Socratic method of communication = Two way communication
Tableau = skit / roleplay
Rate: Measures the occurrence of an event per unit size of population over a specified period.
Ratio: Compares two independent quantities (x:y) where the numerator is not part of the denominator.
Proportion: A specific ratio where the numerator is included in the denominator, usually expressed as a percentage (x/(x+y)×100).
Index case: The first case/patient that comes to the attention of the investigator or health authorities.
Primary case: The individual who introduces the disease/infection into a susceptible population or group (the actual patient zero who may or may not be identified first).
Secondary case: A person who contracts the disease from exposure to a primary case.
💡Designs
Study
Starts with
Direction
Main measure/use
Cross-sectional
Population
Snapshot
Prevalence
Case-control
Disease
Backward → exposure
OR
Cohort
Exposure
Forward → disease
RR, incidence
RCT
Intervention assigned
Forward
Risk, RR, treatment effect
Ecological
Population/group
Usually observational
Population-level association
Case report
One case
Description
Rare/unusual disease
Case series
Several cases
Description
Rare/unusual disease
💡Levels of Prevention
Traditionally there are 3 levels:
Primary prevention → prevent disease from occurring
Secondary prevention → detect disease early + treat it
Tertiary prevention → reduce disability/restore function
There is also primordial prevention, which comes before primary prevention. Prevent the development of risks themselves.
💡Phases of Drug/Clinical Development
Stage
Main purpose
Subjects
Key FCPS clue
Preclinical ⭐
Safety, toxicity, pharmacology before human testing
Laboratory + animals
Before humans
Phase 0
Very early exploratory pharmacokinetic/pharmacodynamic studies
Very small number of humans
Microdose / exploratory
Phase I ⭐
Safety, tolerability, dose range
Small number of humans, often healthy volunteers
Is it safe?
Phase II ⭐
Efficacy + dose finding
Patients with the disease
Does it work?
Phase III ⭐
Confirm efficacy & safety; compare with standard treatment/placebo
Large number of patients
Confirm/compare
Phase IV ⭐
Post-marketing surveillance
General population
After approval
💡Epidemiological Measures
Measure
Formula
Mainly used in
What it tells you
Relative Risk (RR) ⭐⭐⭐
Risk in exposed / Risk in unexposed
Cohort
How many times greater/lower is the risk?
Odds Ratio (OR) ⭐⭐⭐
ad / bc
Case-control
Ratio of odds of exposure/disease
Attributable Risk (AR) ⭐⭐⭐
Riskₑ − Riskᵤ
Cohort
Excess risk due to exposure
💡ACCURACY, PRECISION, VALIDITY AND RELIABILITY
Term
Definition
Dental Example
Key Point
Accuracy
Closeness of a measurement to the true value
A scale reads 70.1 kg when the true weight is 70.0 kg
Systematic error affects accuracy
Precision
Closeness of repeated measurements to one another
Repeated readings are 70.1, 70.1 and 70.2 kg
Mainly reflects random variation
Validity
Extent to which an instrument measures what it is intended to measure
A questionnaire genuinely measures oral-health knowledge
A reliable test may still be invalid
Reliability
Consistency or reproducibility of measurements
Two examiners record similar DMFT scores for the same patient
Reliable results are reproducible
Robustness
Ability to maintain performance under reasonable variations in testing conditions
An instrument remains dependable under routine operating conditions
Not identical to reliability
💡Types of reliability
Type
Definition
Example
Test–retest reliability
Same instrument gives similar results on separate occasions when the measured characteristic has not changed
A patient repeats the same knowledge questionnaire
Intra-rater reliability
Same examiner produces consistent ratings
One dentist records similar periodontal scores on repeat assessment
Inter-rater reliability
Different examiners produce consistent ratings
Two dentists independently assess the same radiographs
Internal consistency
Items intended to measure the same construct show consistency
Questions in an oral-health knowledge scale assess a coherent construct
💡DIAGNOSTIC TESTS
💡SENSITIVITY, SPECIFICITY, PPV AND NPV
These are important concepts in screening and diagnostic testing.
The 2 × 2 table
Disease Present
Disease Absent
Total
Test Positive
True Positive (TP)
False Positive (FP)
TP + FP
Test Negative
False Negative (FN)
True Negative (TN)
FN + TN
Total
TP + FN
FP + TN
Total tested
Definitions:
True positive (TP): Disease is present and the test is positive.
False positive (FP): Disease is absent but the test is positive.
False negative (FN): Disease is present but the test is negative.
True negative (TN): Disease is absent and the test is negative.
💡Formulas and definitions
Measure
Formula
Meaning
Sensitivity
TP / (TP + FN) × 100
Among people with disease, how many test positive?
Specificity
TN / (TN + FP) × 100
Among people without disease, how many test negative?
Positive predictive value (PPV)
TP / (TP + FP) × 100
Given a positive test, what is the probability the person has disease?
Negative predictive value (NPV)
TN / (TN + FN) × 100
Given a negative test, what is the probability the person does not have disease?
Overall accuracy
(TP + TN) / Total tested × 100
Proportion of all tested people correctly classified
Worked example
A dental screening test is performed on 200 people.
40 have disease and test positive (TP).
10 have disease but test negative (FN).
30 do not have disease but test positive (FP).
120 do not have disease and test negative (TN).
Measure
Calculation
Result
Sensitivity
40 / (40 + 10) × 100
80%
Specificity
120 / (120 + 30) × 100
80%
PPV
40 / (40 + 30) × 100
57.1%
NPV
120 / (120 + 10) × 100
92.3%
Overall accuracy
(40 + 120) / 200 × 100
80%
A negative result is reassuring in this example, but a positive result does not mean that disease is certain.
Effect of prevalence on predictive values
When Disease Prevalence Increases
Usual Effect
PPV
Increases
NPV
Decreases
Sensitivity
Does not mathematically depend on prevalence
Specificity
Does not mathematically depend on prevalence
Sensitivity and specificity may nevertheless vary across populations because of differences in disease severity, patient selection and testing conditions.
Likelihood ratios
Measure
Formula
Interpretation
Positive likelihood ratio (LR+)
Sensitivity / (1 − Specificity)
How much more likely a positive result is in someone with disease than without disease
Negative likelihood ratio (LR−)
(1 − Sensitivity) / Specificity
How much a negative result changes the odds of disease
A high LR+ helps rule disease in, while a low LR− helps rule disease out. Predictive values still depend on the patient's pre-test probability.
