For centuries, it was believed that tuberculosis spread primarily when a vulnerable person spends hours in a poorly ventilated space with someone infectious. But new findings suggest that transmission also occurs through casual contact.
Conventional thinking held that enclosed spaces such as households, prisons, and shelters, where people spent long periods of time together, were where most TB transmission took place. Now, writes Biénne Huisman for Spotlight, recent data suggest that casual contact at social settings like shopping malls, restaurants, bars, and places of worship might also play a significant role.
A recent study found that close contact explained only 9% of TB transmission links, while casual contact accounted for 49%. The study, called CONTEXT (Casual Contact and Migration in XDR TB), was conducted in KwaZulu-Natal.
The study’s lead author, Professor Neel Gandhi of Emory University in Atlanta, recently presented the findings at the Conference on Retroviruses and Opportunistic Infections (CROI) in San Francisco. The work has not yet been published in a peer-reviewed medical journal.
The new findings come in the context of other research (much of which was conducted in Cape Town) suggesting TB could be transmitted through breathing, and growing evidence that people with asymptomatic TB can transmit the infection.
Where transmission occurs
Gandhi tells Spotlight that TB transmission has traditionally been linked to prolonged, close contact, with previous studies showing that 9% to 30% of cases could be attributed to this type of contact. A compelling alternative argument, he says, is that the remaining 70% of transmission occurs from casual contact in community settings – which is what their research sought to explore.
“For much of history, we have thought that most transmission occurs through close and prolonged contact, meaning a susceptible person is spending a lot of time in a poorly ventilated area with somebody who is infectious. And so most often we think of households as places where transmission occurs; or congregate settings, places like prisons or homeless shelters.”
On defining casual contact, he says: “In our research, we wanted to understand less intense forms of contact where transmission can occur. So, we understood where people lived, but we also asked them where they spent time in a typical week. The phrase we used was: ‘where do you spend two hours or more, most weeks?’ To try to identify the places people spend substantial amounts of time; and seeing whether they crossed paths with somebody else to whom their molecular fingerprints (of their TB bacteria infection) match.”
Genotyping and geomapping
Gandhi and his colleagues used both genotyping and geospatial mapping to figure out where TB transmission was likely to have occurred.
Genotyping was developed about 30 years ago, allowing us to examine the genetic code of TB bacteria, and to compare similarity between patients’ bacteria.
“TB is a bacteria that keeps its genetic code similar across many generations of replication. In layman’s terms, we call this molecular fingerprinting. If I were to transmit TB to somebody else, my TB bacteria and that person’s TB bacteria’s genetic codes would look similar – almost identical – so we could use this fingerprinting technique by sequencing the genomes of the two bacteria to try to fully get a sense of what the likelihood of transmission was.”
On their geospatial methodology, he says: “When our participants told us where they live or where they spend time in the community, or where they get outpatient healthcare; our team went to those sites and captured a GPS co-ordinates.
“Just like we use GPS for mapping when we’re trying to get around town, we would get specific co-ordinates… If two people went to the same shop, they might have used different names for that shop, or let’s say they went to a shopping mall, they may have used different names for those places; but GPS co-ordinates helped us to determine whether they were at the same place or close to one another. And we used the concept of proximity to try to understand the likelihood that they may have crossed paths.”
In the study they used the metric of “community proximity” defined as a radius of 500m or less.
Essentially, the researchers used genotyping, particularly molecular fingerprinting to help understand the likelihood of transmission between people who have drug resistant TB. And once individuals with similar molecular fingerprints were found, they used geomapping to see whether these patients could be connected through close contact – and if not close contact, then through casual contact.
“The most common place people told us were friends and relatives’ homes. The next most common was shops – so shopping malls.”
Symptoms and disease
He says more work is needed to understand why casual contact transmission is happening. “And it connects to another topic in the TB community that is gaining a lot of attention, which is trying to understand what the association is between symptoms and having the disease.”
He notes that the challenge for researchers is understanding the link between infectiousness and symptoms – specifically, understanding when a person becomes infectious, even if they show no symptoms.
Most TB public health interventions are still based on the assumption that infected people will present at health facilities with classic symptoms such as persistent cough, night sweats, fever, weight loss, and chest pain. However, in recent years, South Africa has been offering TB tests to asymptomatic people thought to be at high risk of TB, as part of its targeted universal testing strategy.
“There’s a concept – what some people have called subclinical TB or asymptomatic TB – so if you were to test a group of people who didn’t come to a health clinic, but you were on a street corner and you tested everybody who went by for TB, we’re coming to appreciate that as many as 50% of people may either not have any symptoms or not have symptoms worrisome enough for them to seek healthcare, but are actually testing positive.”
Gandhi says this reminds him of the early days of Covid-19, when scientists weren’t sure if people only became infectious after showing symptoms.
“Eventually we learned that people were infectious probably for a few days before they developed symptoms. And in the TB world, this may be an area we need to investigate. If there’s the possibility that somebody is infectious when they have absolutely no symptoms, they would go about their regular activities; going to work, school, shopping, religious ceremonies restaurants, and they may unknowingly be infectious with TB. So this is the challenge.”
The bigger picture
Commenting on the findings, Robert Wilkinson, Honorary Professor in the Department of Medicine at the University of Cape Town and director of the Centre for Infectious Diseases Research in Africa, says: “It is interesting, and the proportion of transmission estimated to occur outside the household is a low estimate, but not incompatible with other estimates.”
He says the phenomenon of transmission occurring after brief casual contact is not novel, though, and has been investigated in previous studies.
Asked how the findings presented by Gandhi might affect the outlook on TB interventions, he says: “While close household exposure to infectious TB should prompt clinical evaluation, especially if there are symptoms, finding a close contact by conventional contact tracing approaches is far from invariable. Therefore, in high incidence environments like South Africa, more attention needs to be placed on mass radiographic (X-ray) and, or microbiological screening of asymptomatic persons.”
In a recent public lecture called ‘Hunting Bosons, Finding the Bummock’, Emeritus Professor in Medicine at the University of Cape Town Robin Wood, former CEO of the Desmond Tutu Health Foundation, states: “I think we are changing the paradigm of tuberculosis.”
He notes that research now targets “hidden reservoirs of TB transmission beyond visible, symptomatic cases… (as) TB silently spreads within communities through carriers who exhibit no symptoms yet contribute to transmission”.
On Gandhi’s findings, Wood told Spotlight he would reserve comment until the data are submitted for further peer review and publication.
Study details
The 305 respondents in the CONTEXT study were patients with extensively drug-resistant TB or pre-extensively drug-resistant TB.
They were diagnosed between 2019 and 2022 in the eThekwini, Ilembe, Umgungundlovu, and Ugu regions. The average age was 36, with 137 (45%) women and 216 (73%) of them infected with HIV.
The study was conducted in collaboration with the Durban-based Centre for the Aids Programme of Research in South Africa (CAPRISA).
“CAPRISA played a leadership role in conceptualising the science, development of the protocol and data collection instruments, oversight of all aspects of field work, including screening and enrolling patients, obtaining informed consent from patients or their proxies, field and laboratory data collection, data verification and data clean-up activities for all data used in this study,” says CAPRISA’s deputy director, Professor Kogieleum Naidoo.
CONTEXT was funded through the United States National Institutes of Health (NIH), which recently terminated several grants in South Africa and elsewhere.
“The funding period has ended,” says Gandhi. “Now we’re analysing all of the data, so it won’t be affected by any changes happening at the NIH.”
See more from MedicalBrief archives:
TB transmission in SA – time for a new paradigm?
Most TB patients don’t have persistent cough – global study
Digital X-rays increase TB detection in asymptomatic South Africans
Groundbreaking UCT study upends theory on how TB is spread