Skip Navigation


November 1, 1999

Urine Testing Proves an Accurate Measure in HIV Screening

HIV-1 testing by using urine samples is an accurate, safe and more acceptable way to screen for the disease, according to collaborative researchers at Makerere University, Kampala, Uganda, Columbia University, and the Johns Hopkins School of Public Health. The study, which compared the results of blood and urine samples from study participants, appears in the November issue of the Sexually Transmitted Diseases.

Senior author Ronald Gray, MD, MSc, professor, Population and Family Health Sciences, Johns Hopkins School of Public Health, said, "Screening for HIV in populations is critical for public health and using urine samples has advantages over blood testing because it does not require the use of needles. Also, urine collection is a safer alternative because urine does not contain the infectious HIV virus which can be inadvertently transmitted to health workers through the handling of blood. The disposal of urine samples is also cheaper and simpler than the disposal of potentially infected blood specimens."

The study was conducted in Uganda and used 222 adults between the ages of 15 and 59 years old. The study subjects provided both urine and blood samples during the same visit. Each urine sample was divided so that half could be tested while fresh in Uganda and the other half was refrigerated and shipped Johns Hopkins for testing. The blood and urine samples were then tested in an Ugandan laboratory. All blood and urine testing was conducted by technologists who were unaware of the results in both the Johns Hopkins and in the Ugandan laboratories. An independent data management team was responsible for statistical analyses.

The results of the study showed that of the 96 HIV-positive blood results, 92 refrigerated urine samples tested positive at Johns Hopkins (95.8 percent). All HIV-negative blood results showed up negative in these urine samples (100 percent). Matched fresh urine samples were tested in 94 HIV-positive blood samples and 128 HIV-negative blood samples. Of the 94 samples of fresh urine, all were found to be HIV-positive (100 percent). However, among the 128 fresh urine samples, only 125 were found to be HIV-negative (97.7 percent).

The study results confirm that urine testing is as sensitive and accurate as testing blood for HIV-1. The fact that the Uganda urine specimens had a higher sensitivity rate (100 percent) compared to the Johns Hopkins urine samples (95.8 percent) may show a difference between using fresh and refrigerated urine samples that are stored for prolonged periods of time. However, it is important to note that the differences between these results are not statistically significant and could have been due to chance.

When asked to give both blood and urine samples, 14.3 percent more study subjects provided urine than blood during repeated follow up sample collection. Although the Ugandan subjects were not asked directly about their preferences or attitudes toward urine or blood collection, in a survey of 180 women in a New York family planning clinic, 21 percent declined HIV testing because they were reluctant to have blood drawn. Therefore, using urine to test for HIV appears to be well accepted and could promote population screening in both Africa and in the United States.

Public Affairs Media Contacts for the Johns Hopkins Bloomberg School of Public Health: Tim Parsons or Kenna Brigham @ 410-955-6878 or