MCQ Principles of A&M Flashcards
Validity
Accuracy of a study
Reliability
Consistency and repeatability of a study
Quantitative data
- Numbers
- Statistics
Qualitative Data
- Survey
- Questionaire
Standardised assessments
- No variation
- Same method/procedure
- usually the same outcome
Non standardised Assessment
-Hard to repeat
-Variety/change
-Less valid
- Different outcome
What are the four types of instrument validity?
-Face
-Content
-Criterion
-Construct
What are the two types of study validity?
-Internal
-External
What are the four types of reliability?
- Internal
-Test/Retest
-Inter-rated
-Intra-rater
What are the four levels of measurement? (NOIR)
-Nominal
-Ordinal
-Ratio
-Interval scale
Specificity
How well a test rejects people without the condition or disease.
Sensitivity
How well test detects people with the condition or people
Audit
-Quality improvement process+ tool
-Clinical governance
What are the two types of audit
-Peer review
-Surveys/Focus groups
Central tendency (Mean)
A value that attempts to describe a set of data by identifying the central position.
Sensitivity
TP / TP + FN
Specificity
TN / TN + FP
Accuracy
True/ Everything OR
TP+TN/ TP + TN + FP + FN
How to plan an Audit
Potatoes see chips as an imposter
1.Plan- get permission, gather team + plan
2.Set standards ( Gold standard)
3.Collect data
4. Analyse + interpret fundings
5. Action plan + Present findings
6.Implement change + Reaudit
Internal reliability
All components involved are testing the same thing
Test/Retest reliability
Repeat test to ensure same answers each time.
Inter-rater reliability
Multiple testers all testing the same thing to see if they get the same results
Intra-rater reliability
One single tester repeating test to try and get the same results each time
Internal validity (SV)
Relates to how well a study is conducted and if it answers the rears h question.
External validity (SV)
How well a study can relate to the real world setting
Instrument validity
Making sure the indicator chosen measures appropriately what you want it to measure
Face Content Validity
Wether the content of the test appears to be suitable to its aim
Criterion Validity
Evaluates how accurate a test measures the outcome it was designed to measure.
Construct validity
Wether your test or measure accurately assess what its meant to
Study validity
How well you can generalise a study towards its participants without bias.
Nominal
-Name
-Percentages
-Countries
- Painful and not painful
-Living and dead
E.g Alive/dead
Male/Female
Tall/short
Ordinal
-Names in order
-Mild pain,Medium pain, severe pain
-High income, low income
E.g 1st, 2nd, 3rd
Ratio
Intervals with ‘true zero’
E.g VAS
Interval scale
Numbers with equal intervals
E.g Pain Intensity scale
Diagnostic accuracy
-The accuracy of a test is its ability to differentiate between the patient with a condition and healthy cases correctly.
-To estimate the accuracy of a test we should calculate the proportion of true positive and true negative in all evaluated cases.
High sensitivity =
Low specificity and people may receive a false positive outcome.
Good test ?
Ideally high sensitivity ( Approx > 90%)
AND
High specificity (Approx > 90%)
What validity do the terms sensitivity and specificity relate to ?
Criterion
What is one of the simplest ways to measure criterion validity?
- To compare it to known standard
(Gold standard, reference standard, benchmark and traditional test)
If the test outcome is POSITIVE
SpPin- When specificity is high and the test is positive you can be confident , that its likely to be a true positive
Rule IN the condition
I SpPin
High specificity = test is positive= true positive
Why? A high specificity is only possible when there is a very small number of false positives, therefore ifyou do get a positive result from thetestyou can be confident thatit’slikely to be a true positive
If the test outcome is NEGATIVE
SnNout- Sensitivity is high and the test is Negative you can be confident this is a true negative
Rule OUT condition
High sensitivity = test is negative = true negative
Why? A high sensitivity means (look at the equation) there has to be a very small proportion of false negatives, therefore if you get a negative response it is much more likely to be positive than negativ