Personal, National Income Positively Linked to Rates Of Voluntary HIV Testing

P. Doskoch

First published online:

Wealthier individuals are more likely than poorer ones to be tested for HIV, especially in less affluent countries, according to an analysis of data from a global survey.1 Within countries, the odds that a person in the wealthiest income quintile has been tested for HIV are twice those of someone in the poorest quintile (odds ratio, 2.1). This disparity is substantially greater in lower-income countries (2.4) than in better-off ones (1.4).

Voluntary counseling and testing is a key component of HIV programs throughout the world, and generally includes not only testing but also risk assessment, emotional support and referral of infected individuals to appropriate providers. Some evidence indicates that testing and counseling reduces the likelihood that clients will engage in HIV risk behaviors, such as unprotected sex. However, cost may be a barrier to testing, particularly in developing countries.

To examine the relationship between voluntary testing and wealth—both within countries and among them—investigators examined data from the 2002–2003 World Health Survey. The survey, whose questions were carefully constructed and standardized to ?facilitate cross-cultural comparisons, was ?administered in 70 low-, middle- and high-?income countries. Forty-nine chose to administer the survey's sexual and reproductive health module, which was given only to participants aged 18–49.

Respondents (except those who had given birth in the past two years) were asked whether they had been tested for HIV during the past 12 months. They also provided information on a range of social and demographic measures, including their gender, education, age, residence (urban or rural), marital status and household assets (which the investigators used to classify the respondents into country-specific income quintiles). Three country-level variables were also included in the analysis: gross domestic product per capita, national health expenditures per capita and HIV prevalence. Data for these measures were generally obtained from the World Bank's World Development Indicators database. The researchers conducted multilevel logistic regression analyses that allowed them to simultaneously assess associations between HIV testing and individual and country-level variables. Countries were categorized as higher income or lower income if their gross domestic product per capita was one standard deviation above or below the sample's mean, respectively.

The survey's sexual and reproductive health module was administered to 267,926 men and women, although only 110,638 answered the question about HIV testing. Another 3,933 respondents failed to provide information on one or more covariates, resulting in an analytic sample of 106,705. Globally, 9% of respondents had been tested for HIV in the past year; the proportion ranged from less than 1% (in Bangladesh, Bosnia and Herzegovina, China, Laos and Pakistan) to more than 25% (in Malaysia, Russian Federation, South Africa and United Arab Emirates). In 35 of the 49 countries, the prevalence of testing was less than 10%.

In regression analyses, all of the individual-level variables were associated with HIV testing. The odds of testing increased more than twofold for each increase in wealth quintile (odds ratio, 2.1); they were also greater among females than among males (1.2), and higher among urban residents than among rural dwellers (1.4). Respondents who were currently or formerly married, or who were cohabiting, were more likely than never-married, noncohabiting individuals to have been tested (1.2–1.5). Finally, the odds of testing increased with each year of schooling (1.1) and decreased with each year of age (0.99).

National wealth was also associated with differences in testing rates across countries. Individuals in higher-income countries were more likely than those in lower-income countries to have been tested for HIV; for example, among respondents who were in their country's lowest income quintile, the probability of testing was 1% if they lived in a lower- income country and 5% if they lived in a ?higher-income one. Moreover, disparities by income within countries were greater in lower-income countries than in higher- income ones; in the former, the odds of a respondent in the wealthiest quintile having been tested were 2.4 times those of a person in the poorest quintile, whereas in higher income countries the odds of testing among the wealthiest respondents were only 1.4 times those among the poorest participants. The probability of testing was unrelated to a country's health expenditures but positively related to its HIV prevalence (odds ratio, 1.1 per 1% increase in prevalence). —P. Doskoch

The authors note several limitations of the study, including the age of the data, the exclusion of new mothers and persons younger than 18 or older than 49 and the likelihood that some participants did not answer the question about HIV testing truthfully, even though interviewers explicitly stated that they did not want to know the test results. Nonetheless, the findings indicate not only that wealthier individuals are more likely than poorer ones to be tested for HIV, but that the disparity is especially great in poorer countries, suggesting that in the developing world "public spending disproportionately benefits those in the highest income quintiles." The investigators recommend that international programs and policies aimed at addressing the HIV/AIDS pandemic in less-developed countries "better [target] programs to low-income individuals to help reduce social and economic barriers associated with testing."


1. Larose A et al., Global income-related inequalities in HIV testing, Journal of Public Health, 2011, 33(3):345– 352.