New program developed for contact investigations
New program developed for contact investigations
This mousetrap catches more skin-test positives
Researchers in Alabama have developed a new program for helping field workers do better contact investigations, and the program's creators say its aim is simple: to help contact investigators focus their time and attention on those most likely to be skin-test positive.
"We've been looking at how our field workers do their contact investigations and how we can improve the results," says Michael Kimerling, MD, MPH, director of the project and a professor in the schools of public health and medicine at the University of Alabama in Birmingham. "With the new program, we think our contact investigators will be able to work more effectively and more efficiently."
With a grant from the National Heart, Lung and Blood Institute, Kimerling and his colleagues are using a scoring system to develop a mathematical model that promises to take some of the guesswork out of the process.
The new program is designed for use on a laptop computer, with the data feeding straight into a statewide database. Not surprisingly, state TB managers give the new system high marks because it makes quick work of tedious tasks like generating six-month reports and makes it much easier to keep an eye on the epidemiology of a case. Field workers are testing the program now; as the results come in, investigators will refine the model for the next year, paring down variables until only the most significant remain, Kimerling says.
In something as complex and subjective as contact investigation, it's easy to see why there's a need for more precision. For one thing, there are the problems inherent in the classic "concentric circle" model of contact investigation, in which, the theory goes, contact investigators are supposed to keep skin-testing until the percentage of positive reactions equals the background rate of skin-test positivity.
The problem is that most of the time, nobody really knows what the background rate is, says Kimerling. "Plus, background rates for the general population aren't the same as for high-risk groups, where most cases may be occurring," he adds. "And usually, no one knows what the background rate is, or what it 'should' be, in those high-risk groups, either."
As for those three familiar zones that overlay the concentric circles - home, work/school, and leisure - it's tough to say how much relevance they have to someone living, say, in a homeless shelter.
More weak spots in typical contact investigations have come to light as work on the project has progressed, Kimerling says. Right off the bat, for example, researchers realized that definitions of basic elements of an investigation needed clarification.
When researchers asked invited field workers to take part in focus groups, for instance, they began by asking the participants to explain what was meant by "close contact" or "casual contact." Most of the time, respondents couldn't agree on even those most basic terms, Kimerling says. Definitions of "close contact" ran the gamut from "people who sleep together," to schoolmates and co-workers, to drinking buddies, to pals "who sit outside under the shaded tree together."
The same thing happened when the focus groups were asked to explain the term "infectious." Were people "infectious" when they were smear positive? Culture positive? Did the term refer to patients with a lot of bacilli on their smears, or just a few? And what about X-ray findings?
Likewise, field workers who might resort to a description like "small environment, no ventilation" were hard pressed to say exactly what kinds of situations were encompassed by those words, Kimerling says.
Armed with a sense of what needed fixing, the researchers began looking for ways to replace murkiness with precision. For environment, they started by identifying and quantifying factors related to size and ventilation. Sizes were quantified on a scale ranging from the interior of a car or truck, to a small eating room, to various house sizes, and so forth. To characterize ventilation, researchers used the same process. Was the space open or closed? If closed, was there central air conditioning and heat? If not, were there window units? Fans?
The last task, which is still under way, amounts to deciding which of the identified variables are the most important - that is, which are most strongly correlated to findings of a positive skin test. Knowing that, says Kimerling, not only will make contact investigations more effective, with less guesswork required, it also should lend more flexibility to the process.
"Since most field workers also give directly observed therapy, they are continually having to choose between investigating or doing DOT," he says. "Or say you get a case on a Friday afternoon. Do you have to spend all weekend preparing for it, or can you assign it to fourth or fifth priority, and maybe put it off for a few days?"
The new program also should be helpful when there's a tough judgment call to be made. For example, who does a field worker believe when two accounts contradict one another? Kimerling hopes his new program will shed light on whether cases or contacts, on average, are more likely to give accurate stories.
Battling computer phobia and phone lines
One interesting feature of the program is the way it keeps track of how many people have been interviewed and how many times a field worker has had to go back into the program and make revisions and changes, says Kimerling. That means the program will show how many times a field worker had to interview sources before arriving at an accurate picture of events.
Feedback from field workers testing the new program has been "mixed," Kimerling says, probably in part because the program requires switching from pen-and-paper methods to full-fledged computer literacy. "Also, instead of juggling all this data in their heads, field workers are now required to enter it," he says. It may look like more work at first, even though it really isn't, he adds. Although Kimerling's new program is probably best-suited to rural states such as Alabama, it should work well anywhere, he says. "The fundamentals of a contact investigation are the same everywhere," he says. "The difference is that the respective variables' importance - immigration, say, which doesn't account for many cases here - may need to be adjusted from one state to another."
The only real problem the new program has run into so far concerns the state's phone lines. "How big a file you can transfer by modem varies," Kimerling says. "But we're working on it." Meanwhile, there's a back-up system: Just put the data onto a disk and mail it.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.