Communicable Disease Control and Outbreak Investigation
Community Medicine — Communicable Disease Control, NMC MBBS licence examination syllabus (Nepal Medical Council).
Communicable disease control: finding the source and breaking the chain
Outbreak investigation is a method, and the method is the examinable part.
An outbreak is one of the few situations in medicine with a genuine standard procedure. There is an agreed sequence of steps, each one producing the information the next one needs, and the discipline of following it in order is what separates a solved outbreak from a rumour. That makes this topic unusually reliable to revise — you are learning a method rather than a set of facts.
Two ideas do most of the work. The chain of infection tells you where control measures can be applied, and the epidemic curve tells you what kind of source you are dealing with. Between them they answer most of what an examiner can ask.
🩺 Where this lives: The most important rule in outbreak response is that control does not wait for the investigation to finish. If contaminated water is the obvious source, you act on the water while the analysis continues — you do not withhold intervention pending a case-control study. Investigation and control run in parallel, and treating them as sequential costs cases. This is also why notification happens on suspicion rather than on laboratory confirmation: the days spent waiting for a result are exactly the days when action would have helped most.
💡 A note on lists and numbers. This chapter gives no list of Nepal's notifiable diseases and no incubation periods for named diseases. The notifiable list is set by national policy and revised over time, and published incubation periods are ranges that differ between sources. Both must come from current official guidance. What is taught here is how they are used — which is the part that does not change. For screening, sensitivity and specificity, study designs and the levels of prevention, see the Epidemiology chapter.
The chain of infection
The value of the chain is that it converts "how do we control this?" into a list of concrete options. You do not need to eliminate an organism to stop an outbreak — you need to break one link, and the right choice is usually the link that is cheapest and fastest to break rather than the most scientifically satisfying. Chlorinating a water supply may control a cholera outbreak faster than any measure aimed at the organism itself.
Investigating an outbreak
WHY THE CASE DEFINITION MATTERS MORE THAN IT LOOKS
A case definition states, in advance, exactly who counts as
a case. It has four components:
CLINICAL criteria — what symptoms or findings
PERSON — which population
PLACE — which geographical area
TIME — over what period
Everything downstream depends on it. Case counts, the
epidemic curve, attack rates and the analytic study all use
it, so a definition changed halfway through invalidates the
comparison.
IT IS A TRADE-OFF, and the trade is the same one as in the
screening chapter:
A BROAD (sensitive) definition catches nearly all true
cases but includes people who do not have the disease.
Useful EARLY, when you are trying to find the extent
of the problem.
A NARROW (specific) definition includes mostly true
cases but misses some. Useful for ANALYTIC study, when
a false case would dilute a real association.
Definitions are often graded as SUSPECTED, PROBABLE and
CONFIRMED for exactly this reason — allowing broad case
finding and narrow analysis from the same data.
AND ACTIVE CASE FINDING MATTERS: those who present to
hospital are the severe end of the spectrum. An outbreak
described only from admitted patients will look far more
lethal than it is.
The epidemic curve
💡 Exam angle: the epidemic curve is heavily examined because its shape carries the answer. A single sharp peak means a point source — everyone exposed at one moment, such as one meal. A plateau means a continuous source still operating, such as a contaminated water supply. Successive peaks about one incubation period apart mean propagated spread from person to person, where control must target transmission rather than a single source. And with a point-source curve, counting back one incubation period from the peak estimates when the exposure occurred.
Attack rates
The commonest error in these questions is looking only at the exposed group. A food item eaten by nearly everyone at a wedding will show a high attack rate among those who ate it — but if the rate is just as high among those who did not, the item is innocent. The vehicle is the item with the largest difference between the exposed and unexposed attack rates, and you cannot identify it without both numbers.
Notification, isolation and quarantine
THREE WORDS THAT ARE ROUTINELY SWAPPED
ISOLATION separates people who ARE ILL.
Duration is determined by how long they remain
INFECTIOUS.
QUARANTINE separates people who have been EXPOSED but are
still WELL.
Duration is determined by the INCUBATION
PERIOD — because that is how long it could take
for them to become infectious.
CONTACT TRACING identifies exposed people so that they can
be informed, monitored and offered testing,
prophylaxis or vaccination.
The memory hook: ISOLATE THE ILL, QUARANTINE THE EXPOSED.
NOTIFICATION is a legal duty for specified diseases, and
the rule that matters clinically is that you notify on
SUSPICION. Public health action — contact tracing,
prophylaxis, source investigation — is time-critical, and
laboratory confirmation typically arrives after the window
in which it would have been most useful.
The list of notifiable diseases is set nationally. Look up
Nepal's current list rather than relying on any textbook.
Clinical reasoning: four scenarios
🔍 Case 1 — the curve that names the source
ScenarioForty people develop diarrhoea after a wedding. The epidemic curve shows a single sharp peak, with all cases falling within about a day of each other.
What the shape saysA point source.
ReasoningA single tight peak means everyone was exposed at essentially the same moment — consistent with one contaminated dish rather than an ongoing source or person-to-person spread.
AnswerWork back one incubation period from the peak to estimate when exposure occurred, then compare attack rates for each food item eaten at that meal to identify the vehicle.
🔍 Case 2 — the innocent chicken
ScenarioInvestigating the same wedding, an investigator notes that 90% of those who became ill ate the chicken, and concludes the chicken was the source.
TrapReasoning from the exposed group alone.
ReasoningIf nearly everyone at the wedding ate the chicken, then nearly everyone — ill or well — will report eating it. What matters is the attack rate among those who did not eat it. If that is similarly high, the chicken is not the vehicle.
AnswerCalculate attack rates for exposed and unexposed for every item, and identify the item with the largest difference between them.
🔍 Case 3 — waiting for the laboratory
ScenarioA clinician strongly suspects a notifiable disease in a patient. Samples have been sent, and results are expected in several days. Notification is deferred until confirmation "to avoid a false alarm".
ErrorTreating notification as a report of fact rather than a trigger for action.
ReasoningNotification exists to start time-critical public health action — contact tracing, prophylaxis and source investigation. Several days is precisely the period in which that action would prevent onward cases.
AnswerNotify on suspicion, per the national requirement, and begin appropriate control measures. Confirmation follows and can correct the record if the suspicion was wrong.
🔍 Case 4 — an exposed but well contact
ScenarioA household contact of a confirmed case is entirely well. The team plans to "isolate her until she tests negative".
TerminologyShe is exposed and well — so this is quarantine, not isolation.
ReasoningIsolation applies to people who are ill; quarantine applies to exposed well people, and its duration is set by the incubation period rather than by a test result. A negative test early in incubation does not exclude infection.
AnswerQuarantine for the incubation period per national guidance, with monitoring for symptoms and prophylaxis or vaccination if indicated.
Commonly confused
Confusion
The distinction
Why it matters
Isolation vs quarantine
Ill versus exposed but well
Duration is set by infectivity versus incubation.
Point vs continuous source
Sharp peak versus plateau
One exposure versus an ongoing one.
Continuous vs propagated
Plateau versus successive peaks
Control targets a source versus transmission.
Exposed attack rate alone vs the comparison
You need the unexposed rate too
A common food looks guilty without it.
Broad vs narrow case definition
Sensitive for finding, specific for analysis
The same trade-off as in screening.
Notify on suspicion vs on confirmation
Notification triggers action
Waiting wastes the useful days.
Hospital cases vs all cases
Admitted patients are the severe end
Passive surveillance overstates severity.
Investigating vs controlling
They run in parallel
Control does not wait for the analysis.
Rapid revision
MUST-KNOW FACTS
1. CHAIN OF INFECTION: reservoir, exit, transmission, entry, susceptible host.
2. Breaking ANY ONE link controls the outbreak.
3. Choose the link that is cheapest and fastest to break.
4. Mode of transmission is usually the most practical link to attack.
5. Immunisation acts on the SUSCEPTIBLE HOST link.
6. Confirm the DIAGNOSIS and confirm an OUTBREAK exists first.
7. An outbreak means more cases than EXPECTED for that place and time.
8. The CASE DEFINITION covers clinical criteria, person, place and time.
9. Do not change the case definition midway — it invalidates comparisons.
10. BROAD definitions find cases; NARROW definitions suit analysis.
11. Definitions are graded suspected, probable, confirmed.
12. ACTIVE case finding — hospital cases are the severe end only.
13. Describe by PERSON, PLACE and TIME.
14. POINT SOURCE: single sharp peak — one exposure.
15. CONTINUOUS SOURCE: plateau — the source is still operating.
16. PROPAGATED: successive peaks one INCUBATION PERIOD apart.
17. Count back one incubation period from the peak to date the exposure.
18. ATTACK RATE = ill ÷ at risk over the outbreak period.
19. Calculate attack rates for BOTH exposed and unexposed.
20. The vehicle shows the LARGEST DIFFERENCE between the two rates.
21. A high rate among the exposed alone proves nothing.
22. CONTROL MEASURES DO NOT WAIT for the investigation to finish.
23. ISOLATE THE ILL · QUARANTINE THE EXPOSED.
24. Isolation duration follows INFECTIVITY.
25. Quarantine duration follows the INCUBATION PERIOD.
26. A negative test early in incubation does not exclude infection.
27. CONTACT TRACING offers testing, prophylaxis or vaccination.
28. NOTIFY ON SUSPICION, not on laboratory confirmation.
29. The notifiable disease list is set nationally — check Nepal's current list.
30. Take incubation periods from current official sources.
💡 Exam angle: three things answer most questions here — the shape of the epidemic curve, the attack rate comparison between exposed and unexposed, and the isolation/quarantine distinction. All three are reasoning rather than recall, which makes them dependable marks.
Syllabus points
The chain of infection and its five links
Choosing which link to break
The ordered steps of an investigation
Writing a case definition
Broad versus narrow definitions
Active case finding and why it matters
The epidemic curve and its shapes
Dating the exposure from the peak
Attack rates and the unexposed comparison
Why control does not wait for the investigation
Isolation versus quarantine
Contact tracing and notification on suspicion
Create a free account to tick topics off, take notes as you read, watch the video lessons and get a day-by-day study plan built around your exam date.