Time series analysis comparing mandatory and voluntary notification of newly diagnosed HIV infections in a city with a concentrated epidemic

BACKGROUND In Catalonia, a law was passed in 2010 to incorporate HIV infection as a mandatory disease and to reduce under-reporting, perform follow-up and to improve prevention. Currently, there are studies that describe the surveillance of new diagnoses of HIV infection. However, there are no studies that compare the change from voluntary to mandatory notification. This study evaluates the impact of mandatory notification on the registered cases of newly diagnosed HIV infections in a city with a concentrated epidemic. METHODS We analysed newly diagnosed HIV infections that were included in the city register. A descriptive analysis compared the number and the epidemiological characteristics of cases that were declared in two different periods (when notification was voluntary in 2001-2009 and when mandatory in 2010-2011). Time series analysis was conducted, evaluating trends and changes by fitting a Poisson regression model. The Epidemiology Service from the Public Health Agency was responsible for gathering and analyzing data and producing reports on communicable disease for the city. The data used in this study is openly available. RESULTS Overall, 4510 cases of HIV infection were registered, 81.9% were men and 74.5% of them aged over 30. Among men, 55.6% were men who had sex with men (MSM), and among women, the most common route of transmission was heterosexual (HTS) with 65.4%. An annual average of 560 cases was registered between 2010 and 2011. This represents an increase of 33% from the annual average over the previous period (p<0.001). Time series analysis showed that the probability of notification was 2.8 (95% confidence interval 2.4-3.3) times higher with mandatory notification than in the earlier period. There was a statistically significant decrease of missing values in the period of mandatory notification (p<0.001). CONCLUSIONS Mandatory notification of HIV has resulted in an increase in detection of newly diagnosed infections, reduced the levels of missing data and has provided a more realistic picture of the epidemiology of HIV. This information also helps to improve the suitability of interventions aimed at HIV prevention and control.