
individual: randomly assign each individual to a group (coin flip), may result in unequal group sizes, which may be problematic with small sample sizes
block: predetermine the # of subjects to be contained in each group, assign each member of the sample a #, select the maximum # of subjects per group via random
what is testing and solutions
subjects demonstrate improvement in outcome measures due to exposure to the testing procedure or due to different instructions and cues provided by the administrator
solutions: use outcome measures that have good reliability, provide several practice sessions, provide specific protocols including scripts for use by test administrators
Likelihood Ratio Interpretation for pretest to posttest probability for the following:
LR+>10; LR- <0.1
LR+ = 5 to 10; LR- = 0.1 to 0.2
LR+= 2 to 5; LR-= 0.2 to 0.5
LR+= 1 to 2; LR-= 0.5 to 1.0
Large and conclusive change: LR+>10, LR- <0.1
Moderate change: LR+ = 5 to 10, LR- 0.1 to 0.2
Small but sometimes important change LR+= 2 to 5, LR- = 0.2 to 0.5
Negligible change LR+= 1 to 2, LR-= 0.5 to 1.0
why do we use appraisal forms in EBP 1
provide a structured format for appraising an article
they are helpful as you learn & may be helpful initially in the clinic
they provide the important questions for that component of the patient/client management model that assess for biases & threats to validity
they provide questions about results and magnitude of results
what are the following random assignment techniques:
Systematic assignment and matched assignment
systematic assignment: sample members count off or are numbered repetitively according to group members; those with # 1 go into group 1, those with #2 go into group 2, etc.
matched assignment: subjects are matched or formed into subsets based on important characteristics, and members of these subjects are randomly assigned to groupes
what is stratified random sampling and snowball sampling
Stratified: ensure specific subgroups are represented in adequate #'s (nonproportional) or preserve population of subgroups in the population within the sample (proportional) requires non-overlapping stratification criteria, randomly select from subjects in each stratum
snowball: recruit subjects that go on to recruit more subjects, used when potential members of the population are identified
waht is maturation and the solutions
subjects mature or change over time like with children
solutions: randomly assign subjects to treatment and control group, ensure that time of day for participation is consistent, adequate rest provided between repeated measures, specific intervention techniques presented in random order