A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
Contact tracing interrupts transmission by identifying and isolating exposed individuals before they become infectious. Effectiveness depends on the speed of case identification, proportion of cases identified, completeness of contact elicitation, and timeliness of isolation. For diseases with short incubation periods or presymptomatic transmission, tracing must occur within hours to be viable.
Compare contact tracing effectiveness for different pathogens (tuberculosis, measles, COVID-19) with different generation times and presymptomatic transmission frequencies. Use delay analysis to show how response time affects outbreak control.
Contact tracing is a targeted interruption of transmission chains. Rather than applying the same intervention to an entire population, it finds the specific individuals most likely to be incubating infection — the people who have already been exposed — and removes them from the transmission network before they can infect others. This makes intuitive sense from what you know about transmission dynamics: if R (the effective reproduction number) must fall below 1 to stop an outbreak, contact tracing can push R downward by isolating a portion of secondary cases before they themselves transmit.
The mathematics of contact tracing effectiveness depend on timing. Every pathogen has a characteristic generation time — the interval between when a person is infected and when their contacts are exposed. Contact tracing can only work if exposed contacts are found and isolated before the end of their incubation period (or the onset of infectiousness if presymptomatic transmission occurs). For tuberculosis, with an incubation period of weeks to months, there is ample time: contacts can be identified, tested, and started on preventive therapy well before they become infectious. For COVID-19, where the median incubation period is around 5 days and presymptomatic transmission begins 1–2 days before symptom onset, a contact tracing program that takes 3 days from symptom onset to contact notification will miss a substantial fraction of transmission — the contacts may already have infected others before they are reached.
Completeness is the second determinant of effectiveness. Contact tracing assumes that cases can identify their contacts and that contacts can be reached and persuaded to isolate. In a household or small workplace setting this is feasible. In an urban environment with many anonymous contacts — public transit riders, bar patrons, gym users — exposure events occur that cases cannot recall and contacts cannot be traced. This is why cluster-based or network-aware tracing approaches (tracking superspreading events rather than individual contacts) emerged as important refinements during COVID-19. Digital contact tracing using Bluetooth proximity data was proposed to address anonymity, but required widespread smartphone adoption and raised privacy concerns that limited uptake.
The third pillar — often underemphasized — is that identification without support fails. Even a contact who is correctly identified, promptly notified, and willing to isolate may not be able to do so without paid sick leave, alternative housing, and access to food. Contact tracing programs that operated in jurisdictions with strong social support infrastructure consistently achieved higher isolation rates than those relying on individual compliance alone. This connects to the broader principle from outbreak investigation: identifying who is exposed is a technical exercise, but achieving isolation is a social and economic one. A technically excellent tracing program operating in a context of economic precarity and stigma will underperform a more modest program embedded in a system that makes quarantine feasible and non-punitive.