Efforts to increase HIV testing in urban areas—where the U.S. epidemic remains concentrated and roughly one in six people living with the virus are unaware they’re infected—are showing signs of success, notably in New York City’s Bronx borough, according to telephone survey results published online ahead of print by the Journal of Acquired Immune Deficiency Syndromes (JAIDS).

In June 2008, NYC launched The Bronx Knows, a three-year initiative with dual goals of testing every Bronx resident who had never been screened for HIV and linking those who tested positive to HIV care and services. The Bronx was selected for its high HIV prevalence and death rate—nearly 3 percent of residents of some Bronx neighborhoods are estimated to be living with the virus—and in response to requests from community leaders that NYC Department of Health and Mental Hygiene launch an HIV testing initiative.

A formal evaluation of the program has not yet been published or presented. What the JAIDS paper makes available are results from an annual telephone survey representative of NYC adults to compare 2005 and 2009 estimates of HIV testing prevalence among Bronx residents. Along with these findings, Julie Myers, MD, MPH, of the city health department’s Bureau of HIV/AIDS Prevention and Control and her fellow authors used NYC HIV surveillance data to evaluate changes in the percentage of people being diagnosed with HIV and AIDS at the same time, an indicator of delayed HIV testing.

The proportion of the Bronx population reporting HIV testing increased significantly during the five-year period included in the analysis.

In 2005, 69.3 percent of those surveyed reported having ever been tested for HIV. In 2009, 79.1 percent of those surveyed reporting having been tested at least once, representing a relative increase of 32.2 percent. This difference was statistically significant, meaning it was too great to have occurred by chance.

Of note, the proportion of Bronx residents who have ever tested for HIV is substantially higher than those documented nationwide. According to national estimates, only 40 to 45 percent of all people residing in the United States in 2008 had ever been tested for HIV at any point.

In 2005, 36.9 percent of survey respondents said they had been tested—either for the first time or again—within the 12 months preceding the telephone interview. In 2009, 48.8 percent had been tested within the 12 months preceding the telephone interview, representing a 32.2 percent increase over the 2005 data. Here, too, the difference between the two years was statistically significant.

Looking specifically at those who were tested within the 12 months leading up to the interview, rates between 2005 and 2009 increased significantly among those between 24 and 44 years old, men, blacks, Latinos, those living under or near the poverty level, those with less than a high school education, those who identified as heterosexual or bisexual, those with a primary health care provider and those with private insurance.

The only significant decrease in testing rates in the 12 months leading up to the interview was among whites.

Across the board, the proportion of people being diagnosed concurrently with HIV and AIDS fell from 30.1 percent in 2005 to 23.6 percent in 2009—a 21.8 percent relative decrease. The proportion of concurrent diagnoses fell among all age groups, men and women, all race/ethnicities (except for Asian/Pacific Islanders or other race), and all transmission risk categories.

However, Myers and her colleagues noted that some groups still had very high rates of concurrent diagnoses in 2009. For example, 44 percent of those between 45 and 64 years old were being diagnosed with HIV and AIDS at the same time. Among those who reported heterosexual sex or “other” as their HIV risk category, concurrent diagnoses were documented in about 28 percent.

Other groups had relatively low proportions of concurrent diagnoses in 2009, including those between 18 and 24 years old (7.9 percent), whites (13.8 percent), and men who have sex with men (12.6 percent).

While the authors acknowledge that telephone surveys come with numerous challenges that can dilute a study’s accuracy, they conclude that their findings “underscore the potential effectiveness of large-scale efforts to impact HIV testing, care and treatment.”