term 3 module 4 Flashcards
Bayes’ Theorem
- Probability implies that there is an uncertainty. Bayes’ Theorem
quantifies the uncertainty. - The purpose of pretest and posttest probabilities is to increase
certainty to be more confident in the diagnosis.
Bayes’ Theorem (cont’d)
* Pretest probability (prior probability)
➢ What we think after initial exam and history
➢ Prevalence of the disorder based on population data
Being able to estimate pretest probability is central to next step!
best initial guess of likelihood
Baues theorem cont’d
* Posttest probability (posterior probability)
➢ What we think after diagnostic test
➢ The revised likelihood diagnosis
➢ based on pretest probability & likelihood ratios (LR) via a nomogram
Positive Likelihood Ratio (LR+)
𝐋𝐑+ =
𝒔𝒆𝒏𝒔𝒊𝒕𝒊𝒗𝒊𝒕𝒚/𝟏 − 𝒔𝒑𝒆𝒄𝒊𝒇𝒊𝒄𝒊𝒕𝒚 =
T positive rate / F positive rate
- Negative Likelihood Ratio (LR-)
𝑳𝑹−
𝟏− 𝒔𝒆𝒏𝒔𝒊𝒕𝒊𝒗𝒊𝒕𝒚/𝒔𝒑𝒆𝒄𝒊𝒇𝒊𝒄𝒊𝒕𝒚 =
F negative rate/T negative rate
A good test will have a high positive likelihood ratio and
a ___ negative likelihood ratio.
low
Interpretation of Likelihood Ratios
- Values close to 1.0 do NOT provide useful
information (b/c an LR = 1.0 represents a 50:50
chance of increasing or decreasing the probability of
a diagnosis). - Values below .2 and above 5.0 provide more useful
information
Guidelines for Interpreting LRs
- LR+ > 10 or LR - < 0.1
- Large and conclusive change
- LR+ = 5-10 or LR- = 0.1 - 0.2
- Moderate change
- LR+ = 2-5 or LR- = 0.2 - 0.5
- Small but sometimes important change
- LR+ = 1-2 or LR- = 0.5 -1.0
- Negligible change in pre-test probability
Prevalence =
✓ the best guess for pretest probability
✓ a proportion reflecting the No. of existing cases of a
disorder relative to the total population at a given point
in time
P=
P = number of existing cases at a give point/total population at risk
Using a Nomogram to determine
posttest probability
A nomogram was developed based on Bayes’
Theorem
whats on L middle and R
- Left: Pretest Probability (41%)
- Middle: LR+ (3.71) and LR– (0.27)
- Right: Posttest Probability (72% and 16%)
Receiver Operating Characteristics (ROC) Curves
- For diagnostic purpose, it is Helpful to determine a cutoff
score (e.g., height loss at 4 cm) along continuous scale to
determine positive or negative test. - There is a Trade-off between sensitivity and 1–specificity
- “Signal to noise”
- Coordinates are true positive rate (Y = Sensitivity) against false positive rate (X = 1 - Specificity) at different cutoff points (1 cm, 2 cm, 3 cm …. 6 cm)
Area under the curve (AUC)
represents the ability of the test to discriminate between those with and without the test condition.
A receiver operating characteristic (ROC) curve for a
perfect test (Sn = Sp = 100%)
basically a filled in square,
sensitivity (true positive) on y axis
1 - specificity (false positives) on x axis
on a ROC , whats on x and y axis
sensitivity (true positive) on y axis
1 - specificity (false positives) on x axis
if an a a box between y axis true positives (sn) and x axis false positives (1 - specificity) is split in two by the numbers, what does the ROC/AUC symbolize
50 50 chance to get the test right to appropriate rule in / out the diagnosis
ROC, Point where curve turns is best cutoff point
✓ Largest difference between true positive and false positive rate (i.e., the difference between sensitivity and (1– specificity), known as the Youden index)
✓ Best cutoff at 4 cm height loss
Critical Assessment of Study Credibility, questions to assess the validity of evidence about diagnostic tests
7 questions
Assessment of Study Credibility – Diagnostic Tests or
Measures
1. Did the investigators include subjects with all levels or
stages of the condition being evaluated by the index test
(measure)?
- Determines the usefulness of the test (measure) with
respect to distinguishing between important gradations in
the diagnosis or impairment - Investigators’ prerogative re: who to include in study
Assessment of study credibility - diagnostic tests or measures
2. Did the investigators evaluate (or provide a
citation for) the reliability of the index diagnostic
test (measure)?
- A necessary precondition for validity; may be
tested in a separate investigation
Critical Assessment of Study Credibility
3. Did the investigators compare results from the
index test to results from a “gold” (reference)
standard comparison diagnostic test (measure)?
- Verifies the test (measure) does what it’s
supposed to do - Commonly used gold standards:
✓Radiographs
✓Surgical exploration
✓A test or measure with previously demonstrated
consistency and usefulness
Critical Assessment of Study Credibility
4. Did all subjects undergo both the test (measure)
of interest and the gold standard test (measure)?
- Addresses the issue of bias due to manipulation of
the sample
Critical Assessment of Study Credibility
5. Were the individuals performing and interpreting
each tests results unaware (“masked”, “blinded”) of
the other test’s (measure’s) results?
- Addresses the issue of bias
Critical Assessment of Study Credibility
6. Was the time between application of the index test
(measure) and the “gold standard” comparison
diagnostic test (measure) short enough to minimize the
opportunity for change in the subjects’ condition?
Addresses the potential for misclassification
(inaccurate quantification) due to natural changes in
the subjects’ status (e.g., healing)
Critical Assessment of Study Credibility
7. Was the study repeated on a new set of subjects?
- Addresses the reproducibility of the results
Should You Use this Evidence?
- Is the study high quality (e.g., does the design minimize
bias)? - Are the results important enough to use?
- Is the test or measure of interest available, practical and
safe for application in the clinical setting? - Was your patient or client represented in the study
sample? - Can you estimate the pre-test probability of the disorder?
Should You Use this Evidence?
Patient’s or client’s values and preferences re:
✓Risk of injury or pain during the test or measure
✓Whether knowing the test results will produce an
important benefit
✓Cost – financial, time, personal
✓Confidence in the test & physical therapy
✓Belief in the value of scientific evidence
✓Previous experiences
