TB Manual

3.3.2. Likely period of infectiousness

There is no clear epidemiologic evidence on when infectiousness begins. Pulmonary TB is generally considered to become infectious at the onset of cough (or worsening of a baseline cough), and this should be the priority timeframe for contact investigation. If no cough or other respiratory symptoms are reported, the onset of other symptoms attributable to TB may be used to estimate the onset of infectiousness. In practice, it is often difficult to know with certainty when symptoms began.

The US Centers for Disease Control and Prevention¹⁴ recommends, based on expert opinion, that patients with smear-positive or symptomatic disease should be considered to have been infectious for three months before onset of respiratory symptoms or the first positive finding consistent with TB, whichever is longer. Asymptomatic, non-cavitary TB with a negative smear should be considered infectious four weeks before the first positive finding consistent with TB.

For contact investigation, the period of infectiousness effectively ends when the index patient is placed in isolation from others (this may be before or after diagnosis; at home with no contacts, or on admission to formal airborne isolation in hospital) or is no longer infectious due to TB treatment, whichever comes first. See also Appendix B: De-isolation review and recommendations.

3.3.3. Degree of exposure to the index patient: duration, proximity

Household members are consistently at highest risk of becoming infected, even from index patients with smear-negative disease, as they have very close contact over extended periods.^{10,11,20–24} Beyond this group, there are so many variables in TB transmission that it is difficult to quantify the exact duration of exposure that constitutes a significant risk, and each case should be evaluated on its specific characteristics.^{20,22,23,25,26} In theory, there is no amount of exposure to infectious TB that is absolutely without risk; in practice, almost all transmission occurs with close, prolonged or repeated contact over days or months.

It is not social closeness to the person with TB, but rather, the amount of time in a shared airspace that is the critical issue. For example, IT personnel may report working very closely with team members but spend little time together in shared air space if the work is mainly done electronically; someone who has minimal interaction with the TB patient but works in the neighboring cubicle is at much higher risk. Among household members, those who share a bedroom with the TB patient are at higher risk, independent of other exposure factors.²⁰

For context, 2 large North American studies (each with more than 3,000 contacts) identified 120-250 cumulative hours of exposure as a reasonable threshold for contact investigation.^20,25 Another study found that contacts who had LTBI had a mean of 321 cumulative hours of close exposure to the index patient, compared with 211 hours for uninfected contacts.²³ In school settings, it is generally only students who share classes with an infectious person who are at risk; consistent with the closeness of interactions, schoolchildren are more likely to become infected by a fellow student with TB than by a teacher with TB.^27–31 Two Scandinavian studies documented minimal risk of transmission to children in daycare settings with less than 18-24 hours of cumulative exposure to cavitary, smear-positive adult caregivers.^32,33 A transmission study in a Canadian homeless shelter suggested 5 nights (cumulative) in the same room with a cavitary, smear-positive patient as a pragmatic risk threshold for contact investigation.³⁴ As an extreme example, in a progressively expanded outbreak investigation among university students exposed to an index patient with laryngeal and cavitary pulmonary TB, the risk of infection per hour of classroom exposure was more than 1% for several shared classes; some contacts converted with as few as 3-4  hours of exposure per week over the infectious period.³⁵ Rarely, children less than 5 years old have been infected following extremely short exposure (<30 minutes) in a small space with a highly infectious adult.³⁶

3.3.4. Characteristics of the space where exposure occurred

The room size and ventilation where exposure occurred (eg, large cafeterias or lecture halls vs small seminar rooms) may reduce or facilitate transmission: exposure in cramped, ill-ventilated spaces may lead to transmission in much shorter exposure times. Formal ventilation assessment is not generally necessary. However, in hospitals, where ventilation rates can vary greatly, it may be possible to arrange for facility staff to measure air exchanges per hour in the exposure areas. Exposures in areas with lower ventilation can be prioritized, while those with very high ventilation pose much lower risk outside of unprotected aerosolizing procedures (see Chapter 14: Prevention and Control of Tuberculosis Transmission in Healthcare Settings). Smoking tobacco or other substances with others increases transmission risk, particularly in confined spaces.^37–41 TB transmission is rarely thought to occur outdoors, but has been occasionally been documented in groups who smoke together regularly.⁴²

Figure 2 illustrates the conjunction of time, proximity and characteristics of the shared air space to assess exposure risk. In this example of exposure in an open-plan office, duration of exposure to the infectious patient was relatively long, but in a large, well-ventilated space. A coworker whose desk is very close nearby has a higher risk of transmission than others sitting far away.