RFLP sleuths find clues to Contra Costa increase
Giant cluster found among crack cocaine users
Old-fashioned shoe-leather epidemiology has its strengths, to be sure, but it couldn’t tell investigators in a southern California county why TB rates had jumped 14% between 1995 and 1996. Even more disturbingly, it couldn’t determine why most of the increase occurred among children.
It took newer technology to solve the puzzle, says Charles Crane, MD, MPH, medical director of TB control in Contra Costa County (Rich mond), CA, and former president of the state TB controllers’ association.
When investigators analyzed the pattern of restriction fragment length polymorphism (RFLP) among 149 isolates (representing 93% of 160 culture-confirmed cases, out of a total of 222 cases reported between 1996 and 1997), they were astonished at what they found, he says.
Investigators find epidemiological links
"What we’d thought were cases that had occurred among three or four extended families turned out to be one humongous case cluster," says Crane, who reported the results of the investigation at the 1999 ALA/ATS International Conference held this past April in San Diego.
RFLP analysis, coupled with intensive chart reviews, found that 69 of the isolates, or 31% of total cases, were linked, Crane says. In 18 more cases, even though cultures were negative, investigators were able to find epidemiological links to another case in the cluster. "With RFLP analysis, we were able to go back and figure out all the missing links," he says. "Knowing the fingerprints matched helped us ask the right questions, so that we were able to categorize each secondary case of TB and figure out why it happened."
In turn, that helped TB controllers better understand why cases had been missed in the first place. In some ways, the information wasn’t a total surprise, Crane adds. Investigators long had known, for example, that the county suffered from high rates of crack cocaine use, and that the crack epidemic was fueling the increase in TB rates.
"What we think happens with crack and TB is that when people get together to smoke crack, there’s a lot of coughing," Crane says. "Obviously, if you have TB, you can expose a lot of people that way."
Part of the problem investigators faced was getting crack users to name contacts, Crane adds. "That’s tough, since we’re the authority figures. Practically speaking, about all we could do was try to gain people’s trust and figure out who they’d been with."
Still, because most of the county’s crack abusers are male, routine contact investigations were having a tough time ferreting out contacts and cases that were turning up increasingly among women and children.
"Before we did the fingerprinting, we thought what we had was three or four extended families, all of them unstably housed, with the women living in different households and lots of kids getting exposed at each household," Crane says.
Indeed, the RFLP analysis confirmed what investigators already had suspected: Many cases were unstably housed and had high rates of injection or other kinds of drug use.
Delayed diagnosis of source case
But the analysis revealed a much richer and far more detailed picture as well. For example, investigators found that among the 42 secondary cases, 21 (50%) resulted from delayed diagnosis of the source case.
An additional 10 secondary cases, or 24%, were not uncovered during the initial contact investigation of the source case; another 10 cases (24%) resulted from problems related to evaluation and administration of preventive therapy; and only one case (2%) resulted from the failure of preventive therapy.
Again, no huge surprises here, Crane concedes. "Most transmission happens before you’re aware that someone has TB," he says. Still, it was striking to see how many people had been infected as the result of source cases’ delays in seeking medical care, he adds.
"What this tells us is that we need to work harder at case-finding and try to come up with ways to identify these people earlier," he concludes. "Case-finding is one tool in TB control that hasn’t gotten a lot of attention but that deserves a lot more."
Data from the RFLP analysis don’t mean the county’s TB troubles are over, adds Crane. "Crack cocaine use and its relation to TB is a tough nut to crack. What RFLP does is to give us a more powerful tool. It teaches us lessons, and it leads us to interventions we didn’t know before."
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content