National trends in rates of undiagnosed HIV in key populations in Australia: a retrospective observational analysis from 2008 to 2019

IntroductionDetermining the proportion of people living with HIV who are undiagnosed is critical for Australia to accurately assess the countrys progress toward UNAIDS 95-95-95 interim targets by 2025 and progress with elimination goals. We aimed to investigate the utility of two additional measures, the Yearly Diagnosed Fraction, and the Case Detection Rate, in elucidating trends over time in the rate of undiagnosed HIV. MethodsUsing routinely collected national HIV surveillance data, we produced estimates for the number living with undiagnosed HIV and the number of new HIV infections using the European Centre for Disease Protection and Control (ECDC) HIV modelling tool to calculate the Total Diagnosed Fraction (TDF), the Yearly Diagnosed Fraction (YDF), and the Case Detection Rate (CDR) from 2008 to 2019 for Australian-born and overseas-born individuals who have reported having male-to-male sexual contact, and heterosexual women and men. We calculated rate ratios using Poisson Regression models to compare trends for each sub-population. ResultsOver 2008-2019, each metric for the Australian-born male-to-male sexual contact group improved consistently over the period with the case detection rate rising above one in 2013. The total diagnosed fraction for the overseas born group fell slightly from 85.0% to 81.9%, the yearly diagnosed fraction fell from 23.1% to 17.8% and the case detection rate stayed below one, falling from 0.74 to 0.63. In the heterosexual group, women and men had consistent increasing trends for the total diagnosed fraction and yearly diagnosed fraction but with women having consistently higher estimates than men, 92.6% vs 80.8% and 26.3% vs 17.4%, respectively in 2019. However, heterosexual men had a declining case detection rate, falling below one in 2019 (0.83), compared to an increase for women. ConclusionThe additional metrics presented provide important information on Australias progress toward HIV elimination. The more dynamic changes in the undiagnosed population seen highlight diverging trends for key populations with growing undiagnosed populations not seen in the total diagnosed fraction. Periodic trend analyses will help strengthen the use of the metrics and ease of interpretation for national surveillance reporting.