Testing for human immunodeficiency virus infection among tuberculosis patients in Los Angeles

To investigate patterns of testing for human immunodeficiency virus (HIV) infection in tuberculosis patients, we obtained data from the tuberculosis registry and from medical records of 500 tuberculosis patients in Los Angeles County. Sixty-three percent of tuberculosis patients were tested for HIV infection, and multivariate logistic regression analysis revealed that the likelihood of testing was significantly higher among males, persons between 20 and 49 yr old, individuals with HIV risk factors, and patients treated by public health care providers. A minim um of 12% of the total sample, 2% of females, 2% of those outside the ages of 20 to 49, 7% of patients without recorded HIV risk factors, and 17% of patients treated by private practitioners were infected with HIV. These results suggest that health care providers, particularly private practitioners, do not follow national recommendations for universal HIV testing of tuberculosis patients. Instead, HIV testing is preferentially performed on patients perceived to be at high risk for HIV infection. Because HIV seroprevalence is at least 2 to 7% in ''low-risk'' groups, failure to test these patients may result in significant missed opportunities for diagnosis and treatment of HIV coinfection.