GROUP A - LONG ANSWER QUESTIONS (15 MARKS EACH) - MODEL ANSWERS

Q1. Radiation effects study among nuclear plant workers over 5 years [JNM] (2+8+5=15)

(a) Ideal Study Design (2 marks)

(b) Steps in Conducting this Study (8 marks)

• Exposed cohort: All workers currently employed at the nuclear plant who are exposed to radiation. Classify them by dose levels (low, medium, high) using personal dosimeters and workplace monitoring records.
• Unexposed (comparison) cohort: Non-radiation-exposed workers from the same plant (e.g., administrative staff) or workers from a similar industry without radiation exposure. This group must be comparable in age, sex, smoking history, socioeconomic status.
• Ensure all subjects are free of the disease (leukaemia, cancers) at the start.

• Complete medical examination of all cohort members at entry.
• Collect demographic data: age, sex, duration of employment, type of work.
• Obtain blood counts (CBC), clinical chemistry, and any relevant biomarkers.
• Record cumulative radiation dose from existing employment records.

• Personal dosimetry (thermoluminescent dosimeters, film badges) to measure individual radiation exposure throughout the study.
• Environmental surveys of the workplace for ambient radiation levels.
• Review of radiation exposure records and occupational health registers.
• Classify workers according to: (a) whether exposed or not, and (b) level/degree of exposure.

• Internal comparison: Workers in the same plant with minimal or no radiation exposure.
• External comparison: General population mortality/incidence rates (using national/regional data).
• Both comparisons may be used for different analytical purposes.

• Periodic re-examination (e.g., annually) of all cohort members.
• Active surveillance: regular CBCs, physical examinations, recording of new diagnoses.
• Passive surveillance: monitoring hospital admissions, cancer registries, death certificates.
• Minimize loss to follow-up by maintaining good communication, updating contact details.

• Record incidence of radiation-related diseases: leukaemia, aplastic anaemia, solid tumours (thyroid, lung, etc.), cataracts, infertility.
• Use standardized diagnostic criteria.
• Outcome assessment should be blinded to exposure status where possible.

• Calculate incidence rates in exposed vs. unexposed groups.
• Compute Relative Risk (RR) = Incidence rate in exposed / Incidence rate in unexposed.
• Calculate Attributable Risk (AR) = Incidence in exposed - Incidence in unexposed.
• Use life table methods or Cox proportional hazards regression for time-to-event analysis.
• Adjust for confounders (age, smoking, other occupational exposures).

• Apply Hill's criteria to assess causality.
• Report findings with confidence intervals and p-values.
• Disseminate findings to plant management, regulatory bodies, and the scientific community.

(c) Disadvantages of a Prospective Cohort Study (5 marks)

1. Expensive and time-consuming: Large numbers of people must be followed for years; costs of repeated examinations, laboratory tests, and surveillance are substantial. For rare outcomes, very large cohorts are needed.
2. Loss to follow-up (Attrition bias): Workers may change jobs, relocate, or die from unrelated causes. Systematic loss to follow-up can bias results, especially if those who drop out differ from those who remain.
3. Not suitable for rare diseases: If the disease incidence is very low (e.g., a specific rare cancer), a very large cohort and very long follow-up are required, making it impractical.
4. Changes over time: During the 5-year follow-up, diagnostic criteria, measurement methods, or occupational practices may change, introducing inconsistencies in data collection.
5. Healthy worker effect: Workers are typically healthier than the general population (they have to be fit enough to work), which may underestimate the disease risk when comparing to general population rates.
6. Confounding: Multiple occupational and lifestyle exposures (smoking, chemical exposures) co-exist, making it difficult to attribute disease solely to radiation.
7. Protracted interval before results: Unlike a case-control study, results are not available for 5+ years, delaying public health action.

• Park's Textbook of Preventive and Social Medicine, pp. 89-95

Q2. Village outbreak of acute diarrhoea after community feast [MMC] (2+8+3+2=15)

(a) Epidemic or Outbreak? (2 marks)

• An epidemic is defined as the occurrence of cases of illness in a community or region, clearly in excess of normal expectancy (Park).
• An outbreak is a localized epidemic, restricted to a specific community, institution, or small geographic area.

(b) Steps in Investigating this Occurrence (8 marks)

• Collect stool specimens from affected individuals for culture and microscopy (identify pathogens: V. cholerae, Salmonella, Shigella, E. coli, Staph. aureus, etc.).
• Confirm that the cases represent a genuine outbreak and not a reporting artifact.
• Define a case definition: "Any resident of the village who attended the community feast on [date] and developed ≥3 loose stools within 24 hours between [date range]."

• Collect data on current cases vs. expected baseline (endemic) level for that village.
• Establish the epidemic: 85 cases in 3 days far exceeds normal expectancy.

• Prepare an epidemic curve (histogram of cases by time of onset).
• If the epidemic curve shows a sharp peak with rapid rise and fall, it indicates a point-source epidemic (consistent with a community feast).
• Collect data on: name, age, sex, address, date and time of symptom onset, foods eaten at the feast, clinical features.

• Person: Who is affected? Attack rates by age, sex, occupation. Are all attendees at risk, or only those who consumed specific items?
• Place: Are all cases clustered at the feast location or in the village? Map cases geographically.
• Time: Plot epidemic curve to confirm point-source pattern (incubation period peak).

• Based on foods served at the feast, prepare a food-specific attack rate table.
• Hypothesize which food item is the likely vehicle (e.g., rice, curry, sweets, water supply).

• Calculate food-specific attack rates for those who ate vs. those who did not eat each food item.
• Calculate Relative Risk for each food item.
• The food with the highest attack rate among those who ate it and lowest among those who did not eat it is the likely vehicle.
• Collect food samples (if available) for laboratory analysis.

• Inspect cooking area, water source, food storage and preparation methods.
• Interview cooks and food handlers (check for illness, wounds, hygiene).
• Test water supply (for coliform count, V. cholerae).

• Prepare a report for the health authorities.
• Identify the source and vehicle of infection.
• Institute control measures (see below).
• Continue surveillance.

(c) Epidemiological Study Design to Identify Source of Infection (3 marks)

• The exposed group (feast attendees) is well-defined and identifiable.
• You can enumerate all feast attendees (cohort = all who attended).
• Among this cohort, identify who ate each food item (exposure) and who developed diarrhoea (outcome).
• Calculate food-specific attack rates and Relative Risk for each food item.
• This design directly identifies the source/vehicle of the outbreak.

(d) Two Immediate Control Measures (2 marks)

1. Source control: Identify and immediately stop distribution/consumption of the implicated food item. If a water source is implicated, chlorinate or seal it. Quarantine remaining food samples for laboratory testing.
2. Treatment of cases: Oral rehydration therapy (ORT)/IV fluids for dehydrated cases; appropriate antibiotics if indicated (e.g., for cholera). Notify the district health officer; set up treatment facilities in the village.

Q3. Time trends in disease occurrence; Transmission dynamics; Indications of cohort study [MldMCH] (6+6+3=15)

(a) Types of Time Trends in Disease Occurrence (6 marks)

• Changes in disease frequency observed over years or decades.
• Reflect gradual shifts in agent, host, or environmental factors.
• Examples:
  • The dramatic decline in tuberculosis mortality in England and Wales over the 20th century (before BCG or antibiotics), attributed to improved nutrition and living conditions.
  • Rising incidence of coronary heart disease in developed nations during the 20th century, linked to dietary changes and sedentary lifestyle.
  • Declining polio incidence following introduction of vaccination.
  • Rising cancer incidence due to increased tobacco use (lung cancer secular trend in males in the 1950s-1980s).

• Regular recurrences over predictable time intervals.
• Caused by changes in herd immunity, climate cycles, or periodic introduction of new susceptibles.
• Two subtypes:
  • Seasonal variation: Repeating pattern every year based on season. Example: Influenza peaks in winter months; cholera peaks in summer/monsoon in India; measles peaks in late winter/spring.
  • Cyclic variation: Recurrence over multiple years. Example: Measles epidemics used to recur every 2-3 years (before vaccination) as new susceptibles accumulated. Influenza pandemics recur every 10-40 years. Meningococcal meningitis shows 8-12 year cycles in the African meningitis belt.

• Sudden, unpredictable changes in disease frequency.
• Include point-source epidemics (e.g., a food-borne outbreak), propagated epidemics (person-to-person spread), and mixed epidemics.
• Example: The 1854 Broad Street cholera outbreak (Snow's investigation) - a sudden sharp peak from a contaminated pump. COVID-19 pandemic (2019-present) represents a sudden global irregular fluctuation.

• Cases cluster sharply over a period equal to one incubation period.
• Reflects exposure to a single common source at one point in time.
• Example: Food poisoning at a wedding feast.

• Cases rise and fall gradually over multiple incubation periods.
• Reflects person-to-person spread.
• Example: Chickenpox in a school, hepatitis A in a community.

(b) Application of Transmission Dynamics in Prevention and Control (6 marks)

• Destroying the agent at source (disinfection, sterilization, pasteurization of milk).
• Reducing virulence through attenuated vaccines (live vaccines like MMR, OPV).
• Example: Chlorination of water eliminates V. cholerae, breaking the fecal-oral route.

• Eliminating animal reservoirs: culling infected animals (rabies in dogs, avian influenza in poultry).
• Treating human reservoirs: treating carriers and cases (e.g., treating typhoid carriers with antibiotics).
• Example: Mass treatment campaigns for malaria reduce the human reservoir.

• For droplet/airborne infections: ventilation, masks, UV air disinfection (TB, influenza).
• For fecal-oral route: safe water supply, sanitation, hand washing (cholera, typhoid, hepatitis A).
• For vector-borne diseases: insecticide-treated bed nets, larval control, adult mosquito control (malaria, dengue).
• For sexual transmission: condom use, partner notification (HIV, STIs).
• For blood-borne: blood screening, needle exchange programs (HIV, hepatitis B).

• Personal protective equipment in occupational settings (masks, gloves for healthcare workers).
• Skin barrier (protective clothing against ticks for Lyme disease).

• Immunization (most powerful tool): example - eradication of smallpox through ring vaccination exploiting the fact that the virus has no animal reservoir and humans are the only host.
• Chemoprophylaxis: malaria prophylaxis in travellers.
• Improving nutrition and general resistance.
• Herd immunity: when a critical proportion of the population is immune (herd immunity threshold = 1 - 1/R₀), transmission chains break. For measles (R₀ ~15), ~93-95% population immunity is needed.

• Based on knowledge of incubation period and communicable period.
• Example: 14-day quarantine for COVID-19 was derived from the maximum known incubation period.

(c) Indications of a Cohort Study (3 marks)

1. Good prior evidence of association - When clinical observations and preliminary case-control or descriptive studies already suggest an association between exposure and disease (cohort study is needed to confirm and quantify it).
2. Rare exposure, high incidence among exposed - When the exposure is uncommon in the general population but disease incidence is high among exposed groups (e.g., special occupational exposure groups like radiologists, asbestos workers, nuclear plant workers).
3. Attrition can be minimized - When the study population is stable, cooperative, accessible, and easy to follow up (e.g., doctors, civil servants, insured persons).
4. Ample funds available - Cohort studies are expensive and resource-intensive; they are feasible only when adequate funding and infrastructure exist.

• Park's Textbook of Preventive and Social Medicine, p. 87

Q4. Framingham Study - Type, Steps, Advantages/Disadvantages, Causal Criteria [KPC] (1+5+4+5=15)

(a) Type of Study (1 mark)

(b) Steps for Conducting this Study (5 marks)

1. Defining the study population and cohort: Residents of Framingham, Massachusetts were identified as the cohort (a geographically defined general population cohort). A random sample of town residents without heart disease was enrolled.
2. Baseline examination: All cohort members underwent comprehensive clinical examination at enrollment: ECG, blood pressure measurement, serum cholesterol, blood glucose, physical examination, and detailed history (smoking, diet, physical activity, family history).
3. Defining exposure variables: Multiple risk factors were recorded: hypertension, hypercholesterolaemia, smoking, diabetes, obesity, physical inactivity, and family history of heart disease.
4. Follow-up: Participants were re-examined at regular intervals (every 2 years in the Framingham study). Active surveillance for cardiac events was maintained through hospital records, physician notifications, and death certificates.
5. Outcome assessment: The incidence of coronary heart disease (angina, myocardial infarction, sudden cardiac death), stroke, and heart failure was recorded using standardized diagnostic criteria.
6. Analysis: Incidence rates in groups with and without each risk factor were compared. Relative risks, attributable risks, and multivariate risk models were developed. The Framingham Risk Score emerged from this analysis.

(c) Advantages and Disadvantages (4 marks: 2+2)

1. Direct measurement of incidence: Provides true incidence rates and Relative Risk (stronger measure of association than Odds Ratio from case-control studies).
2. Temporal relationship established: Since exposure is measured before disease onset, the sequence cause → effect is clearly established, strengthening causal inference.
3. Multiple outcomes studied simultaneously: From one cohort, multiple outcomes (MI, stroke, heart failure) can be examined against multiple exposures.
4. Less subject to recall bias: Exposure data is collected prospectively, not relying on retrospective recall.
5. Identifies new risk factors and generates hypotheses for future research.

1. Expensive and time-consuming: Decades of follow-up with repeated examinations require enormous resources.
2. Loss to follow-up: Over decades, participants migrate, withdraw consent, or die from unrelated causes, potentially biasing results.
3. Not suitable for rare diseases: A large cohort and long follow-up are needed to observe sufficient outcomes.
4. Healthy cohort/survivor bias: Those who remain in the study for decades may represent healthier individuals.
5. Changes in diagnostic criteria over time can create inconsistencies.

(d) Criteria for Judging Causal Association - Hill's Criteria (5 marks)

1. Strength of association: A strong association (high RR or OR) is more likely to be causal. Example: RR of 9-10 for lung cancer in heavy smokers.
2. Consistency: The association has been repeatedly observed by different researchers, in different populations, at different times and places.
3. Specificity: One cause leads to one specific effect (one exposure → one disease). Less absolute criterion today as many diseases are multifactorial.
4. Temporality (Correct time sequence): The cause must precede the effect. Exposure must occur before disease onset. This is the only essential criterion.
5. Biological gradient (Dose-response relationship): Greater exposure leads to greater disease frequency. Example: more cigarettes smoked → higher lung cancer incidence.
6. Plausibility: The association should be biologically plausible - consistent with known biological mechanisms and existing knowledge.
7. Coherence: The association should not conflict with what is generally known about the natural history and biology of the disease.
8. Experiment (Experimental evidence): When the cause is removed (preventive trial) or disease frequency falls, or when the exposure is introduced and disease increases, this provides strong support. Example: Stopping smoking reduces lung cancer risk.
9. Analogy: Similar known causal relationships exist that support the proposed causal mechanism (e.g., if thalidomide causes birth defects, perhaps another drug in pregnancy might also).

Q5. Outbreak of fever with rash and bronchopneumonia in under-five children, border district West Bengal [SRIMS] (2+6+7=15)

(i) Probable Diagnosis (2 marks)

• Under-five age group - highest susceptibility
• Fever with rash (maculopapular rash spreading centrifugally)
• Bronchopneumonia - the most common fatal complication of measles
• Border district setting suggests possible importation from across the border (cross-border movement of unvaccinated children)
• Deaths reported - consistent with complicated measles in malnourished/unvaccinated children

(ii) Outbreak Investigation (6 marks)

1. Immediate notification and team deployment:
   • Notify the District Health Officer, State Surveillance Unit, Integrated Disease Surveillance Programme (IDSP).
   • Deploy a Rapid Response Team (RRT) to the field.
2. Case verification and diagnosis:
   • Clinically confirm measles cases: fever ≥38.3°C, characteristic maculopapular rash lasting ≥3 days, cough/coryza/conjunctivitis (3 C's) and/or Koplik's spots.
   • Collect nasopharyngeal swabs, throat swabs, blood for IgM antibody (ELISA), urine for virus isolation.
   • Confirm with laboratory testing.
3. Case finding and defining:
   • Establish a case definition: "Any child under 5 years in [district] presenting with fever and rash from [date], with or without respiratory symptoms."
   • Active case search: visit all health facilities, anganwadis, schools, and villages in the affected area.
   • House-to-house survey.
4. Epidemiological characterization (Person, Place, Time):
   • Plot epidemic curve: determine onset, peak, duration. Is it a propagated epidemic (person-to-person spread - expected for measles)?
   • Map geographic distribution of cases.
   • Age-specific attack rates: which age groups are most affected?
   • Vaccination status of all cases: proportion unvaccinated/incompletely vaccinated.
5. Identify risk factors:
   • Cross-border movement, crowding, malnutrition, school/anganwadi attendance.
   • Vaccination coverage survey in the affected area using lot quality assurance sampling (LQAS).
6. Case management:
   • Identify severe cases needing hospitalization (pneumonia, dehydration, encephalitis).
   • Ensure Vitamin A supplementation for all confirmed measles cases (reduces morbidity and mortality).
   • Establish isolation facilities.

(iii) Containment of the Outbreak (7 marks)

1. Emergency immunization (Ring vaccination / Mass vaccination):
   • Conduct outbreak response immunization (ORI): administer measles vaccine (MR/MMR) to all children 6 months to 5 years (expand to 15 years if needed) in the affected area and surrounding villages, regardless of prior vaccination status.
   • Extend vaccination to border area villages on both sides of the border (coordinate with counterpart health authorities if possible).
2. Case management:
   • Hospitalize severe cases (pneumonia, encephalitis, severe dehydration).
   • Vitamin A supplementation for all confirmed cases (200,000 IU for children >1 year; 100,000 IU for 6-12 months).
   • Supportive care: antipyretics, antibiotics for secondary bacterial pneumonia.
3. Isolation of cases:
   • Isolate confirmed measles cases for 4 days after rash onset (infectious period: 4 days before to 4 days after rash onset).
   • Restrict movement of ill children from school/anganwadi.
4. Surveillance intensification:
   • Strengthen fever-rash surveillance through IDSP.
   • Weekly case reporting from all health facilities, anganwadis, and community health workers (ASHAs/AWWs).
   • Monitor for complications and deaths.
5. Nutritional support:
   • Screen for malnutrition; provide therapeutic feeding for SAM/MAM children (malnutrition increases measles fatality significantly).
   • Vitamin A distribution (also for contacts as prophylaxis in some situations).
6. Health education and community mobilization:
   • Communicate risk to community through local leaders, ASHAs, panchayat.
   • Emphasize importance of vaccination, hand hygiene, isolating sick children.
   • Address vaccine hesitancy if identified.
7. Cross-border coordination:
   • Notify border district health officials (in liaison with state and national level) to conduct parallel vaccination campaigns and surveillance on the other side of the border.
8. Review and feedback:
   • Prepare an outbreak investigation report.
   • Identify gaps in routine immunization coverage; plan remedial action.
   • Strengthen cold chain and vaccine supply in the district.

Q6. Epidemiology vs. Clinical Medicine; Point-source epidemic; Periodic fluctuations; Geographic distribution; Function of descriptive epidemiology [SSKM] (2+3+4+5+1=15)

(a) Two Distinct Differences: Epidemiology vs. Clinical Medicine (2 marks)

1. Unit of study: Clinical medicine focuses on the individual patient; epidemiology focuses on population groups and compares rates of disease between groups.
2. Purpose: Clinical medicine aims at diagnosis and treatment; epidemiology aims at identifying causes and risk factors to enable prevention and control at the population level.

(b) Primary Epidemiological Characteristics of a Point-Source Epidemic (3 marks)

1. Explosive onset: A sudden, rapid rise in the number of cases over a very short period.
2. Epidemic curve shape: The histogram of cases plotted against time shows a sharp, single peak (bell-shaped or skewed) with most cases clustered within one incubation period of the exposure.
3. Cases confined to exposed period: The time from exposure to the peak of cases corresponds to the typical (mean/median) incubation period of the causative agent.
4. Rapid decline: Cases fall as steeply as they rise, since no further exposure occurs after the point source is eliminated.
5. Common exposure: All or most cases share a definite, identifiable exposure (e.g., a contaminated food item at a feast, a contaminated water source).
6. Limited geographic distribution: Cases are concentrated among those who attended the common event or used the common source.

• Example: Food poisoning outbreak at a wedding feast; cholera from a single contaminated well (Broad Street pump, 1854).

(c) Periodic Fluctuations of Epidemics with Real-World Examples (4 marks)

• Disease incidence regularly rises and falls within each calendar year, peaking in a particular season.
• Caused by seasonal changes in: survival and multiplication of pathogens, vector activity, host behavior and crowding, rainfall, temperature, and humidity.

• Influenza: Peaks in winter months in temperate regions (November-February in the Northern Hemisphere) because cold weather promotes crowding, low humidity favors aerosol survival, and vitamin D levels are lower.
• Cholera in India: Peaks twice a year - just before the monsoon (April-June, when water sources concentrate) and during/after monsoon (August-September, with flooding and water contamination).
• Malaria: Peaks 1-2 months after the monsoon season (September-November) when Anopheles mosquito breeding is maximal.
• Measles in pre-vaccine era: Peaked in late winter and spring in crowded urban settings.
• Rotavirus diarrhoea: Peaks in winter months in India.

• Disease incidence fluctuates over a cycle of several years, driven primarily by the accumulation of susceptibles after each epidemic depletes herd immunity.
• As immunity wanes in the community (or new susceptibles are born), the susceptible pool builds up until another epidemic becomes possible.

• Measles (pre-vaccination era): Epidemics recurred every 2-3 years in urban areas as new birth cohorts reached school age unimmunized.
• Influenza pandemics: Occur every 10-40 years (1918, 1957, 1968, 2009), driven by antigenic shift producing a new subtype against which the population has no immunity.
• Meningococcal meningitis (Sub-Saharan Africa - "Meningitis Belt"): Epidemics recur approximately every 8-12 years, linked to the accumulation of susceptibles and specific climate conditions (dry harmattan winds).
• Whooping cough (pertussis): In the pre-vaccine era, cyclic peaks every 3-4 years.

(d) How Geographic Distribution Affects Descriptive Epidemiology Findings (5 marks)

• Striking differences in disease rates between countries suggest the importance of environmental, dietary, genetic, or socioeconomic factors.
• Example: Stomach cancer is very common in Japan but rare in the USA. Japanese migrants to the USA who adopt American diets show reduced rates, pointing to dietary/environmental rather than genetic causes.
• Coronary heart disease rates are higher in Finland/UK than in Japan, pointing to dietary fat differences.

• Within a country, geographic clustering of disease indicates local environmental factors or behavioral patterns.
• Example: Fluorosis is concentrated in states of India (Rajasthan, Gujarat, Andhra Pradesh) with high natural fluoride in groundwater.
• Goitre belt corresponds to iodine-deficient soil in mountainous regions (Himalayan foothills, Western Ghats).

• Urban: higher rates of cardiovascular disease, lung cancer, STIs, mental illness, trauma.
• Rural: higher rates of vector-borne diseases, diarrhoeal diseases, malnutrition, respiratory infections.
• This reflects differences in lifestyle, pollution, access to healthcare, and environmental exposures.

• Geographic clustering of cases within a community points to a localized source.
• John Snow's spot map of 1854 Broad Street cholera outbreak showed cases clustering around the contaminated pump - a landmark application of geographic epidemiology.

• Migrants who adopt the disease rates of their new country (rather than their country of origin) suggest environmental (not genetic) causes.
• These findings fundamentally shape descriptive epidemiology conclusions - a disease assumed to be genetic may actually be environmental once geographic data and migrant data are combined.

(e) One Major Function of Descriptive Epidemiology (1 mark)

Q7. Urban air pollution and bronchial asthma in school children [SCCGMCH] (3+6+3+3=15)

(a) Most Appropriate Study Design and Justification (3 marks)

1. The hypothesis involves long-term exposure (urban air pollution) preceding an outcome (development of bronchial asthma) - a temporal, prospective design is required.
2. The exposure (urban air pollution) can be objectively measured at baseline and quantified longitudinally (air quality monitoring, personal samplers).
3. Among school-going children (a defined, accessible cohort), those in high-pollution urban areas (exposed cohort) can be compared to those in low-pollution areas (unexposed cohort).
4. Cohort study allows direct calculation of incidence rates and Relative Risk - the most valid measure of association.
5. Since asthma in children is not extremely rare, a cohort study is feasible; sufficient incident cases will be observed over time.
6. The study allows investigation of dose-response relationship (degree of air pollution exposure correlated with asthma incidence).

(b) Step-by-Step Methodology (6 marks)

• Select school-going children (e.g., 5-15 years) from two areas: a high-pollution urban area and a low-pollution area (suburban/rural).
• Alternatively, within one city, select children living near busy roads/industrial zones (high exposure) vs. children living in greener, low-traffic areas (low exposure).

• Screen all children at enrollment:
  • Medical history: any pre-existing asthma, allergies, atopic disease, family history.
  • Physical examination.
  • Spirometry (FEV₁, FVC, FEV₁/FVC ratio).
  • Skin prick tests for common allergens.
  • Exclude children with pre-existing asthma (or enroll them separately to study exacerbations).

• Measure air quality at school and home locations:
  • Ambient PM₂.₅, PM₁₀, NO₂, SO₂, ozone levels using fixed monitoring stations.
  • Personal air quality monitors worn by children for sub-samples.
  • Geographic Information System (GIS) mapping of proximity to roads, industries.
• Assign each child an exposure level (low, medium, high) based on residential and school location air quality data.
• Record time spent outdoors, commuting patterns (potential confounders).

• Follow all children prospectively for a defined period (minimum 3-5 years).
• Annual re-examination: spirometry, questionnaire on respiratory symptoms, hospitalizations.
• Record incident asthma cases using standardized criteria (GINA criteria: recurrent wheeze, breathlessness, chest tightness, and cough with variable expiratory airflow limitation).
• Track loss to follow-up (school transfers, migration, withdrawal).

• Collect data on potential confounders at baseline and follow-up:
  • Socioeconomic status, parental smoking, indoor air quality (biomass fuel, passive smoking), family history of atopy, obesity, physical activity.
  • School environment (ventilation).

• Calculate incidence rate of asthma in high-exposure vs. low-exposure groups.
• Compute Relative Risk (RR) and 95% confidence intervals.
• Assess dose-response relationship: does higher PM₂.₅/NO₂ exposure correlate with higher asthma incidence?
• Multivariate Cox proportional hazards regression to adjust for confounders.
• Sensitivity analyses for missing data and lost-to-follow-up subjects.

• Obtain ethical committee approval.
• Written informed consent from parents/guardians, assent from older children.
• Ensure no harm from monitoring procedures; refer cases identified during study for treatment.

(c) Strengths of this Study Design in Establishing Temporal and Causal Relationship (3 marks)

1. Temporal sequence established: The exposure (air pollution level) is measured before the outcome (new asthma cases) appears. This is the most critical criterion for causality (Hill's temporality criterion). Case-control studies cannot establish this as clearly.
2. Direct incidence measurement: Cohort studies measure the incidence (new cases) of asthma directly in exposed vs. unexposed groups, yielding a true Relative Risk - the most direct measure of causal strength.
3. Dose-response relationship: By quantifying pollution exposure at multiple levels, a biological gradient (more pollution → more asthma) can be demonstrated, further supporting causality.
4. Multiple outcomes: The same cohort can be studied for multiple respiratory outcomes (rhinitis, COPD precursors, lung function decline), adding efficiency.
5. Minimized recall bias: Since exposure is measured prospectively with objective monitoring equipment, it does not rely on subjects' memories.

(d) Potential Sources of Bias (3 marks)

1. Selection bias:
   • If the two comparison groups differ systematically at baseline (e.g., children in high-pollution areas may also be poorer, with higher indoor pollution exposure), the effect of outdoor pollution is confounded.
   • Volunteers may be healthier than non-participants (healthy volunteer effect).
2. Attrition bias (Loss to follow-up bias):
   • Children who develop respiratory problems may move away from the polluted city (migration bias), causing underestimation of the true effect.
   • Children who are sicker may drop out, or healthier children may be more likely to complete follow-up.
3. Information bias:
   • Exposure misclassification: Using fixed monitoring stations to estimate individual exposure ignores time spent indoors, school vs. home environments, and commuting. Personal monitor data is more accurate but expensive.
   • Outcome misclassification: Asthma diagnosis may be inconsistent; wheeze questionnaires have variable sensitivity and specificity; parents may over- or under-report symptoms.
4. Confounding:
   • Indoor air pollution (cooking fuel, passive smoking), socioeconomic status, dietary patterns, obesity, and genetic susceptibility (atopic family history) are major confounders that can bias the association between outdoor air pollution and asthma.
5. Hawthorne effect:
   • Awareness of being in a study may lead families to change behaviors (e.g., reduce outdoor activity in polluted areas), reducing true exposure and underestimating the association.

• Park K. Park's Textbook of Preventive and Social Medicine, 26th edition. Jabalpur: Banarsidas Bhanot; 2023. pp. 62-100 (Epidemiological studies), pp. 105-140 (Epidemiology of communicable diseases).
• Hill AB. The Environment and Disease: Association or Causation? Proc R Soc Med. 1965;58:295-300.