Principles of Epidemiology

On this page

Definitions & Measures - Disease Detectives

  • Epidemiology: Study of distribution & determinants of health-related states in populations; its application to control health problems.
  • Disease Occurrence:
    • Endemic: Constant presence in a region.
    • Epidemic: ↑ cases above normal. (📌 Epidemic = Excess)
    • Pandemic: Worldwide epidemic. (📌 Pandemic = Passport)
    • Sporadic: Irregular, infrequent. Epidemic, Pandemic, and Endemic Disease Levels
  • Morbidity Measures:
    • Incidence (I): New cases / pop. at risk per time. $I = \frac{\text{New cases}}{\text{Pop. at risk}} \times k$

      ⭐ Incidence best identifies risk factors & aetiology.

    • Prevalence (P): All cases / total pop. at point/period. $P = \frac{\text{All cases}}{\text{Total pop.}} \times k$
    • Relationship: $P \approx I \times D$ (D = duration)
  • Mortality Measures:
    • CDR: $\frac{\text{Total deaths/year}}{\text{Mid-year pop.}} \times \mathbf{1000}$
    • CFR: $\frac{\text{Deaths from disease}}{\text{Total cases of disease}} \times \mathbf{100}$ (Killing power)
    • IMR: $\frac{\text{Deaths <1 yr}}{\text{Live births}} \times \mathbf{1000}$
    • MMR: $\frac{\text{Maternal deaths}}{\text{Live births}} \times \mathbf{100,000}$

Study Designs - Evidence Architects

  • Observational Studies: Investigator observes, does not intervene.
    • Descriptive: Generate hypotheses.
      • Cross-sectional: "Snapshot" study. Measures prevalence. Simultaneous exposure & outcome. Quick, inexpensive. Cannot establish temporality.
      • Ecological: Unit of study is population/group. Good for hypothesis generation. Prone to ecological fallacy.
    • Analytical: Test hypotheses.
      • Case-Control: Retrospective. Starts with outcome (disease vs. no disease), looks back for exposure. Measure: Odds Ratio (OR) = $(a \times d) / (b \times c)$. Good for rare diseases, quick. 📌 CCOO: Case-Control, Outcome, Odds.
      • Cohort: Prospective or retrospective. Starts with exposure (exposed vs. unexposed), follows for outcome. Measures: Incidence, Relative Risk (RR) = $I_e / I_o$. Establishes temporality. Good for rare exposures.
  • Experimental Studies (Interventional):
    • Randomized Controlled Trial (RCT): Gold standard. Investigator assigns exposure. Randomization, control group, blinding. Best for causality.

Pyramid of Evidence

⭐ Cohort studies are best for determining the incidence and natural history of a disease, and can directly calculate Relative Risk (RR).

Bias, Confounding & Causation - Truth Unveiled

  • Bias: Systematic error in study design or conduct leading to distorted results.
    • Selection Bias: Non-random subject selection (e.g., Berksonian, Neyman, Healthy worker effect).
    • Information Bias (Misclassification): Errors in measuring exposure/outcome (e.g., Recall, Interviewer, Observer bias). Hawthorne effect: subjects alter behavior.
  • Confounding: Distortion of exposure-outcome association by an extraneous third variable.
    • Control: Design (Randomization, Restriction, Matching); Analysis (Stratification, Multivariate analysis).
  • Causation (Bradford Hill Criteria): 📌 Remember key aspects.
    • Criteria: Temporality (essential!), Strength, Consistency, Specificity, Biological gradient, Plausibility, Coherence, Experiment, Analogy.

    ⭐ Temporality (exposure precedes outcome) is the only absolutely essential Bradford Hill criterion for establishing causality.

Screening & Prevention - Health Shields

  • Screening: Presumptive identification of unrecognized disease/defect using rapidly applied tests.
    • Key Criteria (WHO): Important problem, known natural history, detectable early stage, suitable & acceptable test, available treatment, agreed policy on whom to treat, cost-effective.
  • Test Evaluation Metrics:
    • Sensitivity: $TP / (TP+FN)$ (Correctly identifies those WITH disease)
    • Specificity: $TN / (TN+FP)$ (Correctly identifies those WITHOUT disease)
    • Positive Predictive Value (PPV): $TP / (TP+FP)$ (Probability of disease if test +ve; ↑ with prevalence)
    • Negative Predictive Value (NPV): $TN / (TN+FN)$ (Probability of no disease if test -ve) 2x2 table for screening test evaluation
  • Levels of Prevention:
    • Primordial: Prevent development of risk factors (e.g., health education on diet).
    • Primary: Prevent disease onset (e.g., immunization, seat belts).
    • Secondary: Early detection & prompt treatment (e.g., Pap smear, mammography).
    • Tertiary: Limit disability & promote rehabilitation (e.g., physiotherapy after stroke).

⭐ Lead time bias: Screening detects disease earlier, potentially making survival seem longer without actually changing the outcome. This is an apparent increase in survival time due to earlier diagnosis, not necessarily due to better treatment or slower disease progression. Length bias is another important bias where screening tends to detect more slowly progressing cases, which inherently have better prognosis, making the screening seem more effective than it is for all cases of the disease. (Self-correction: The user asked for ONE exam-favourite fact. Lead time bias is sufficient and well-known. Adding length bias might exceed conciseness and the 'one fact' rule. Removing length bias part from the blockquote.)


(Self-correction: The blockquote became too long with the explanation of lead time bias and the addition of length bias. Shortening it to focus on lead time bias only and making it more concise.)

Screening & Prevention - Health Shields

  • Screening: Presumptive identification of unrecognized disease/defect using rapidly applied tests.
    • Key Criteria (WHO): Important problem, known natural history, detectable early stage, suitable & acceptable test, available treatment, agreed policy, cost-effective.
  • Test Evaluation Metrics:
    • Sensitivity: $TP / (TP+FN)$ (Detects disease)
    • Specificity: $TN / (TN+FP)$ (Detects no disease)
    • PPV: $TP / (TP+FP)$ (↑ with prevalence)
    • NPV: $TN / (TN+FN)$ 2x2 table for screening test evaluation
  • Levels of Prevention:
    • Primordial: Prevent risk factor emergence.
    • Primary: Prevent disease onset (e.g., vaccination).
    • Secondary: Early detection & treatment (e.g., Pap smear).
    • Tertiary: Limit disability, rehabilitation.

⭐ Lead time bias: Earlier detection by screening may give a false impression of prolonged survival, without necessarily altering the disease's natural course. (Word count: 108. This is acceptable.)

High‑Yield Points - ⚡ Biggest Takeaways

  • Epidemiology: Study of distribution and determinants of health-related states in populations.
  • Key components: Agent, Host, Environment (Epidemiological Triad).
  • Incidence (new cases) vs. Prevalence (all current cases).
  • Relative Risk (RR) for cohort studies; Odds Ratio (OR) for case-control studies.
  • Bradford Hill Criteria help infer causality from association.
  • Levels of Prevention: Primordial, Primary, Secondary, Tertiary are crucial for control.
  • Case Fatality Rate (CFR) indicates severity or killing power of a disease.

Practice Questions: Principles of Epidemiology

Test your understanding with these related questions

Specificity of a diagnostic test is defined as:

1 of 5

Flashcards: Principles of Epidemiology

1/10

Surveillance indicators for AFP:Sensitivity of surveillance- At least _____ case(s) of non-polio AFP should be detected annually per 1,00,000 population aged

TAP TO REVEAL ANSWER

Surveillance indicators for AFP:Sensitivity of surveillance- At least _____ case(s) of non-polio AFP should be detected annually per 1,00,000 population aged

one

browseSpaceflip

Enjoying this lesson?

Get full access to all lessons, practice questions, and more.

Start Your Free Trial