Simulating HIV transmission dynamics: An agent-based approach using NetLogo.

Many studies have employed Agent-Based Modeling (ABM) to study the complex dynamics of HIV transmission. However, these studies often focus narrowly on specific subpopulations and limited parameters, restricting the potential of ABM to capture the intricate interrelationships between diverse subpopulations. This paper proposes an improved ABM to simulate HIV epidemic dynamics, exploring parameters such as sexual behaviors, drug use, condom usage, testing frequency, and treatment-seeking behavior. Calibrated with empirical data from the Philippines, the simulation closely aligns with national HIV infection trends from 2010 to 2018, achieving a Mean Absolute Error (MAE) of 3.5 and a Mean Squared Error (MSE) of 14.9. Findings indicate that extended commitment duration, consistent condom use, sexual inactivity, high testing frequencies, and strict adherence to treatment significantly lowers HIV transmission rate. The simulation results mirror trends observed in other studies, suggesting that the enhanced model provides reliable and expected outcomes. The results also illustrate the relationships between different factors, highlighting the model's comprehensive approach. Furthermore, the model effectively captures the trends within a 10-year period, predicting the cyclical rise and fall of new infections every 2 to 3 years, along with the overall decline in the percentage of new infections over time. This paper represents an initial step, prompting further efforts to enhance understanding and public health interventions.