• Bullet-point format for quick revision
• Key definitions, classifications, and mnemonics
• Important numbers/values highlighted
• Exam-relevant points emphasized

---

Steps in the Investigation of an Epidemic

---

Objectives of Epidemic Investigation (remember: DICE + R)

• Define magnitude in terms of time, place, person
• Identify cause, source of infection, mode of transmission
• Conditions and factors responsible for the epidemic
• Establish control measures
• Recommend prevention of recurrence

---

Steps (10 Steps - Mnemonic: "Very Clever Dogs Race Past Six Big Orange Trees Rapidly")

---

Step 1 - Verification of Diagnosis

• Confirm the diagnosis on the spot immediately
• Clinical examination of a sample of cases suffices (not all)
• Laboratory investigations confirm diagnosis - but do NOT delay epidemiological investigation waiting for lab results
• Rule out spurious reports (misinterpretation by laypublic)

---

Step 2 - Confirmation of Existence of an Epidemic

• Compare current disease frequency with same period in previous years
• Epidemic exists when observed frequency > expected frequency for that population
• Arbitrary limit: 2 standard errors above endemic occurrence = epidemic threshold (e.g., influenza)
• Common-source epidemics (cholera, food poisoning, Hep A) are easily recognized - no comparison needed
• Modern epidemics (cancer, CVD) are not easily recognized

---

Step 3 - Defining the Population at Risk

• (a) Obtain a map of the area - note landmarks, roads, dwelling units (numbered)
• (b) Count the population - house-to-house census by age and sex
  • Lay health workers used
  • Needed to compute attack rates in various subgroups
  • No denominator = no attack rate calculation

---

Step 4 - Rapid Search for All Cases and Their Characteristics

• (a) Medical survey - identify ALL cases including those not seeking care; administer epidemiological case sheet
• (b) Epidemiological case sheet (Case Interview Form) - collects:
  • Name, age, sex, occupation, social class
  • Time of onset, signs and symptoms
  • Travel, previous exposure, personal contacts (home, work, school)
  • Special events (parties, foods eaten, common vehicles - water/food/milk)
  • Blood products, injections, large gatherings attended
• (c) Searching for more cases - ask index case about others with similar illness in home, neighbourhood, school, workplace within the incubation period; check local hospital admissions
• If outbreak is large → examine a random sample

---

Step 5 - Data Analysis (Characterize by Time, Place, Person)

• Plot an epidemic curve (X-axis = time of onset; Y-axis = number of cases)
• Determines: type of epidemic, incubation period, likely exposure time
• Point source = single sharp peak
• Propagated = series of peaks at intervals of ~1 incubation period

• Prepare a spot map (dot map) - each dot = one case at address
• Identifies geographic clustering and clues about source/mode of spread

• Compute attack rates by age, sex, occupation, etc.
• Identify high-risk groups to focus control measures
• Formula: Attack Rate = (Number of new cases / Population at risk) × 100

---

Step 6 - Formulation of Hypotheses

• Based on analysis of time, place, person data
• Hypothesize:
  • Probable source of infection
  • Probable mode of transmission
  • Probable causative agent
• Should explain the majority of cases

---

Step 7 - Testing of Hypotheses

• Use analytical epidemiology - case-control or cohort study
• Case-control study most commonly used in food-borne outbreaks
  • Compare food histories of cases vs. controls
  • Calculate odds ratios or relative risks
• Statistical tests applied (chi-square, etc.)
• Hypothesis accepted if statistically significant

---

Step 8 - Instituting Control Measures

• Should not wait until investigation is complete - begin as soon as probable source/mode is identified
• Directed at:
  • Agent - treatment, disinfection
  • Host - vaccination, chemoprophylaxis
  • Environment - water purification, food safety, sanitation
• This is the primary purpose of epidemic investigation

---

Step 9 - Writing a Report

• Document all findings systematically
• Include: background, methods, results, discussion, conclusions, recommendations
• Serves as a permanent record and guides future action
• Helps in legal matters if required (food poisoning, compensation)

---

Step 10 - Recommendations for Prevention of Recurrence

• Based on identified cause and contributing factors
• Long-term measures: improved sanitation, water supply, immunization programs, health education
• Submitted to health authorities

---

Key Points to Remember for Exam

---

Source: Park's Textbook of Preventive and Social Medicine - the gold standard for PSM (MBBS)

---

Case-Control Study

---

Definition

---

3 Distinct Features (Must Know)

1. Both exposure and outcome (disease) have already occurred before the start of the study
2. Study proceeds backwards - from effect → cause
3. Uses a control/comparison group to support or refute an inference

---

Design Diagram

---

The 2×2 Contingency Table (Framework)

• Exposure rate in cases = a/(a+c)
• Exposure rate in controls = b/(b+d)
• If exposure rate in cases > controls → association exists

---

Measure of Association: ODDS RATIO (OR)

Odds Ratio = (a × d) / (b × c)

• OR = 1 → No association
• OR > 1 → Positive association (risk factor)
• OR < 1 → Negative association (protective factor)
• In case-control studies: cannot calculate Relative Risk directly - only estimate it via Odds Ratio
• OR is a good approximation of RR when disease is rare (rare disease assumption)

Classic Example (Smoking & Lung Cancer):

• Exposure rate in cases = 33/35 = 94.2%
• Exposure rate in controls = 55/82 = 67.0%
• P < 0.001 (statistically significant)

---

4 Basic Steps in Conducting a Case-Control Study

Step 1 - Selection of Cases and Controls

• Diagnostic criteria - must be defined before study (e.g., histologically confirmed cancer stage)
• Eligibility criteria - preferably newly diagnosed (incident) cases, not prevalent cases
• Sources: (i) Hospitals (single or network), (ii) General population

• Must be comparable to cases in known confounding factors (age, sex, occupation, social class)
• Sources:
  • Hospital controls (admitted for unrelated conditions)
  • General population controls
  • Relatives/neighbors (neighborhood controls)
  • Dead controls (from death certificates)
• Controls must be free from the disease under study
• Ratio: usually 1 case : 1 control (can go up to 1:4 to increase statistical power)

---

Step 2 - Matching

• Process of making cases and controls comparable with respect to confounding variables
• Types of matching:
  • Individual matching (Pair matching) - each case matched to one or more controls for specific characteristics
  • Group matching (Frequency matching) - controls selected so the overall distribution of confounders is similar
• Variables commonly matched: age, sex, race, occupation, socioeconomic status
• Overmatching (matching on too many variables) reduces efficiency

---

Step 3 - Measurement of Exposure

• Information collected about past exposure to suspected risk factors
• Methods:
  • Interviews / questionnaires
  • Medical records
  • Biological specimens
• Data collected on: diet, smoking, occupational exposure, medications, family history, etc.

---

Step 4 - Analysis and Interpretation

• (a) Exposure rates - compare frequency of exposure in cases vs. controls
• (b) Odds Ratio - estimate of disease risk associated with exposure
• Statistical test: Chi-square test (for discrete variables); p ≤ 0.05 = statistically significant
• Statistical significance ≠ causation

---

Sources of Bias in Case-Control Studies

---

Advantages vs Disadvantages

---

Key Comparisons to Remember

---

High-Yield Exam Points

• Case-control = retrospective study
• Proceeds backwards: effect → cause
• Measure of association = Odds Ratio (OR) = ad/bc
• OR approximates RR when disease is rare
• Most common bias = recall/memory bias
• Berkesonian bias = unique to hospital-based case-control studies
• Best for: rare diseases, chronic diseases, multiple risk factor analysis
• Cannot measure incidence rates

---

Source: Park's Textbook of Preventive and Social Medicine