Epidemiology of Hiv/aids in Cuba,1986-2008: Modeling Changes in Detection with Temporally Varying Parameters

Background HIV/AIDS prevalence in Cuba has remained consistently the lowest in the Caribbean region. The Cuban HIV/AIDS program features a detection system which has experienced many changes over the years, including the gradual discontinuation of sanatorial system in the 1990’s and increased detections through the family doctors after 2000. Methodology/Major Findings A nonlinear compartmental model with several types of detections is proposed. Analysis reveals that local asymptotic stability for disease-free equilibrium (DFE) can be achieved if: (i) random screening is sufficiently effective; (ii) infection by detected HIV-positive individuals is minimal. If the average number of infections by a known infective exceeds unity, the endemic equilibrium is always unstable and the total number of infectives could increase without bound, provided that the initial infective population sizes are sufficiently large. On the other hand, if the number of infections by a known infective is less than one, then either the DFE or the endemic equilibrium is globally asymptotically stable, leading to a more manageable epidemic, even if the disease is not eradicated. We make use of the Cuban HIV/AIDS data to fit the model during two separate time periods in 1986-2008 to reflect the implementation of different types of detections. The reproduction numbers for each time period are then computed from the two sets of estimated parameter values. Conclusions The results indicate that the HIV/AIDS epidemic in Cuba is currently near an endemic equilibrium. Random screening is still the most important mean of surveillance and detection, while infections by the known infectives must be monitored closely and kept in check, thus highlighting the importance of education and behavior change. The combined use of theoretical analysis and data-oriented application in this work provides new insight into the dynamics of a sexually transmitted disease and its epidemiology.