Communicable disease epidemiology and outbreak investigation

In short

Communicable disease epidemiology studies how infectious agents spread through populations and how that spread can be interrupted; outbreak investigation is its field application — a standard sequence of steps used to detect, confirm, describe and stop an epidemic. India carries a very large infectious disease burden (tuberculosis, dengue, malaria, diarrhoeal disease, hepatitis A/E, measles, leptospirosis, scrub typhus, rabies) and runs formal surveillance through the Integrated Disease Surveillance Programme (IDSP) on the Integrated Health Information Platform (IHIP), the National Centre for Disease Control (NCDC) and the India-EIS field epidemiology training programme. The topic is examined in BPH/B.Sc Public Health epidemiology papers, in the Epidemiology paper of MPH and AIIMS/ICMR entrance tests, and tested in technical rounds for Surveillance Officer, Epidemiologist (NHM), Monitoring & Evaluation Officer and research roles at ICMR institutes, WHO-NPSP, PATH, Jhpiego and clinical research organisations. Interviews focus on the ten investigation steps, epidemic curve reading, attack-rate tables and the difference between a point-source and propagated outbreak.

Explained

Concepts, agent-host-environment and the chain of infection

• Communicable (infectious) disease: illness caused by transmission of a specific infectious agent or its toxic products from an infected person, animal or reservoir to a susceptible host, directly or indirectly.
• Epidemiological triad: agent (infectivity, pathogenicity, virulence, immunogenicity), host (age, sex, nutrition, immunity, behaviour) and environment (water, sanitation, housing, vectors, climate); disease occurs when the balance shifts.
• Chain of infection — six links: (1) infectious agent, (2) reservoir, (3) portal of exit, (4) mode of transmission, (5) portal of entry, (6) susceptible host. Control means breaking the weakest link.
• Reservoir types: human (cases, carriers — TB, measles), animal/zoonotic (rabies in dogs, plague in rodents, brucellosis in cattle, Nipah in fruit bats), and non-living/environmental (soil for tetanus and botulism, water for legionella).
• Carrier states: incubatory (measles before rash), convalescent (typhoid after recovery), healthy/asymptomatic (polio, diphtheria), and chronic (hepatitis B, "Typhoid Mary"). Carriers matter more than cases in typhoid and hepatitis B control.
• Modes of transmission: *Direct* — contact, droplet (large droplets, under ~1 m), transplacental/vertical, inoculation, bites. *Indirect* — vehicle-borne (water, food, blood, fomites), vector-borne (biological: Anopheles-malaria, Aedes-dengue; mechanical: housefly), airborne (droplet nuclei under 5 µm — TB, measles, varicella), and fomite-borne.
• Key time intervals: incubation period (infection → first symptom), latent period (infection → start of infectiousness), infectious/communicable period, serial interval (symptom onset in case → onset in its secondary case; the generation time is the corresponding infection → infection interval), pre-patent period (infection → detectable parasite).
• Terms of occurrence: *sporadic* (scattered cases), *endemic* (constant expected presence — malaria in Odisha), *hyperendemic* (intense, all ages), *epidemic/outbreak* (occurrence clearly in excess of expectation), *pandemic* (multi-country), *exotic* (imported), *epizootic* (animal epidemic), *enzootic*.
• Spectrum / iceberg of disease: sub-clinical and carrier states far outnumber clinical cases; surveillance based only on hospital cases sees the tip of the iceberg.

Herd immunity, reproduction number and dynamics

• Basic reproduction number R₀: average number of secondary cases one case generates in a wholly susceptible population — measles 12-18, pertussis 12-17, diphtheria 6-7, polio 5-7, influenza 1.5-2, SARS-CoV-2 ancestral strain roughly 2-3 (variant-dependent).
• Effective reproduction number R (Rt): R₀ × proportion susceptible. R > 1 → epidemic grows; R = 1 → endemic plateau; R < 1 → epidemic dies out. All control measures aim to push R below 1.
• Herd immunity (community immunity): when a high enough proportion is immune, transmission chains break and even susceptibles are protected. Herd immunity threshold Hc = 1 − 1/R₀.
• Herd immunity works only for diseases with human-only reservoirs and person-to-person transmission; it does not protect against tetanus (soil reservoir) or rabies (animal reservoir).
• Epidemic ends when susceptibles fall below threshold, not when everyone is infected; waning immunity, births and migration replenish susceptibles and cause cyclical epidemics (measles inter-epidemic interval).

Surveillance in India

• Surveillance: continuous, systematic collection, analysis and interpretation of health data, with timely dissemination for action — "information for action". Not the same as research or a one-off survey.
• Types: *passive* (routine reporting from facilities — cheap, incomplete), *active* (investigator seeks cases — complete, costly), *sentinel* (selected high-quality sites — ICMR virology network for dengue/influenza), *syndromic* (symptom clusters, used for early warning and in disasters), and *event-based* (media, rumour registers).
• IDSP / IHIP: reporting is through three forms — S (syndromic, by health workers/ANMs), P (presumptive, by medical officers) and L (laboratory-confirmed). District and State Surveillance Units feed the Central Surveillance Unit at NCDC; the IHIP portal moved reporting to case-based, near-real-time entry. Weekly outbreak reports are published by NCDC.
• Key supporting systems: National Centre for Disease Control (NCDC), Delhi; ICMR-NIV Pune for virology; National Vector Borne Disease Control Programme (NVBDCP, now under NCVBDC); NIKSHAY for TB notification (TB is a notifiable disease nationwide); National Rabies Control Programme; India-EIS (Epidemic Intelligence Service) two-year field epidemiology training; National One Health Mission for zoonoses.
• Notifiable diseases: notification is governed by state public health Acts and lists, so the list varies by state; nationally notifiable items include TB (since 2012/2018 orders), and AFP, measles, neonatal tetanus and others under programme requirements. Because lists differ, name the state Act rather than assume a uniform national list.
• Legal framework: the Epidemic Diseases Act, 1897 (as amended in 2020 to protect healthcare workers) and the Disaster Management Act, 2005 are the instruments used during epidemics; many states have their own Public Health Acts, and a model national public health Bill has been under consideration for years without being enacted — status should be verified before quoting.
• International: WHO International Health Regulations (IHR 2005) oblige India to maintain core capacities and notify events of potential international concern through a National IHR Focal Point (NCDC); a Public Health Emergency of International Concern (PHEIC) is declared by the WHO Director-General.

Steps of an outbreak investigation

1. Establish the existence of an outbreak: compare current counts with the expected baseline for that place and season (previous 3-5 years, endemic channel). Rule out pseudo-epidemics from a new test, a new doctor, better reporting or a change in case definition.
2. Verify the diagnosis: review clinical records, examine some patients, confirm with laboratory tests; collect specimens (stool, blood, serum, CSF, water) before treatment where possible.
3. Define and count cases: write a case definition with clinical criteria plus restrictions by time, place and person; classify as suspected / probable / confirmed. Keep it sensitive early (find all cases) and tighten it for analysis. Prepare a line list (name, age, sex, address, date of onset, symptoms, exposure, outcome).
4. Describe the data by time, place and person (descriptive epidemiology): epidemic curve; spot map/GIS; attack rates by age, sex, occupation and ward. Identify the index case (first case that comes to notice), primary case (brings infection in) and secondary cases.
5. Develop a hypothesis about the agent, source, vehicle, mode of transmission and population at risk, consistent with the curve, the incubation period and the descriptive picture.
6. Test the hypothesis analytically: a retrospective cohort study when the exposed population is enumerable (a wedding, hostel, school — compute food-specific attack rates and relative risk); a case-control study in an open community (compute odds ratios).
7. Refine hypotheses and do additional studies: environmental and laboratory investigation — water sampling and residual chlorine, food handler stool culture, entomological survey (house index, container index, Breteau index for Aedes), veterinary investigation for zoonoses.
8. Implement control and prevention measures: act in parallel from day one, do not wait for analysis — chlorinate or stop the implicated water source, withdraw implicated food, isolate cases, vaccinate, treat contacts, vector control (fogging for adults, source reduction for larvae).
9. Evaluate and continue surveillance: check that incidence falls, continue active case search for at least two incubation periods after the last case before declaring the outbreak over.
10. Communicate findings: oral briefing to administration for immediate action, and a written report (introduction, methods, results, discussion, recommendations) that becomes the legal and scientific record; risk communication to the community and press.

Epidemic curves and types of outbreak

• Epidemic curve: histogram of number of cases (y-axis) against date/time of onset (x-axis); the x-interval should be about one-quarter to one-third of the incubation period.
• Point source (common source, single exposure): sharp rise, single peak, rapid fall; all cases within one incubation period — food poisoning at a marriage feast. The curve's spread estimates the incubation range, and the peak minus one average incubation period estimates the exposure time.
• Continuous common source: plateau over many incubation periods — a persistently contaminated piped water supply causing cholera or hepatitis E.
• Intermittent common source: irregular peaks following irregular exposures.
• Propagated (progressive, person-to-person): successive peaks one serial interval apart, each taller than the last until susceptibles are exhausted — measles, diphtheria, COVID-19.
• Mixed: a point-source outbreak followed by secondary person-to-person spread (hepatitis A, shigellosis).
• Reading the curve gives the probable mode of spread, the exposure period, outliers (early case = source or separate exposure; late case = secondary case or long incubation) and the stage of the epidemic.
• Spot map shows clustering around a pump, canteen, hostel block or borewell (John Snow's Broad Street investigation, 1854, remains the classic).

Levels of prevention and control measures

• Control of the reservoir: early diagnosis and notification, isolation, treatment, quarantine of contacts, disinfection; for animal reservoirs — mass dog vaccination, culling of infected poultry, control of rodents.
• Isolation vs quarantine: isolation separates known cases for the duration of communicability; quarantine restricts movement of apparently healthy exposed contacts for the maximum incubation period.
• Interrupt transmission: safe water and chlorination, sanitation and safe excreta disposal, food safety (FSSAI standards under the Food Safety and Standards Act, 2006), hand hygiene, vector control, safe blood and injection practices, air disinfection and ventilation.
• Protect the host: active immunisation (Universal Immunisation Programme vaccines — BCG, hepatitis B birth dose, OPV, fIPV, pentavalent, rotavirus, PCV, MR, DPT boosters, Td, JE in endemic districts; vaccination is free and voluntary, not legally compulsory), passive immunisation (immunoglobulin for rabies, hepatitis B, tetanus), chemoprophylaxis (rifampicin for meningococcal contacts, doxycycline in leptospirosis/scrub typhus settings per protocol), and non-specific measures — nutrition, personal hygiene, PPE.
• Cold chain: vaccines are carried at +2 to +8 °C in ice-lined refrigerators and vaccine carriers; OPV and some others are kept frozen at deep-freeze temperatures in stores; the vaccine vial monitor (VVM) and shake test indicate heat and freeze damage. eVIN tracks vaccine stock and temperature digitally.
• Levels of prevention: primordial, primary (immunisation, sanitation), secondary (early diagnosis and treatment, DOTS/NTEP, screening), tertiary (disability limitation and rehabilitation).
• Disease elimination vs eradication: *eradication* = permanent worldwide reduction to zero with no further intervention needed (smallpox 1980; rinderpest in animals); *elimination* = reduction to zero or to a defined low level in a defined geographical area (India certified polio-free in March 2014; yaws and maternal/neonatal tetanus eliminated; measles-rubella elimination and TB reduction targets are active programme goals — the target dates have been revised more than once, so quote the current NTEP/NHM document rather than a remembered year).

Key points

• Communicable disease - disease transmitted from a reservoir to a susceptible host by a specific agent.
• Chain of infection - agent → reservoir → exit → transmission → entry → susceptible host; control breaks a link.
• Incubation period - infection to first symptom; it sets quarantine length and dates the exposure from an epidemic curve.
• Latent period - infection to becoming infectious; may start before symptoms (pre-symptomatic spread in COVID-19).
• Carrier - infected person shedding the agent without clinical disease; incubatory, convalescent, healthy or chronic.
• R₀ - average secondary cases per case in a fully susceptible population; R > 1 means a growing epidemic.
• Herd immunity threshold - Hc = 1 − 1/R₀; about 92-95% immunity needed for measles.
• Attack rate - cases ÷ population at risk during an outbreak; a cumulative incidence, expressed as a percentage.
• Secondary attack rate - new cases among household contacts ÷ (contacts − primary cases); measures infectiousness.
• Case definition - standard criteria of clinical features plus time, place and person; sensitive early, specific later.
• Epidemic curve - histogram of cases by onset date; shape reveals point-source, continuous, intermittent or propagated spread.
• Point-source curve - one sharp peak, all cases within one incubation period; peak minus mean incubation ≈ exposure time.
• Propagated curve - progressively taller peaks one serial interval apart.
• Index case - first case to come to the investigator's notice; not necessarily the primary case.
• Isolation vs quarantine - cases separated for the infectious period; healthy contacts restricted for the maximum incubation period.
• IDSP on IHIP - S, P and L forms; district and state surveillance units reporting to NCDC's Central Surveillance Unit.
• Case fatality rate - deaths among cases; measures severity, unlike the attack rate which measures spread.
• Control before conclusion - control measures start immediately, in parallel with the analytic study, never after it.

Formulas and examples

Example 1 — Food-specific attack rates at a wedding (retrospective cohort). 250 guests attended a wedding in Pune; 80 developed vomiting and diarrhoea within 2-6 hours.

• Overall attack rate = 80/250 × 100 = 32%.
• Food histories for the suspected items:

• Paneer curry: RR = 60.0/6.2 ≈ 9.7 — strongly implicated (high AR among eaters, low AR among non-eaters).
• Rice RR < 1 and salad RR = 1 — not implicated.
• Short 2-6 hour incubation with vomiting points to a preformed toxin, most consistent with *Staphylococcus aureus* food poisoning from a milk product handled by a carrier.

Example 2 — Case-control study of a jaundice outbreak. 60 hepatitis E cases and 120 neighbourhood controls were asked about drinking untreated ward-tap water: 48 cases and 48 controls were exposed.

• a = 48, b = 48, c = 60 − 48 = 12, d = 120 − 48 = 72.
• OR = (48 × 72)/(48 × 12) = 3456/576 = 6.0.
• Exposed persons had six times the odds of illness; with a plateau-shaped curve this indicates a continuous common-source piped-water outbreak — check residual chlorine and sewage cross-connection.

Example 3 — Secondary attack rate and herd immunity for measles. A measles case in a village had 25 household contacts; 18 were unvaccinated susceptibles, and 12 of them developed measles within 7-18 days.

• Secondary attack rate among susceptible contacts = 12/18 × 100 = 66.7% (using susceptible contacts as the denominator; if all 25 contacts are used, 12/25 = 48%, which understates infectiousness because immune contacts are not at risk).
• If R₀ = 15, Hc = 1 − 1/15 = 0.933 → 93.3% immunity needed.
• With a vaccine efficacy of 85% for a single dose, coverage needed = 0.933/0.85 = 1.098 → over 100%, i.e. unachievable with one dose — which is exactly why India gives two MR doses.

Example 4 — Epidemic curve and exposure dating. A hostel outbreak of hepatitis A: cases ran from 2 to 30 March, peaking on 16 March, with the mean incubation period of hepatitis A taken as 28 days (range 15-50).

• Estimated exposure = 16 March − 28 days = 16 February.
• Cross-check with the range: first case 2 March − 50 days = 11 January; last case 30 March − 15 days = 15 March; so exposure lies in the overlap, around mid-February.
• All cases falling within one incubation span supports a point source, probably a single contaminated water-tank filling, and the investigation should target mid-February tank cleaning and supply records.

Example 5 — CFR and comparison of measures. In a district outbreak, 4,000 of 50,000 residents fell ill with cholera and 40 died.

• Attack rate = 4,000/50,000 × 100 = 8%; CFR = 40/4,000 × 100 = 1%; mortality rate = 40/50,000 × 100 = 0.08%.

Where it is used

• National Health Mission and district health offices: District Epidemiologist, District Surveillance Officer and Public Health Specialist posts — weekly IDSP/IHIP analysis, outbreak response team (RRT) deployment and outbreak reports.
• NCDC, ICMR institutes (NIV Pune, NIE Chennai, NIRT, NICED) and India-EIS: field investigations of Nipah in Kerala, Zika in Rajasthan/Maharashtra, scrub typhus in the sub-Himalayan belt, acute encephalitis syndrome in Bihar and Uttar Pradesh.
• Programme management: NTEP (TB notification via NIKSHAY and contact investigation), NVBDCP/NCVBDC (dengue, malaria, kala-azar, lymphatic filariasis MDA), National Rabies Control Programme, UIP and intensified Mission Indradhanush microplanning.
• Hospital infection control and antimicrobial stewardship: infection control nurse/officer roles — surveillance of hospital-acquired infections, outbreak investigation of ICU clusters, NABH accreditation requirements and ICMR's AMR surveillance network.
• International agencies and NGOs: WHO-NPSP surveillance medical officers, UNICEF, PATH, Jhpiego, Gates Foundation projects — measles-rubella and polio surveillance, AEFI investigation.
• Food, water and industry: FSSAI food-safety investigations of foodborne outbreaks, municipal water-quality surveillance (drinking water quality per the Indian Standard for drinking water, IS 10500), and occupational health teams investigating clusters in factories and hostels.

Interview questions

• Q: What is the difference between an outbreak and an epidemic? A: Technically the same — occurrence in excess of expectation — but "outbreak" is used for a localised, limited event and "epidemic" for a wider one; "outbreak" is preferred in public communication as less alarming.
• Q: How do you decide an outbreak exists? A: Compare the current number of cases with the expected number for that place and season from the past 3-5 years or an endemic channel, after ruling out artefacts such as a new diagnostic test, a changed case definition or improved reporting.
• Q: Index case or primary case — which is which? A: The primary case brings the infection into the population; the index case is simply the first one that came to the investigator's attention, and the two are often different people.
• Q: Why keep the case definition sensitive at the start? A: To avoid missing cases while the picture is unclear; a specific definition is applied later, for the analytic study, to reduce misclassification.
• Q: From an epidemic curve with a single sharp peak, what do you conclude and how do you date the exposure? A: A point-source outbreak; subtract one average incubation period from the peak date, and cross-check using the first and last cases with the minimum and maximum incubation periods.
• Q: Cohort or case-control for a wedding-party outbreak? A: Retrospective cohort, because the exposed population (the guest list) is fully enumerable, so food-specific attack rates and relative risks can be computed directly.
• Q: What does a secondary attack rate tell you that an attack rate does not? A: It measures the agent's infectiousness from person to person among exposed susceptible contacts, and is used to judge whether spread is propagated.
• Q: Why does herd immunity not protect against tetanus? A: Tetanus has an environmental (soil) reservoir and no person-to-person transmission, so immunising others does not reduce your exposure.
• Q: Difference between elimination and eradication, with Indian examples? A: Elimination is reduction to zero or a defined low level in a defined area (India certified polio-free in 2014); eradication is permanent global reduction to zero so interventions can stop (smallpox, 1980).
• Q: Would you wait for laboratory confirmation before acting? A: No — control measures run in parallel from the first day on the strength of the clinical and epidemiological picture; waiting for the laboratory costs cases.
• Q: How does India report an outbreak officially? A: Through IDSP S/P/L reporting on the IHIP portal from peripheral health workers and facilities up to the District and State Surveillance Units and NCDC's Central Surveillance Unit, with rapid response team investigation and, if internationally relevant, notification under IHR 2005.
• Q: When can an outbreak be declared over? A: When no new cases occur for at least two maximum incubation periods after the last case, with active case search continuing through that period.

Common mistakes

• Using the whole population as the denominator for an attack rate instead of only the population at risk — or including already-immune people among susceptible contacts in a secondary attack rate.
• Confusing attack rate (spread) with case fatality rate (severity), and reporting one as the other.
• Treating the index case as the source of infection.
• Plotting the epidemic curve by date of reporting or date of diagnosis instead of date of symptom onset, and choosing an x-axis interval far too wide, which flattens a propagated curve into a single peak.
• Confusing isolation with quarantine, and incubation period with latent or infectious period.
• Calculating an odds ratio in a cohort study or a relative risk in a case-control study, where the denominators of exposure are not known.
• Deferring control measures until the analytic study and laboratory results are complete.

Related topics

Epidemiology: principles, measures and study designs · Biostatistics for health sciences · Immunisation and vaccine-preventable diseases · Environmental health: water, sanitation and vector control · Health programmes and policy in India · One Health and zoonotic disease control