Test 1 Flashcards
Define “Evidence-Based Medicine”
also referred to as “evidence-based health care” and “evidence-based practice”
The use of evidence in clinical decision making rather than unquestioning reliance on knowledge gained from experts or teachers or familiar practice habits.
What are the EBPT focus areas?
Examination, Evaluation, Diagnosis, Prognosis, Intervention, Outcomes
What is the EBPT process?
- Question formation
- Search for relevant evidence
- Critical appraisal of evidence
- Determination of applicability of evidence
- Consideration of evidence in combination with clinical expertise, patient/client values and preferences
What are the factors important to EBPT?
- Therapist knowledge and willingness to accept new evidence
- Access to evidence through electronic databases
- Time to participate in the EBPT process
What is evidence?
Empirical observation about the apparent relation between events
List the types of research designs from most bias control to least.
- Experimental designs
- Quasi-experimental designs
- Nonexperimental designs
- Case report/ anecdote
Why should you publish whether or not the experiment works?
You can prevent someone else from wasting their time and resources.
Explain an experimental design.
Groups are randomly assigned to an experimental group or a control group
Explain a quasi-experimental design.
It is similar to an experimental design but does not have a control group.
Explain a nonexperimental design.
surveys, questionnaires, expert opinions, historical reviews, curriculum design, etc.
Explain a case report.
An example would be reviewing patient records.
What does a blind study mean?
The patient doesn’t know which group they are being assigned to.
What does a double blind study mean?
Neither the patient nor the researcher know which group the patient is assigned to.
What is a variable?
A characteristic of a group or a person.
Example- demography, gender, age, weight
What is an independent variable?
A variable that can be changed based on the type of research you are going to do.
What is a dependent variable?
The variable that depends on the independent variable.
What are the different ways to think about research designs concerning time?
- Retrospective (already existing data)
- Prospective (new data in real time)
- Cross-sectional (a single point in time)
- Longitudinal (over an extended period of time)
What is the hierarchy of evidence quality from highest to lowest?
- Randomized clinical trials
- Cohort studies
- Case control studies
- Case-series study
- Expert opinion
Which type of research is often used in PT?
Systematic Reviews
Why do we do research and how to we start?
We do research when we have a question. We look up information in order to answer that question. We begin with a small-scale pilot study.
What is a SLAP injury?
Superior Labrum Anerio-Posterior
What might EBPT questions pertain to?
- Pathology
- Medical and surgical management options
- Tests and measures
- Predictive factors for health status
- Benefits and risks on interventions
- Outcomes
What kind of questions can we ask concerning pathology in PT?
Why does and injury happen?
What kind of questions can we ask concerning medical and surgical management options?
What type of treatment has been done?
What are some examples of tests and measures in PT?
Range of motion, proprioception, etc.
What kind of questions can we ask concerning predictive factors for health status?
How will the person recover from the injury?
What kind of questions can we ask concerning benefits and risks of interventions?
Does the treatment work?
How can we question clinical prediction rules?
Develop strategies for a person to reach certain standards (like a certain level=go back to work).
What kind of questions can we ask concerning Outcomes?
What is the end result?
What does a background question do?
Reflect desire to understand the nature of the problem, help clincians better understand and treat situations
This can affect precautions, contradictions, and exercise limits.
What does a foreground question do?
Allow clinicians to make decisions about specific clinical management issues
These can involve diagnostic tests, clinical measures, prognostic factors, interventions, clinical prediction rules, and outcomes.
What are the 4 key elements of a foreground question?
- Patient/Client details
- A specific factor of interest
- A basis for comparison of that specific factor
- The consequence of interest for the factor
What are the major EBPT databases?
- US National Library of Medicine (pubmed)
- Cumulative Index of Nursing and Allied Health Literature (CINAHL)
- Cochrane Library
- Physiotherapy Evidence Database (PEDro)
- APTA’s Hooked on Evidence
Which is the database most often used by PTs?
PEDro
What is research?
The studying of a problem in pursuit of a definite objective through employing precise methods with due consideration to the adequate control factors other than the variable under investigation and followed by analysis according to acceptable statistical procedures.
Define and give an example of deductive reasoning
Top-down (more general–>more specific)
Every day I leave for work at 8.
Every day I arrive on time.
If I leave for work at 8 today, I will be one time.
Define and give and example of inductive reasoning.
Bottom-Up (specific observation–>broader generalization)
Today I left at 8 and arrived on time
Therefore, every day I leave at 8 I will be on time.
Who can do research?
ANYBODY
What are the 5 general phases of the research process?
- Develop a research question
- Design of Study/ Methods
- Data collection
- Analysis
- Summary and Conclusion
What does the development of the research question involve?
rationale, review, research hypothesis
What does the design of study/ methods involve?
subject selection, testing methods, methods
What does the analysis involve?
reduction of data, testing hypothesis
What does the summary and conclusion involve?
reports, presentation
Which sections do you include in a research proposal?
Intro, Review, and Methodology
Which hypothesis are you striving to reject?
The null hypothesis
Where does statement of the hypothesis go?
Under methods
Where does the decision to accept or reject the null hypothesis go?
After analysis of data, before interpret findings
Explain basic research.
to obtain empirical data that can be used to develop, refine, and test theory. (Acquisition of new knowledge)
Explain applied research.
directed toward functional applications, testing
What are the types of nonexperimental research?
Descriptive, Historical, Evaluation, Methodology
What are some ethical issues in clinical research?
- Integrity of the researcher
- Use of control groups
- Protection of human rights
What are the elements of informed consent?
- Subjects must be fully informed
- Subjects information should be confidential and anonymous
- The informed consent form must be written in a simple language
- The researcher must offer to answer questions at any time.
According to the institutional review board, what are consent elements?
- consent must be voluntary
- Special consideration to subjects who are vulnerable
- subjects must be free to withdraw consent at any time
What are the different levels of measurement?
Nominal
Ordinal
Interval
Ratio
Describe the nominal level of measurement.
Mutually Exclusive- male and female, yes and no
Describe the ordinal level of measurement.
Manual muscle test (0,1,2,3,4,5) rank.
Describe the interval level of measurement.
Intervals between are consistent, but nothing is zero (equal intervals between numbers).
Describe the ratio level of measurement.
Uses zero (Units with equal intervals…zero is true.)
What is the mean?
The average
What is the median?
Middle number(s)… 50th percentile
What is the mode?
The number with the highest frequency of occurrence.
What is the range?
The difference between the lowest and the highest
What is the standard deviation?
The square root of the variance, measured from the mean
What is the variance?
The variability of a score within a distribution (How far it is spread from the middle)
What percentage of the distribution lies within 1 standard deviation of the mean?
68%
What percentage of the distribution lies within 2 standard deviations of the mean?
95%
What percentage of the distribution lies within 3 standard deviations of the mean?
99.7%
How do you find the z-score?
(x-μ)/σ
(#-mean)/SD
What type of error is it when you reject a true null hypothesis?
Type I (alpha)
What type of error is it when you accept a false null hypothesis?
Type II (beta)
You have H0: μ1=μ2 and H1: μ1≠μ2. What type of test do you use?
a two-tailed t-test (you use 0.25 on each side)
You have H0:μ1 ≤μ2 and H1:μ1>μ2. What type of test do you use?
a one-tailed t-test (you use 0.5 on that side)
You have Ho:μ1=μ2 and H1:μ1>μ2. What type of test do you use?
A one-tailed t-test (you use 0.5 on that side)
What is the difference between using a one-tailed t-test and using a two-tailed t-test?
When you use 0.5 on the one-tailed t-test it means there is a greater margin of error.
What is the most common error made my clinicians and what type of test do they most often use?
Type 1 errors (reject null)
Two-tailed t-tests
What is the P-value?
The odds of not rejecting the null hypothesis
What are the two different ways of writing the null hypothesis?
H0: μ1=μ2
H0: μ1≤μ2
What is it called when a normally distributed set of data is converted to z scores?
The standard normal distribution
Which direction is the long tail on a positively skewed distribution graph?
The right (the graph leans to the left)
Which direction is the long tail on a negatively skewed distribution graph?
The left (the graph leans to the right)
On a positively skewed graph, where are the parameters located?
Mode, Median, Mean from the left
On a negatively skewed graph, where are the parameters located?
Mean, Median, Mode from the left
Where is the mean located when a graph is skewed
toward the long tail side
What is criterion referencing?
Interpretation of a score based on its actual value
What is norm referencing?
Interpretation of a score based on its value relative to a standard or “normal score”
What is criterion referencing?
Interpretation of a score based on its actual value
What is norm referencing?
Interpretation of a score based on its value relative to a standard or “normal score”
What is a body of interpreted principles that present a systematic of phenomena?
A theory
A theory is _____ and _____.
testable and tentative
What is a hypothesis?
A statement of the expected relationship between variables.
What are norm-referenced findings derived from?
previously tested subjects
What are criterion-referenced findings compared to?
a previously established ‘absolute’ standard
What is measurement reliability?
The amount of variability in a measure
What is internal consistency?
The extent to which tests of procedures asess the same
What is parallel forms?
Used to asess the consistency of the results of two tests constructed in the same way from the same content data (surveys, questionnaires)
What is split half?
randomly divide all items that purport to measure the same construct into 2 sets. Administer the instrument to a sample of people and calculate the total score for each randomly divided half
What does inter-rater reliability measure?
the consistency of measurement values between different individual raters
What does intra-rater reliability measure?
the consistency of measurement values of one individual
What is measurement validity?
the ability of a measure to capture what it is intended to capture.
What is face validity?
The simplest and weakest form of measurement validity. (Does this instrument appear to be the appropriate choice to measure this variable? Yes/No)
What is content validity?
When the instrument represents all of the relevant facets of the variable it is intended to measure (Fitness variables is validated by using running, swimming, jumping, body fat, etc. tests)
What is construct validity?
The degree to which the measure reflects the definition of the concept or construct it represents. (example- you have to construct what patient satisfaction is in order to measure it. Like a questionnaire)
What is concurrent criterion validity?
If there is a relationship/correlation between the gold standard test and the one you create, then you can use your test.
What is predictive criterion validity?
Prediction of a variable if the relationship exists
What is responsiveness to change?
The smallest amount of difference the instrument can measure (the more values on a scale the more responsiveness to change)
What need tos be established in order for responsiveness to be determined?
Construct validity
What is the standard error of measurement?
The extent to which observed scores are disbursed around a “true score” [Range for which a measure can fall based on SD]
What is it called when no change is seen in scores even with lower performance?
Floor effect
What is it called when no change is seen in scores even with better performance?
Ceiling effect
What is the degree to which the relationship between the independent and dependent variables is free from the effects of extraneous factors? (AKA- the independent variable is the only on changing the dependent variable)
Internal validity
What is the degree to which results of a study can be generalized to persons or settings outside the experimental situation?
External validity
What is the total group of interest?
the population
What is the subgroup of the group of interest?
the sample
What is the group to whom researchers hope to generalize their findings?
the target population
What is the group of potential research participants who are actually available for a given study?
the accessible population
What is a probability sample as opposed to a non-probability sample?
Probability involves some kind of randomization and is unbiased. These samples tend to be less variable and better approximations of the population.
What is simple random sampling?
When you put all of the possible names of a group you are studying into a computer and have it randomly select a certain number, like 100.
What is systematic sampling?
Choose participants from a population using specific intervals (5,10,15,20, etc)
What is stratified sampling?
Grouping individuals into homogenous groups… Subjects are proportional based on their category (like 10% freshmen 20 %sophomore so 10 f and 20 s out of 100)
What is Cluster/Multistage sampling?
Large subgroups are randomly selected first, and smaller units from these clusters. (Selected 10 random schools, select groups from them like DPT 1 2 or 3)
What is convenient sampling?
Like using our DPT group to represent all DPT students because we are easily accessible
What is snowball sampling/chain referral?
Choose one person who recruits his friends who recruit their friends
What is purposive sampling?
Want to study the population of Leavenworth so you choose the school. Specific choices are made and this is very biased.
What are the types of probabilistic sampling?
Simple random, Stratified, Systematic, Cluster/Multistage
What are types of non-probabilistic sampling?
Convenient, snowball, purposive
What is the effect size?
Change in status/how much change has taken place. The mean change score divided by the standard deviation of the pre-treatment score.
What is (1-beta)?
The ability of a statistical test to find a significant difference that really does exist
What is the statistical concept indicating the number of values within a distribution that are free to vary?
Degrees of freedom (n-1)
In a single case design, what does A=
baseline data…. can also be withdraw the protocol
In a single case design, what does B=
protocol/independent variable
What is used to quantitatively describe the strength and direction of a relationship between two variables?
Correlation Coeffecients
Y=a+bx
Y=dependent variable, X=independent variable, a=constant (intercept), B=slope
What are the general guidelines for correlation?
0-0.25 no relation
- 26-0.49 low/fair correlation
- 5-0.69 moderate
- 7-0.89 high
- 9+ very high
What does the r-square value (coeffecient of determination) indicate?
The percentage of variance shared by the variables
What is regression?
prediction
What is degrees of freedom in a chi2 table?
(r-1)(c-1)
What type of test do you use with observed and expected frequency?
one tailed test (chi square)
What does a chi-square test determine?
whether there is a significant different between the expected frequencies and the observed frequencies
Categories in a chi2 test are:
exhaustive and mutually exclusive
Frequencies represent ______ not ______.
represent individual counts, not ranks or percentages.
What does the null hypothesis in a chi2 test say?
There is no significant difference between expected and observed frequencies.
What is the point at which you can say with 95% confidence that the difference is not due to chance alone?
The level of significance (set at .5)
What is the Chi2 formula?
X2= Sum(O-E)2/E
What happens when the calculated value is less than the critical value?
accept the null
