Tobacco and TB linked, Indian research suggests

Reports: Smoking drives up cases, deaths

In India, clinicians have long observed a relationship between TB and smoking, but evidence to support or dispute a cause-effect linkage between the two tends to be weak. New data from India and other countries, though still preliminary, show a strong cause-and-effect relationship between tobacco used and TB disease and death, an Indian researcher says.

"Until recently the topic got little attention, partly because there is little TB in countries where the effects of smoking and tobacco have been extensively researched," says Prakash Gupta, PhD, a senior research scientist at the Tata Institute of Fundamental Research in Mombai. "In India, of course, there are still many deaths from TB. So if smoking is causing an increase in those deaths, that is certainly a matter of interest and concern."

U.S. TB experts familiar with previous studies on the subject agree there may be a connection, though data from previous published studies suggest it may be only modest. "When you look at countries like India, China, and Vietnam, all places where there’s a lot of tobacco use, it seems as if it’s a potentially important risk factor," says Michael Iademarco, MD, associate director for science at the Center for Disease Control and Prevention’s Division of TB Elimination. "There are many confounding factors at work, so it’s hard to tease this one out. But certainly, the connection is one we’re looking at."

Gupta conducted a cohort study of about 100,000 people in Bombay by going house-to-house in working-class and poor neighborhoods. The original purpose was not to focus on possible links between TB and tobacco use, but rather to estimate tobacco-attributable mortality of every kind, he adds.

Study shows twice the risk

The cohort included approximately 60,000 women and 40,000 men, all ages 35 or older; after five years, researchers did a follow-up, again going house-to-house. Clinical records were not examined, but subjects reported on their own disease status, if any; and deaths that had occurred in the interim were matched according to cause by referring to records of the Bombay Municipal Corp. The analysis showed there were at least twice as many cases of active TB, as well as more deaths from TB, among smokers than nonsmokers, says Gupta. Other unpublished studies that have been recently completed, another of which also comes out of India, suggest TB risks are elevated by tobacco use as much as fivefold, he adds.

In Gupta’s study, increased risks for death and disease due to TB applied to subjects who used all three kinds of tobacco consumed in India — that is, cigarettes, bidis (an Indian "smoking stick"), and smokeless tobacco — although risks were highest among those who used cigarettes and bidis. Why smokeless tobacco should also drive up risks "is not yet totally understood," adds Gupta, "but we’re seeing the risk there at lower level, and we don’t see it as a causal relationship."

With cigarettes and bidis, evidence for causality is much stronger, he adds. For one thing, data show a positive relationship between response and the dose and duration of smoking. In addition, Gupta notes, a causal link is biologically plausible. "Cigarette smoking affects the lungs in many ways, reducing the immune response and affecting the function of the cilia," he says.

Data from Gupta’s study, along with that data from other recent and earlier studies, were presented at a conference held in November last year in India. The Tata Institute hosted the conference, with additional sponsorship from the Centers for Disease Control and Prevention and the World Health Organization. Three reports on TB and tobacco presented at the conference — two from India, and another from South Africa — have been submitted as a group for publication.