Abstract
BACKGROUND: In the context of an outbreak of HIV among people who inject drugs in Glasgow, Scotland, identified in 2015, our objectives were to: (1) develop epidemiological methods to estimate HIV incidence using data linkage, and (2) examine temporal changes in HIV incidence to inform public health responses. METHODS: This was a retrospective cohort study involving data linkage of laboratory HIV testing data to identify individuals with a history of drug use. Person-years (PY) and Poisson regression were used to estimate incidence by time period (pre-outbreak: 2000-2010 and 2011-2013; early outbreak: 2014-2016; ongoing outbreak: 2017-2019). RESULTS: Among 13 484 individuals tested for HIV, 144 incident HIV infections were observed from 2000 to 2019. Incidence rates increased from pre-outbreak periods (1.00/1000 PY (95% confidence interval, CI: 0.60-1.65) in 2000-2010 and 1.70/1000 PY (95% CI: 1.14-2.54) in 2011-2013) to 3.02/1000 PY (95% CI: 2.36-3.86) early outbreak (2014-2016) and 2.35 (95% CI 1.74-3.18) during the ongoing outbreak period (2017-2019). Compared with the pre-outbreak period (2000-2010), the incidence rates were significantly elevated during both the early outbreak (2014-16) (adjusted incidence rate ratio (aIRR) = 2.87, 95% CI: 1.62-5.09, p < 0.001) and the ongoing outbreak periods (2017-19) (aIRR = 2.12, 95% CI: 1.16-3.90, p = 0.015). CONCLUSIONS: Public health responses helped to curb the rising incidence of HIV infection among people with a history of drug use in Glasgow, but further efforts are needed to reduce it to levels observed prior to the outbreak. Data linkage of routine diagnostic test data to assess and monitor incidence of HIV infection provided enhanced surveillance, which is important to inform outbreak investigations and guide national strategies on elimination of HIV transmission.