UCLA

Objective

This study examined temporal trends in HIV testing among U.S. older adults (50-64 years of age) before and after the release of CDC's routine HIV testing recommendations in 2006.

Methods

The sample (n=872,797; 51.4% female) comprised 2003-2010 Behavioral Risk Factor Surveillance System respondents in the oldest categories to which the recommendations apply: 50-54 years (34.5%, n=301,519), 55-59 years (34.1%, n=297,865), and 60-64 years (31.3%, n=273,413). We calculated (1) four-year pooled prevalences of past-year HIV testing before and after 2006, when the recommendations were released; and (2) annual prevalences of HIV testing overall and by age category from 2003-2010. Using weighted, multivariable logistic regression analyses, we examined binary (pre- vs. post-recommendations) and annual changes in testing, controlling for covariates. We stratified the data by recent doctor visits, examined racial/ethnic differences, and tested for linear and quadratic temporal trends.

Results

Overall and within age categories, the pooled prevalence of past-year HIV testing decreased following release of the recommendations (p<0.001). The annual prevalence decreased monotonically from 2003 (5.5%) to 2006 (3.6%) (b=-0.16, p<0.001) and then increased immediately after release of the recommendations, but decreased to 3.7% after 2009 (b=0.01, p<0.001). By race/ethnicity, testing increased over time among non-Hispanic black people only. Annual prevalence also increased among respondents with recent doctor visits.

Conclusion

CDC's HIV testing recommendations were associated with a reversal in the downward trend in past-year HIV testing among older adults; however, the gains were neither universal nor sustained over time.