EOR Flashcards
1
Q
The integration of best research with clinical expertise and patient values
A
evidence based medicine
2
Q
The conscientinous, explicit, and judicious use of the current best evidence in making decisions about the care of individual patients
A
evidence based medicine
3
Q
What are the 4 domains of practice?
A
Diagnosis
Prognosis
Intervention
Harm
4
Q
What are the 7 general steps of EBM method?
A
- Acknowledge there is something I don’t know
- Formulate a foreground question (PICO)
- Search online datebases
- Select best evidence
- Critically appraise evidence
- Integrate evidence with clinical practice and patient values
- Evaluate self: how am I doing as an evidence based practitioner?
5
Q
What are the 3 fundamentals of EBM?
A
1. optimal clinical decision making
* awareness of best available evidence (systematic summaries)
2. assess whether evidence of trustworthy
* confidence of properties of diagnostic tests, patient prognosis or impact of therapeutic options
3. evidence alone is never sufficient to make a clinical diagnosis
* weight risks vs benefits, alternatives, values and pt preferences
6
Q
What does PICO stand for?
A
Patient / Population
Interventions / Exposures
Compare - control, alternative, or comparison
Outcome we are interested in
Time - only used in PECOT to provide restrictions to begining and end of studied period
7
Q
What are the 5 categories of EBM resources?
A
- Summary
- Guidelines
- Pre-appraised research
- Non pre-appraised research
- Federated searches
8
Q
An EBM resource that is updated regularly to integrate the body of evidence of several related questions and provide actionable recommendations for practice.
Give 3 examples.
A
Summaries
UpToDate, Best Practice, DynaMed
9
Q
An EBM resource that focuses on providing specific topic/disease focused recommendations for optimal patient management.
Give an example.
A
Guidelines
US national guidelines clearinghouse
More difficult to search for since they are scattered across specialty journals and organization websites.
10
Q
An EBM resource that looks directly at research findings; search sources that select only Systematic Reviews and studies that meet defined methadological criteria and provide synopses.
GIve 4 examples.
A
Pre-appraised research
ACP journal club, Cochrane, McMaster, DARE
11
Q
An EBM resource that is the largest compilation of controlled trials
A
pre-apprased research
12
Q
An EBM resource that searches all 3 categories without pre-vetting. Must use filters like ____.
Give 3 examples.
A
Non pre-appraised research
clinical queries (broad filter is more sensitive, narrow filter is more specific)
pubmed, medline, CINAHL
13
Q
An EBM resource that provides all layers of research at once, including summaries, guidelines, preappraised and nonpreappraised; requires advanced searching skills.
Give 4 examples.
A
Federated search engines
ACCESS, Trip, SumSearch, Epistemonikos
14
Q
Differentiate between Level and Grade of evidence.
A
Levels = individual study (1a, 1b, 1c, 2a, 2b, 2c, 3a, 3b, 4, 5)
Grades = summarize multiple studies (A, B, C)
“Multiple letter Grades on a report card”
15
Q
what does GRADE stand for?
what is it used for?
A
Grading and Recommendation, Assessment, Development, and Evaluation
for multiple studies
16
Q
several high-quality studies w/ consistent results or one large, high quality multi-center trial; further research is very unlikely to change our confidence in estimate of effect
A
A (high quality evidence)
17
Q
one high quality study or several studies with some limitations; further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate
A
B (moderate quality evidence)
18
Q
one or more studies with several limitations; further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate
A
C (low quality evidence)
19
Q
one or more studies with very severe limitations; expert opinion - any estimate of effect is very uncertain and there is no direct research evidence
A
D (very low quality evidence)
20
Q
What is SORT? what kind of studies is it used for?
A
Strength of Recommendation Taxonomy
independent studies
21
Q
independent study with consistent, good quality patient-oriented evidence
A
A; sort
22
Q
independent study with inconsistent or limited quality patient-oriented evidence
A
B; sort
23
Q
independent study with consensus, disease-oriented evidence, usual practice, expert opinion, or case series for studies of diagnosis, treatment, prevention and screening
A
C; sort
24
Q
What is the hierarchy of evidence? (5)
A
- N-of-1 RCT
- Multiple patient RCT
- Observational studies (outcomes)
- Basic research (lab, animal, human physiology)
- Clinical experience (expert opinion)
25
what are the 2 types of study designs?
Observational
Experimental
26
what are the 2 types of observational studies?
descriptive
analytical
27
observational study that inclues case reports, surveys, qualitative and case series
descriptive
28
observational study that includes cross sectional, case control, and cohort
analytical
29
RCT, quasi experiements, and single subject design are all examples of what kind of studies?
experimental studies
30
what type of study design is from a given point in time (snapshot) without follow up period? (made on a single occasion)
cross-sectional study
31
what statistic is used for a cross-sectional study?
estimate prevalence
32
what is the estimate prevalence equation?
number of existing cases at a single point in time / total population
33
what does incidence mean?
new cases
i**N**cidence = New cases
34
what does prevalence mean?
all cases
pre**AL**ence = ALL cases
35
which type of study is always retrospective and the outcomes of two groups have already happened?
case control study
36
what kind of study is good for studying rare diseases?
case control study
37
what statistic is used for case control studies?
odds ratio
38
what is the equation for odds ratio?
ratio of the odds of an event in exposed group vs odds of same event in control group
OR = ad / cb
39
which type of study can be retrospective or prospective with observations made and followed over time?
cohort study
40
what statistic is used in cohort studies?
estimate incidence
*I am IN the cohort"*
41
what are the 4 steps of experimental research?
1. identify participants
2. place in common context
3. intervene
4. observe effects
42
what study design is usually used in experimental research?
randomized blinded trial
43
what type of study is putting together all research over a certain question?
systematic reviews
44
what is used to organize all systematic reviews and narrows down the research question?
meta-analyses
45
what statistic is used in each study type?
cross sectional
case control
cohort
cross sectional = prevalence
case control = odds ratio
cohort = incidence
46
what are the 3 criteria for choosing EBM resources?
1. based on current best evidence
2. coverage and specificity
3. availability and access
47
what are the 5 A's for the efficient use of current medical literature?
Ask
Acquire - systemic retrieval of best evidence
Appraise - for validity, clinical relevence, and applicability
Apply to practice
Assess performance
48
what kind of questions ask about definition or pathophysiology of a syndrome or mechanism of a treatment modality?
when are they usually used?
background questions
beginning stages
49
what kind of questions ask target questions of therapy, harm, diagnosis, or prognosis that provide the evidentiary basis for decision making?
what is the 1st step?
foreground questions
structure question in PICO format
50
what are the 5 types of foreground questions?
1. therapy
2. harm
3. differential diagnosis
4. diagnosis
5. prognosis
*"Folklore = The Tortured Poetry Department"*
*Foreground = TTPD (therapy, harm, prognosis, Dx/DDx"*
51
what are the 3 study design questions we should use to assess research on best treatment/therapy? (questions we should ask when looking into research about a treatment we want to use)
1. Validity = how serious was the risk of bias?
2. Outcomes = what are the results?
3. Application = can I apply these results to patient care?
VOA = validity, outcomes, application
52
what kind of trial has the primary objective to determine the magnitude of increased benefit of the experimental intervention over the standard therapy?
superiority trial
53
what is the strength / weakness of a superiority trial?
strength = getting a superior tx compared to the standard
weakness = expensive or inconvenient
54
what statistic is used in superiority trials?
P value
"suPeriority trials"
55
the idea that nothing special is happening; the claim that the effect being studied does not exist
null hypothesis
56
what does p < 0.05 mean?
reject the null (intervention > standard)
57
what does a smaller P value mean?
the saying that there is no difference between two treatments = false
58
what kind of trial provides an alternative to equivalent trials, but is unconcerned if experimental treatment is better as long as it is "not much worse"?
non-inferiority trials
59
what is the weakness in non-inferiority trials?
investigators set their own thresholds for accepted values
*without universally accepted method for defining an appropriate threshold*
60
what is the statistic used in non-inferiority trials?
confidence interval
"non-Confident in non-inferiority"
61
If the upper boundary of the confidence interval is ____ your threshold = the balance between desirable and undesirable consequences is a close one - will involve full exploration of your pt's views of the trade off at hand
near
62
If the upper boundary is _____ ___ than your threshold, very few pts would choose the intervention
substantially greater
63
what are the 3 questions to ask when evaluating a non-inferiority trial?
1. what are the results?
2. are the results valid?
3. how can I apply the results to patient care?
64
what kind of trials establish that an experimental treatment is neither better nor worse than the standard?
equivalence trial
65
occational mistakes that are likely to skew measurements from the study in either direction
random error
66
systematic errors that distort the findings in one direction
bias
67
means it is free from random error
precision
*"r and p are close in the alphabet"*
68
means it is free from systematic error
accuracy
*"systematic error = bias, b and a are close in the alphabet"*
69
how can we reduce the chances of random error?
manipulate sample size (make it larger)
70
how can we control for bias?
improve study design and implementation
71
accuracy vs precision
accuracy (validity)
precision (reliability/reproducible)
"PRAV"
72
distortion in the perception or reporting of the measure by observer; experimenter bias
observer bias
73
how can we reduce observer bias?
double blinding
74
bias that can result from faulty function of a mechanical instrument
instrument bais
75
how can we reduce instrument bias?
use only validated instruments
76
bias that's described as distortion of measurements by the study subject
subject bias
77
how can we reduce subject bias?
compare it to a "gold standard" to assess accuracy
78
what are the 5 groups of people that should be blinded in a study?
patients
clinicians
data collectors
adjudicators of outcomes
data analyst
"we will be blinded as a PAACC"
79
what can eliminate differential bias that affects one study group more than another?
blinding
80
the risk of an event
absolute risk
81
what is the equation for risk (absolute risk)?
a/a+b
82
the ratio of the risk of an event among exposed population vs the risk among the unexposed population
relative risk
83
what is the relative risk equation?
RR = experiment/control group
(a/(a+b)) / (c/(c+d))
84
indicates that the exposure / treatment has no excessive risk for the outcome
RR = 1
85
indicates that the exposure / treatment increases the risk for the outcome
RR > 1
86
indicates that the exposure / treatment decreases the risk of the outcome
RR < 1
87
the **difference** in risk of a harmful outcome between experimental and control group
risk difference (absolute risk reduction (ARR))
88
What is the most useful way of presenting research results to help your decision making? (2)
Absolute Risk Reduction (ARR) or Risk Difference (RD)
89
what is the equation for ARR or risk difference?
control - experimental group
[c/(c + d)] – [a/(a + b)]
90
what is the patient ultimately interested in?
risk difference (RD) or absolute risk reduction (ARR)
91
the proportional reduction in risk of harmful outcomes between the experimental and control group
relative risk reduction
92
what is the most commonly reported statistical measure of dichotomous treatment effects?
relative risk reduction (RRR)
93
what is the relative risk reduction equation? (2)
RRR = 1-RR
RRR = (CER-EER) / CER
94
what does the degree of risk reduction help us decide?
whether a treatment is worth pursuing based on the likelihood that the outcome will be successful
95
tells us the % of patients in the experimental group with a bad outcome
experimental event rate (EER)
96
what is the equation for experimental event rate (EER)?
a / (a+b)
97
tells us the % of patients in the control group with a bad outcome
control event rate (CER)
98
what is the equation for control event rate?
c / (c+d)
99
how do you set up contingency table and what are they used for?
used for RCTs which monitor dichotomous outcomes
100
the odds of an event in an exposed group vs the odds of the same event in a control group
odds ratio (OR)
101
what is the odds ratio equation? (2)
ad / bc
102
the number of patients who must receive an intervention of therapy during a specific time period to prevent 1 adverse outcome or produce 1 positive outcome
Number Needed to Treat (NNT)
103
what is the equation for NNT?
(1 / ARR) x 100
104
do you want a small or large NNT? why?
small number
only need to treat 1 pt to prevent 1 adverse outcome; every pt will benefit
105
the measure of how many people need to be treated (or exposed to a risk factor) in order for one person to have an adverse effect
Number Needed to Harm (NNH)
106
what is the equation for NNH?
(1 / ARI) x 100
ARI (absolute risk increase) = EER - CER
107
do you want the NNH to be large or small? why?
large
the less likely a pt will suffer an adverse event
108
a range of values within which the true value of a parameter lies in
confidence intervals
109
more precision (narrower confidence intervals) result from what? (2)
larger sample sizes
larger number of events
110
How do you interpret confidence intervals in a positive trial of a meta analysis?
1. check the lowest number in that range
2. if lowest # is high enough, then trust that it's a positive trial
3. if the lowest # is too low for the range, then we need more studies
111
a study in which there is no important difference between the treatment and control group
negative study
112
How do you interpret confidence intervals in a negative trial of a meta analysis?
1. look at the highest # in that range
2. if the highest # is too low, then trust it's a negative trial
3. if the highest # is high enough, we need more studies
113
the analysis of accumulated data that takes into account the timing of events; displayed with a survival curve of a group of patients to describe their status at different times after a defined starting point
survival analysis
114
What is the advantage of using survival analysis?
can account for different lengths of follow-up
115
In many trials of a fixed duration, some patients are enrolled early and thus have a long follow-up and some later with shorter follow-up, by a process called ____
censoring
116
the probability of events occurring at any point in each group
hazard
117
The weighted relative risk of an outcome during the entire study period
hazard ratio
118
Which two measures allow you to estimate the patient's risk with treatment?
relative risk (RR) or relative risk reduction (RRR)
119
tells us the difference between the risk with and without treatment
risk difference (RD)
120
Which two measures in a individual patient will be most useful in guiding the treatment decision?
risk difference (RD) & number needed to treat (NNT)
121
what are the 3 P's or considerations when creating differential diagnoses?
probability *(most likely)*
prognostic *(more serious if left undx or untx)*
pragmatic *(more responsive to tx)*
122
What are the 2 complementary approaches to diagnosis?
Pattern recognition
Probabilistic diagnostic reasoning
123
see and recognize it
ex. shingles classic presentation
pattern recognition
124
too complex for pattern recognition; dx requires expert diagnosticians - takes more time, money, and knowledge
Ex: pt presents with weight loss and do not have an easily identifiable cause
Probabilistic diagnostic reasoning
125
What are the 3 steps clinicians use to perform probabilistic diagnostic reasoning?
1. pre-test probability
2. post-test probability
3. compare post-test probability with thresholds
126
done before any formal testing, assemble short list of prausible target disorders to be investivated; forming a differential diagnosis
pre-test probability
127
the revised likelihood of a dx based on test outcome
post-test probablitiy
128
what post-test probability indicates the diagnosis would be absolutely certain?
post-test probability = 1
129
what post-test probability indicates that the dx becomes more and more likely and reaches a threshold of probability in which the clinician would recommend starting treatment for the disorder?
post-test probability approaches 1
130
what post-test probability indicates the diagnosis would be disproved?
post-test probability approaches 0
131
What is used to assess value of performing a diagnostic test (determined by the sensitivity and specificity of the test) and moves us from the pretest probability to a posttest probability?
likelihood ratio (LR)
132
indicates that the post-test probability is exactly the same as the pre-test probability (test is NOT worth doing)
LR = 1
133
Increases the probability that the target disorder is present
LR > 1
134
Decreases the probability of the target disorder
LR < 1
135
the proportion of people with a positive test result, among those with the target condition
sensitivity
*"SNout" = rules out negatives*
136
high sensitivity = low false _____ test
negative
*"SnNout = Negative rules OUT"*
137
the proportion of people who are truly free of a disorder with a negative test result
specificity
*"SPin = Positive, rules IN"*
138
high specificity = low false ____ test
positive
"SPin = Positive rules IN"
139
what are the 2 thresholds in the diagnosis process?
test threshold
treatment threshold
140
in a test threshold, what does a post-test probability nearing 0 indicate?
diagnosis is less likely and may be excluded
141
in a treatment threshold, what does a post-test probability approaching 1 indicate?
diagnosis is likely and treatment can be initiated
142
what does an in between thresholds indicate?
further testing is required
143
studies that are most useful if individual members of the entire group are similar enough, that the outcome of the group is applicable to each participant
prognostic studies
144
a study that enrolls patients at a point in time and follows them forward to determine the frequency and timing of subsequent events
prognostic studies
145
variables or factors that influence whether a patient does better or worse
prognostic factors
146
The separation of pts in two groups based on a prognostic factor to better understand the prognosis in each group
adjusted analysis
147
if a large number of variables have a major effect on prognosis, investigators use statistical techniques like ____ ____ to determine the most powerful predictors
regression analysis
148
summary of research that addresses a focused clinical question in a systematic, reproducible manner
systematic review
149
a statistical pooling or aggregation of results from difference studies in order to increase precision, and is the single best effect estimate to facilitate clinical decision making
meta-analysis
150
what is most clinically useful to combine, in order to provide the best estimate of effect that increases precision and facilitates clinical decision making?
systematic review + meta-analysis
151
the extent to which the design and conduct of the review are likely to be protected against misleading results
credibility
152
Why is credibility important in meta-analyses?
to determine how narrow or wide the scope of the question is (were eligibility criteria for inclusion appropriate?)
153
Systematic review w/o a meta-analysis usually presents results from individual studies. Meta-analysis adds a ____ estimate of effect, w/ an associated confidence interval for each relevant outcome
pooled (combined)
154
____ estimates could be for therapy outcomes, estimates of the properties of diagnostic tests, or estimates of patients' likely outcomes
Pooled
155
clinicians need to know the extent to which they can trust pooled estimates. What is a fundamental problem that can undermine this trust?
credibility of the review
156
____ may be undermined by inappropriate or unspecified eligibility criteria, inadequate research, and the omission of risk of bias assessments of individual studies
credibility
157
when can a highly credible review leave us with a low confidence in estimates of effect? (4)
risk of bias / inconsistent results
small sample size
imprecision (wide CI)
enrolled pts differ from those we're interested in (eligibility criteria)
158
what statistic is used for estimate of effects?
what numbers represent small, medium, and large effect size?
standard deviation (SD)
0.2 = small effects
0.5 = medium effects
0.8 = large effects
159
the effect size that results from the calculation of mean difference between treatment and control and dividing this by the standard deviation, provides a summary estimate of what?
treatment effect
160
statements that include recommendations to optimize patient care
practice guidelines
161
Practice guidelines are ideally informed by what?
Systematic review of evidence and an assessment of the benefits/harms of alternative options
162
What must be done to make a recommendation for practice guidelines? (4)
1. define clinical questions
2. select relevant outcome variables
3. retrieve and synthesize relevant evidence
4. rate confidence in effect estimates
163
an examination of the differences in effect estimates across included studies, in an attempt to explain differences in results (used in meta-analyses)
heterogeneity
164
What two statistical tests are used for heterogeneity? which is preferred?
I ^2 (preferred)
P-value
165
tells us the magnitude of variability across primary studies (what amount of difference is due to chance vs due to true variation between studies)
I ^2
166
shows statistical significance of variability
P-value
167
an I ^2 that indicates that the differences in results are more likely due to chance (homogenous)
I ^2 closer to 0%
168
an I ^2 that indicates that the differences in results is less likely due to chance (more heterogenous)
I ^2 closer to 100%
169
in I ^2 < 50% (homogenous, more likely due to chance), what model is used in the meta-analysis?
fixed effect model
170
in I ^2 > 50% (heterogenous, less likely due to chance), what model is used in the meta-analysis?
random effect model
171
When looking at a meta-analysis, the starting assumption is that if a summary estimate of tx is provided (across range of pts, interventions, and outcomes) the effect of the interest is _________
about the same
172
if combining diverse studies violates the starting assumption that the effect of interest is about the same, it may lead to false conclusions. what is the solution?
evaluate the variability/heterogeneity of study results
173
what does a large difference in variability indicate in terms of confidence?
less confidence
174
what does a small difference in variability indicate in terms of confidence?
more confidence; effect is more or less the same
175
if the cofidence intervals overlap widely, the differences are likely due to ____
random error / chance
176
if the cofidence intervals don't overlap, random error is ____ the cause of the differences
unlikely
177
Failure to report or publish studies with a negative results is an example of _____ bias
reporting
178
When an entire study remains unreported, we have ____ bias
publication
179
bias when investigators measure a number of outcomes but report those that favor the experimental intervention or those that favor the intervention most strongly
selective outcome reporting bias
180
Network meta-analyses which simultaneously include both direct and indirect evidence are sometimes called a __________ ____
closed loop
*"your network is a closed loop of people"*
181
what type of systematic review allows the comparison of multiple interventions, and provides estimates of effect for all possible pairwise comparisons
network meta analysis
182
example of closed loop / network meta-analyses?
comparing 7 different triptans
183
Network meta-analyses are subject to what 3 chief considerations?
1. are the studies sufficiently homogenous to combine for each intervention?
2. are the trials in the network sufficiently similar, except the intervention?
3. are the findings sufficiently consistent to allow confident pooling of direct and indirect evidence together?
184
formal method that integrates the evidence regarding the beneficial and harmful effects of treatment options with values or preferences associated with those effects
decision analysis
185
how are clinical decision analysis built?
decision trees (built as structured approaches)
186
What are the 4 clinical decision making approaches?
1. Paternalistic or parental
2. Clinician-as-perfect-agent
3. Informed decision making
4. Shared decision making
187
what type of decision making approach is when the clinician makes minimal effort to establish patient values and preferences, makes decision on behalf of patient?
Paternalistic or parental approach
188
what is the con with paternalstic or parental decision making?
violates patient autonomy
189
what type of decision making approach is when the clinician ascertains patient's values and preferences, makes decision on behalf on patient?
Clinician-as-perfect-agent
190
what is the con with clinician-as-perfect-agent decision making?
some experts think it's impossible to implement
191
what type of decision making approach is when the clinician provides patient with the information; patient makes the decision?
Informed decision making
192
what type of decision making approach is when the patient and clinician both bring information/evidence and values and preferences to arrive at the best course of action?
Shared decision making (bidirectional exchange)
193
What are the 3 types of talk that clinicians can have with their patients in shared decision making approach?
team talk
option talk
decision talk
*"shared DOT"*
194
type of talk that offer choices then re-assess based on pt reaction
team talk
195
type of talk where you explain all aspect of options; pros/cons
option talk
196
type of talk where you assess pt values first then offer options
decision talk
197
talk that facilitates patients' awareness that reasonable options exist and that the clinician will help the patient understand how to consider these options in detail
team talk
198
what are the 7 parts of team talk?
1. stepping back
2. offering choices
3. justify choices
4. explain different issues matter more to some people than others
5. evidence may be lacking, outcomes are unpredictable at the individual level
6. check patient's reaction
7. postpone closure
199
the act of being clear about reasonable treatment alternatives and helping patients compare them
option talk
200
what are the 6 parts of option talk?
1. check patient's knowledge
2. list options
3. describe options
4. explain harms and benefits
5. provide decision aids
6. summarize options with teach-back method
201
when you make an effort to ask about what matters most to the patients and help them form their own views and try to work with the patients to see how best to take the next steps in order to make a wise and well-considered decision
decision talk
202
what are the 4 parts of decision talk?
1. focus on values and preferences
2. elicit a preference
3. move to a decision
4. offer review of decisions
203
What are 3 methods that protect human subjects during and after clinical investigation?
belmont principles
common rules
regulatory criteria
204
contain all elements necessary to protect human subjects
regulatory criteria
205
the ethical principles and guidelines for the protection of human subjects of research, of which the procedural requirements of the Common Rule are based on
Belmont Report
206
What are the 3 principles of the Belmont Report?
1. respect for persons
2. beneficence (do no harm)
3. justice (be fair, equitable)
207
How do the requirements of the Common Rule follow the Belmont Principles?
1. respect for persons = informed consent
2. beneficence = risk minimized
3. justice = equitable subject criteria
208
What are the IRB decisions based on for approval?
the 8 criteria under Common Rule
209
What are 3 ways to minimize risk based on the IRB approval criteria?
1. precautions (additional monitoring, specific criteria, pregnancy test)
2. safeguards (restricted access to date, team with expertise)
3. alternatives (less invasive procedures, de-identified data)
*"PSA if you want to minimize risks & get IRB approval"*
210
Hypothesis must contain what 3 things to be complete?
population
variables
relationship of variables
211
Research must pass the "so what" aka FINER test which stands for what?
F = feasible
I = interesting
N = novel
E = ethical
R = relevant
212
defined as no significant difference between 2 populations? (any difference was due to small sample size or error)
Null hypothesis (Ho)
213
Which hypothesis is defined as difference between populations was due to an observed effect?
alternative hypothesis
214
the basis for all inferential statistics
probability
215
If P < 0.05 then ____ the null hypothesis
reject
*the results are significantly different*
216
If P > 0.05 then ____ the null hypothesis
retain
*results are not significantly different*
217
an alpha of 0.05 means there is a 5% probability of ____ error
type 1
218
____ error is rejecting the null when the null is actually true
type 1
219
___________ error is retaining the null when the null is actually false
type 2
"reTain Two"
220
refers to the potential for confounding factors to interfere with the relationship between the independent and dependent variables, or the degree to which the investigator draws the correct conclusions about what actually happened in the study
internal validity
221
What are the 3 threats to internal validity?
1. Intended sample vs actual subjects
2. Intended variables vs actual measurements
3. Confounding variables
222
refers to the extent to which results of a study can be generalized outside the experimental situation
external validitiy (generalizability)
223
What are the 2 threats to external validity?
1. If the intended sample and variables do not sufficiently represent the target population
2. Phenomena of interest
224
the use of a study plan with minimizing ____ can help protect the internal and external validity of the study
errors
225
What type of validity is defined as how well the assessment represents all aspects of the phenomenon under the study and uses **subjective judgements** about whether the measurements seem reasonable?
content validity
226
What type of validity is defined as whether or not a measurement seems inherently reasonable? ie does your instrument make sense?
face validity
227
What type of validity determines if a measurement conforms to theoretical constructs? ie internal performance of survey instruments
construct validity
228
What type of validity is defined as the ability of a measure to predict an outcome? ie agrees with accepted norms / measures
predictive validity
229
What type of validity is defined as a new measurement as good as the gold standard? ie agrees with already accepted methods
criterion-related validity
230
What is the extent to which results can be reproduced when the research is repeated under the same conditions and the degree to which any measuring tool controls random error?
reliability
231
unordered categories. what are 4 examples?
nominal data (names)
sex, blood type, hair color, alive vs dead
232
ordered categories. what are 2 examples?
ordinal data (ordered)
GCS, 10 pt pain scale
233
a numeric scale in which we know the difference between the intervals but there is no absolute 0
what is an example?
interval scale
degrees celcius
234
a numeric scale where we know the difference between the intervals and there is an absolute 0 representing lack of an attribute
what are 2 examples?
ratio scale
height, weight
235
Dependent variable is also called the _______ variable
outcome
236
Independent variable is also called the _______ variable
predictor
237
What are the 3 types of IRB applications?
1. Exempt from certain research regulations, but not IRB
2. Expedited; minimal risk
3. Full board; greater than minimal risk
238
Exempt
1. is IRB required?
2. does it expire?
3. does it require re- approval/continuing review?
4. can it be FDA regulated?
1. yes
2. no, but has 3 yr institutional expiration
3. exempt from continuing review
4. no
239
what are the 6 categories of Exempt?
1. educational settings
2. collecting existing data that is de-identified
3. benign behavioral interventions
4. surveys, educational tests, interviews, observations of public behavior
5. projects approved by federal dpt designed to eval public benefit or service programs
6. taste and food quality eval or consumer acceptance studies
240
Expedited
1. who reviews it?
2. can it be FDA regulated?
3. does it require re- approval/continuing review?
1. single member of IRB
2. yes
3. no, unless it's FDA regulated
241
what are the 7 categories of Expedited?
1. approved drugs, in vitro dx testing
2. blood samples under 550 mL in 8 wk period for healthy, and 50 mL in 8 wk period for unhealthy, preg, or children
3. non invasive collection of biological specimens
4. non invasive data collection
5. retrospective or prospective materials
6. data collected from recordings
7. individual or group characteristics
242
Full Board
1. who reviews it?
2. does it expire?
3. does it require re- approval/continuing review?
1. convened IRB meeting
2. approved x 1 yr
3. yes
243
what are the 2 categories of Full board?
1. greater than minimal risk
2. minimal risk; but not eligible for expedited review
244
What are the 5 steps required to submit an IRB protocol
1. complete human subject protection training
2. determine which application to submit
3. obtain appropriate signatures and approvals
4. assemble study related documents for IRB review
5. email application to IRB
245
What are the 8 components of a standard research article?
1. title
2. abstract
3. introduction
4. background and significance
5. design / methods
6. results
7. discussion/conclusion
8. references
246
What are the 10 steps to create a manuscript for submission to a medical journal?
1. Choose appropriate journal for publication
2. Title and authorship
3. Abstract
4. Introduction
5. Materials and methods
6. Results
7. Discussion +/- conclusion
8. References
9. Acknowledgements
10. Conflicts of interest
247
Where do your baseline demographics go in a standard research article?
results
248
Where is research designs, location/setting of research, IRB and ethical review, population, variables, sample size and statistical analyses in a standard research article?
methods
249
Where does inclusion and exclusion criteria go in a standard research article?
methods
250
Where do descriptive studies like stats tables and graphs go in a standard research article?
results
