Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
126 Cards in this Set
 Front
 Back
 3rd side (hint)
3 Sources of evidence that inform EBP

Patients unique values and circumstances
Best research evidence Clinical expertise 


What is the PECO approach to formatting clinical questions

P:patient/problem
E:exposure/ C: comparison O: outcomes 


What is the highest level of evidence according to the evidence pyramid? Level 1a and b

Systematic reviews
Randomized clinical trials 


What is the lowest level of evidence according to the evidence pyramid?
Level 5 
Narrative reviews, expert opinion and textbooks



Level 2b on the evidence pyramid

Cohort studies



Level 3b on evidence pyramid

Case control studies



Level 4 on evidence pyramid

Case series
Case studies 


5 steps of EBP

1 ask a focusedclinical question
2 search for the best research evidence 3 appraise quality of research evidence 4integrate research evidence with information about pt 5 reflect on process to improve the future 


2 essential components of a background question

Question root (who, what, why....)
Condition 


2 types of clinical questions

Background
Foreground 


3 essential components of foreground questions

Patient and or problem
Exposure ( intervention)/ comparison intervention Clinical outcomes 


Source of knowledge described?
Something is thought to be true simply because people have always known it to be true 
Tradition



Source of knowledge described?
Information based on specialized training or experience 
Authority



Source of knowledge described?
Try something if it seems to work satisfactorily keep doing it 
Trial and error



Source of knowledge described?
Combines personal experience, intellectual faculties and formal systems of thought 
Logical reasoning



Sources of information described?
Systematic, empirical, controlled and critical examination of hypothetical propositions about associations among natural phenomena 
The scientific method (research)



T/F among peer reviewed journals there is no consistent process/ level of rigor by which the articles are evaluated

True



Hierarchy of study types (7)

1. Evidence based clinical guidelines
2. Systematic review and meta analysis of RCTs 3.RCTs 4. Non randomized intervention studies 5. Observational studies 6. Qualitative studies 7. Case series, case reports 


Literature reviews focused on a single question that try to identify appraise, select and synthesize all high quality research relevant to that question

Systematic review



Statistical analyses of a large collection of results from individual studies for the purpose of integrating the findings

Meta analysis



Experimental studies of cause and effect relationships between treatments and outcomes

Randomized controlled / clinical trials



Observational studies in which a defined group of people is followed over time. Aka longitudinal study

Cohort studies



Observational research where no Intervention is provided rather people who have the condition of interest are compared to people who do not

Case control studies



Descriptive research. A group/ series of case reports involving patients who were given similar treatment

Case series



Type of research or evidence that:
Describe practice. Can focus on patients or facilities. Can't prove effectiveness or test hypotheses or prove cause and effect 
Case studies/ reports



Clinical guidelines are considered obsolete after how many years?

5.8



Cochrane collaboration policy states that systematic reviews should be updated how often?

Every 2 years



Level of data measurement?
Qualitative/categorical level of data with no mathematical interpretation 
Nominal measures



Level of data measurement?
You specify only the order of the cases in greater than and less than e.g. Patient satisfaction 
Ordinal measures



Level of data measurement?
Numbers represent fixed measurement units but have no absolute zero point e.g, temperature 
Interval measures



Level of data measurement?
Fixed measuring units with an absolute zero. Numbers can be added, subtracted etc e.g. Goniometric measure of ROM 
Ratio measures



T/F
Reliability is not a property of an experiment/ study 
True. It's a property of a measurement instrument



Extent to which a measure produces the same result under different conditions ( consistency)

Reliability



Extent to which a measure indicates what it is supposed to measure

Validity



Type of validity?
The methods used on the study are correct and the results are accurate 
Internal validity



Type of validity?
The findings are applicable beyond that particular study 
External validity



Type of validity?
Measure looks like its going to measure what it's supposed to measure 
Face validity



Type of validity?
Measure covers full range of concepts meaning 
Content validity



Type of validity?
Scores obtained with one measure can be accurately compared with scores obtained with another measure that's more established 
Criterion validity



Type of validity?
Measure fits well with other measures of similar theoretical concept 
Construct validity



Measure of central tendency that can be used with any type of data

Mode



Measure of central tendency that can be used with interval, ratio and sometimes ordinal data

Mean
Median Mode 


Type of data that cannot be assessed with mean or median

Nominal



Which measures of central tendency are unaffected by extreme score?

Median and mode



Mean is the preffered measure of central tendency except?

Extreme scores or skewed distributions
Non interval data Discrete variables 


Measure of spread and variability

Range and deviation



A number that measures how far away each number in a set of data is from the mean

Standard deviation



Approximate standard deviations within 1,2,3 standard deviations of the mean for any data set?

1: 68%
2: 95% 3: 99.7% 


Parameters are?

Mathematical characteristics of populations



Statistics

Mathematical characteristics of samples



What is the EPSEM method of sampling?

Equal Probability Of Selection Method



For any trait or variable as the sample grows larger the means will will be normally distributed. This describes?

Central limit theorem



What is the implication if p value<a (alpha)?

Reject null hypothesis
Report p value Estimated effect Precision of estimate 


What is the implication if p value > a?

Fail to reject null hypothesis
Report p value Power of your study How big an effect you may have missed 


A null hypothesis states

There will be no effect on the dependent variable due to independent variable and any observed effect is due to chance unless proven otherwise



Purpose of inferential statistics?

Test whether results achieve statistical significance, eliminating chance and sampling error as cause of change of dependent variable



You could be making a mistake whether you reject or fail to reject the null hypothesis. True or false

True



When you reject the null hypothesis concluding there is a diffence between groups when in fact there is none

Type 1 error



When you fail to reject the null hypothesis concluding there is no difference between groups when in fact there is

Type 2 error



In research when we say there is a real difference between 2 groups when it was just a chance what kind of error have we made?

Type 1



When we use our data to calculate the probability that our finding is just due to chance under the null hypothesis, we calculate the?

P value



What is the alpha value?

Level of significance
How small the p value can be for the null hypothesis to be rejected How much I am willing to make a type 1 error 


A p value larger than the alpha value indicates?

Our finding is largely due to chance
We should fail to reject null hypothesis otherwise we will make a type 1 error 


A p value smaller than the alpha value indicates?

We can reject null hypothesis concluding there is a real difference between research groups



Probability of rejecting the null hypothesis when the alternative hypothesis is true ie not making a type 2 error

Power



Difference between high power and low power?
The commonly accepted power is? 
High power means decreased likelihood of type 2 error
Low power increased likelihood of type 2 error 0.80 or higher 


Factors affecting power (4)

1 sample size: larger sample, greater power
2. Large or small difference in means ( what you are looking for) 3. Variation within the groups you are studying 4. Alpha level if it's really low e.g. 0.01 it's harder to reject so power goes down 


Statement that population parameter will meet some test of difference for some specified probability / significance level for any sample

Hypothesis test



Statement that population parameter will fall within interval some specified probability / confidence level for any sample

Confidence interval



Hypothesis tests are more informative than confidence levels
True or False 
False
CI levels are more informative since they provide a range of plausible values for unknown parameters 


If alpha level is .05 confidence interval is?

95%



Samples that have no effect on each other are known as

Independent samples



Samples which are matched pairs and one group is tested more than once are known as

Dependent samples



Which type of statistical test is appropriate for
2 independent samples 2 dependent samples >2 independent samples > 2 dependent samples 
2 independent samples: Unpaired T test
2 Dependent samples: Paired T test >2 Independent samples: ANOVA >2 Dependent samples: Repeated measures ANOVA 


Type of hypotheses where the researcher specifies which of the group means he expects to be greater than the other

Directional hypothesis



Type of hypothesis stating the group means will differ but not stating which ones will differ?

Non directional hypothesis



Which type of hypothesis uses a one tailed test/ analysis

Directional hypothesis



Which type of hypothesis uses a 2 tailed test / analysis?

Non directional hypothesis



How would an alpha level of .05 be interpreted in a one tailed vs a two tailed analysis?

One tailed it would simply stand as is
2 tailed : overall probability would be split so each tail will have an alpha level of .025 


For an alpha level of .05 it is easier to reject the null with a 1 tailed test compared to a 2 tailed test? True or false

True. With a 1 tailed the .05 is in one direction but in a 2 tailed the probability has to reach .025 at one of the ends for the data to be statistically significant



If a researcher specified a directional hypothesis and uses a 1 tailed test but the direction turns out to be opposite than he expected what happens to null hypothesis?

Fail to reject the null hypothesis



What measure fits this description:
Examines relationships between variables with co efficients 1 to 0 to +1 quantifying strength and direction of association 
Correlation



Quantities for a medium correlation and
High correlation 
Medium correlation: 0.300.45
Strong correlation: 0.50  1.00 


Type of correlation?
Direct association between 2 variables (as one becomes large the other one becomes large) 
Positive correlation



Type of correlation
As the value of one variable increases the other decreases 
Negative correlation



What 3 assumptions must data fit to be analyzed using parametric statistics?

Must have a normal distribution therefore sample must be chosen randomly
Variances must be roughly equal Data are interval or ratio scale ( can be manipulated mathematically) 


Parametric or non parametric?
Test that can be done without an assumption of normality 
Non parametric



Parametric or non parametric ?
Deals with mean rather than median 
Parametric



Parametric or non parametric ?
Deals with median rather than mean 
Non parametric



Category of research that identifies cause and effect between variables and comprises of: RCTs, non RCTs, single subject designs

Experimental research



Best control for bias is ?

Randomization



3 ways to randomize for a study

1. Random selection of a sample
2. Random assignment of sample to groups 3. Random assignment of groups to treatment and control conditions 


What category of research do these fall in: developmental, normative, qualitative, evaluation research ?

Descriptive research



Which category of research identifies relationships between variables usually using correlation statistics e.g. Cohort and case control studies, methodological studies

Exploratory research



Category of research that rigorously integrates findings from more than one study on the same topic e.g. Meta analysis, evidence based clinical guidelines, systematic reviews

Integrative research



3 explanations for an observed effect in RCTs

Tx had an effect
Chance variation between 2 groups Bias 


Sources of potential bias

Natural history of a disease
Placebo effect Drop outs ( intention to treat analysis) 


Efficacy vs effectiveness

Whether an intervention works under ideal circumstances vs in the real world



Systematic or random error?
Captured in the validity of the inference rather than its precision 
Systematic error



Systematic or random error?
Causes distortion of truth in a predictable direction 
Systematic error or bias



Systematic or random error?
Occurs because one cannot study everyone, obscures real difference and is reduced with larger sample sizes 
Random error or chance



Quantitative or qualitative research?
Explicitly acknowledges bias instead of attempting to eliminate it 
Qualitative research



Selection bias is?

Bias resulting from a problem in selection of subjects or in the process that determines which subjects drop out of the study



Sources of selection bias

Inappropriate population studied
Inadequate participation Selection of most accessible subjects or of volunteers 


what is prospective subject recruitment?

Selecting subjects as they come along and present themselves to researchers



Retrospective subject recruitment is?

Potential subjects are identified and contacted by the researchers to participate in the study and may involve review of pts charts to collect data

* can also occur such that actual patients are not seen but their information is compared to test hypothesis


Identify common types of sample selection
Consecutive sample Selective sample Convenience sample 
Consecutive: subjects chosen on a first come first serve basis
Selective: sampling that screens out certain characteristics Convenience: selected at the convenience of researcher 


True or false?
Reducing number of observers increases the bias 
False. Having the same observer gives greater reliability than having 2 or more



Placebo effect

Measurable, observable or felt improvement in health not attributable to actual treatment



Why does blinding increase internal validity?

Decreases placebo effectl



What is NNT?

Number needed to treat for one person to benefit associated with confidence interval levels and where CI levels overlap there is no difference between groups



Relative risk is?

Probability of developing a disease ( or other event) relative to exposure
RR =1 no difference in risk between 2 groups RR <1 less likely in experimental than control (.8 or 80%) RR>1 more likely in experimental than in control (1.25 or 125%) 


Odds ratio?

Ratio of the odds of an event occurring in one group to the odds of it occurring in another
= 1 event equally likely in both groups (1:1) >1 more likely in first group (2:1) <1 less likely in first group (.5:1) 


Which is recommended as an accurate measure when the disease in consideration is more common,Relative risk or odds ratio?

Relative risk because odds ratio can overestimate and magnify risk



What level on the evidence pyramid do literature reviews fall under and why?

Level 5, grade D (expert opinion)
Because the possibility of bias is large since researchers can pick whichever articles they want 


The statistical analysis of a a large collection of results from individual studies for the purpose of integrating the findings

Meta analyses



Which one is more comprehensive and transparent?
Traditional peer reviewed literature review OR textbook summary of literature 
Traditional peer reviewed literature review



Which has more potential for bias?
Course review paper OR clinical expert opinion 
Clinical expert opinion



What is the main outcome of a meta analysis?

Overall magnitude of the effect



Why are meta analysis subject to publication bias?

They reflect only what is published and statistically significant effects are more likely to be published hence biased toward Positive outcomes



What is the cohens d measure?

A generic dimensionless measure for standerdized change in mean used in meta analysis



On a forest plot what do the following represent?
Line Box Size of box Line w/ arrow Diamond Horizontal line crosses vertical line 
Line:CI
Box: point estimate Size of box: magnitude of effect Line w/ arrow: CI so large it goes off the page Diamond: summary of point estimates Crossing line: effect could be due to chance 


Recommended time span of publications for a systematic review?

2025 years if not researcher should indicate why they limited it more



How many raters are typically used to select studies for a systematic review?

At least 2 each blinded to the other
A 3rd rater used to break ties 


Process of assessing methods and results of each study to determine if the study is adequate to answer the question is known as

Critical appraisal



Recommended number of databases to collect literature for a systematic review

More than 2

