HIV testing in the US: those at highest risk also most likely to put off testing

A large-scale study of the population in the United States has found that HIV testing rates remain relatively low, both nationally and in higher-risk groups. Actual rates of testing also fall short of individual intentions to be tested, especially among those who are most at risk. The findings were published in the October 22, 2007 issue of the Archives of Internal Medicine.

Studies in the United States and elsewhere have consistently shown that many people with HIV are undiagnosed and unaware of their infection. Of the estimated 1.1 million HIV-positive people in the US, an estimated 24 to 27% are unaware of their HIV status.

Policy in the US and in other countries is moving toward more widespread use of routine, opt-out HIV testing, especially outside the "traditional" risk groups of men who have sex with men (MSM) and injection drug users (IDUs). )

The National Health Interview Study (NHIS) is an ongoing large-scale study which collects detailed, wide-ranging health information from over 35,000 adult US civilians annually via in-person interviews. For this report, health policy researchers used pooled data from 154,302 NHIS respondents, gathered in six consecutive surveys between 2000 and 2005.

Participants were asked whether they had ever had an HIV test, when they had most recently been tested, the reason for seeking the test (voluntary, part of routine medical care, or mandatory), whether they planned to be tested in the next twelve months, whether they identified as having any "lifetime HIV risk factor", and how they perceived their own current risk of HIV infection (high, medium, low, or none).

Lifetime risk factors included receiving blood clotting factors, MSM, IDU, sex trade, or having a sex partner with any of the same risks; respondents were only asked if any factor applied and were not asked to specify which. Questions also included demographics, alcohol use, depression, and receipt of other medical care such as hepatitis B vaccination.

Between 2000 and 2005, rates of HIV testing "remained low and relatively unchanged". Overall, those with greater self-perceived risk were more likely to have been tested: see table (p<.001 for all comparisons):

Voluntary tests made up 23.7% of the most recent tests, 44.2% were part of routine medical care (including 17% as part of prenatal care); 20.9% were required for insurance, immigration, marriage or military service. The remaining 11.2% were due to other or unknown reasons. White men consistently reported the lowest rates of testing; between 2000 and 2005, there were modest increases in lifetime testing rates (having ever been tested) among women of all races.

Alcohol use was strongly correlated with testing, perceived risk, and reported risk factors. Compared to no alcohol use, light or moderate drinkers were 1.35 times more likely to have a risk factor or higher perceived risk. People with any level of alcohol use were more likely to have been tested; odds ratios ranged from 1.42 to 1.45 for lifetime testing and 1.12 to 1.14 for past-year testing, with no significant difference between heavier and lighter use. Heavy drinkers were more than twice as likely to have a lifetime risk factor.

Higher depression scores (rated on a scale of 0 to 4) were also correlated with risk, testing, and testing plans. An increased depression score corresponded to an odds ratio of 1.27 of having been tested, 1.20 of a test in the past year, 1.11 of perceived risk, and 1.83 of a lifetime risk factor.

Besides alcohol use, depression score, and self-reported risk, rates of testing also varied significantly with sex, age, education, and race. (Odds ratios for lifetime testing and past-year testing were generally comparable; only past-year testing will be described here.)

Women were more likely to have been tested in the past year than men (OR=1.27, 95% confidence interval [CI] 1.22 to 1.32), as were younger respondents (OR=0.70 per 10 years of age, 95% CI 0.66 to 0.69), college graduates or higher (OR=1.19 vs high school, 95% CI 1.13 to 1.25), Hispanics (OR=1.52 vs whites, 95% CI 1.42 to 1.62) and non-Hispanic blacks (OR 2.55, 95% CI 2.40 to 2.71). Those who had had a flu shot in the past year or a lifetime hepatitis B vaccination were also more likely to have been tested for HIV in the past year (OR 1.35 and 1.48 respectively). (p<.001 for all.)

The other main component of the analysis was a comparison of planned and actual testing. Rates of actual reported testing differed from rates of intention to be tested: considering only voluntary tests, 8.2% of the overall sample reported an intention to be tested in the coming year, but only 2.1% had actually received a test in the past year – a shortfall of 6.1%. (This difference vanished when routine and mandatory testing were also considered.)

Notably, the gap between planned and actual testing was wider for those with greater HIV risk factors. Those with a lifetime risk factor, medium or high self-perceived risk, or heavier alcohol use were all more likely to report an intention to be tested than to actually have received a test in the past year.

For voluntary tests, actual reported testing fell short of intention by 16.0% for those with a lifetime risk factor, 15% for self-perceived medium risk, 15.2% for perceived high risk, and 8.8% for heavier alcohol use. Significant planned-vs-actual gaps were also seen for blacks (15.2%) and Hispanics (10.4%). All of these differences were smaller but remained statistically significant even when routine and mandatory testing were also considered.

In the study methodology, respondents who were already HIV-positive were coded for analysis as "high-risk" rather than as a distinct category. As HIV-positive people (an estimated 0.31% of the US population) would be unlikely to report any intention of further testing, this would introduce a bias toward lower intention in the high-risk group.

The researchers concluded that, in the United States, "HIV testing rates remain low… likely contributing to a substantial number of undiagnosed cases of HIV… populations considered to be at increased risk … still demonstrate the need for improved access to and utilization of testing… [and] factors that inhibit the translation of testing intention into testing behaviour may disproportionately affect groups at higher risk for HIV." However, they note that nearly half of HIV tests occurred as part of medical checkups or pre-natal care, "suggesting that policy initiatives to integrate testing into routine medical care have had some success."