CDC national study of CIs finds gaps

Results will be used in training

A large-scale study of contact investigations (CIs) at 11 U.S. sites found considerable variation in how the investigations are performed and what data are gathered, says Suzanne Marks, MPH, MA, an epidemiologist with the Division of TB Elimination (DTBE) at the Centers for Disease Control and Prevention. She served as the study’s principal investigator.

In many instances, even high-risk contacts (such as children under age 6 and contacts who are HIV-positive) are not started on preventive therapy (PT) regardless of their skin-test results, Marks says.

Her study, along with a second one directed by Mary Reichler, MD, a medical epidemiologist in the DTBE, are "the first major, nationwide studies that have been done on contact investigations," Marks says. The CDC will use results of the two studies to improve contact investigation and training, she adds.

Researchers for the Marks study reviewed records from CIs for 1,048 cases of infectious pulmonary TB. The total number of contacts investigated was 13,029, including 5,625 close contracts and 7,404 casual (or status "unknown") contacts.

Investigators found a mean number of five close contacts per case and a median of four. The number of close contacts in the study ranged from zero to 75. As for casual contacts (or those of unknown status), the mean was seven, and the median was zero (reflecting the fact that the number of casual contacts ranged from zero to a mind-boggling 822).

On the basis of the study, investigators reached the following preliminary conclusions:

    • Procedures for CIs differed widely among sites, including who does the CI, who supervises it, and what data are collected.

    • Cases with zero contacts were more likely to be U.S.-born, males, substance abusers, or homeless.

    • For contacts, high-risk factors for disease (especially HIV) and for nonadherence (substance abuse, homelessness) often are not recorded.

    • Only half of close contacts known to be at high risk for disease were placed on PT regardless of tuberculin skin-test (TST) results.

    • Only two of the 11 sites recorded the date of the contact’s last exposure to the infectious case, which forms the basis for determination of the follow-up TST date.

    • There was no record of a follow-up TST for 43% of initial TST negatives.

    • For a third of those who started on PT, completion status was listed as either "refused/uncooperative," "lost to follow-up," or "unknown."

    • Fifty-seven percent of those started on PT were known to have completed it, but only 44% of those eligible for PT were known to have done so.

    • Those started on PT who didn’t complete took a median of two to three months of PT.

    • Asian and foreign-born persons were more likely to complete PT.

    • Children under age 6 were not more likely to complete PT.

    • For other high-risk groups (HIV-positive individuals and substance abusers), data were too limited to draw conclusions about rates of PT completion.