The first inclination of public health researchers responding to the call for increasing use of PrEP among populations at high risk for HIV is to focus on research and interventions that target patients in those populations. However, this is only partially addresses the problem. Patients who seek PrEP from their physicians will only be successful if their care providers know about and are willing to prescribe PrEP. That is to say, we cannot increase PrEP utilization in high risk populations unless PrEP is accessible and available to them. Thus, our focus in this R34 is on care providers. We propose to develop a novel tool that uses medical claims data to infer physician networks and then use those networks to determine how to best implement targeted interventions to encourage primary care providers to offer Pre-Exposure Prophylaxis to their patients at risk for HIV. This will increase access and availability of the treatment, thereby increasing the likelihood of uptake and adherence, critical for strengthening the clinical impact of this highly effective treatment modality. We will use regression models that incorporate data gleaned from the probabilistic physician network we create from claims data to explore whether PrEP prescribing clusters by local physician subnetwork, is associated with connections to infectious disease specialists, or to geographic variation related to features of the HIV epidemic, such as stigma. These models, adjusted for social network statistical dependencies, will help us operationalize and understand the relative contribution of three alternative processes: social influence among physicians, social learning (physician education and diffusion of information), and non-network geographic variation. The results of these analyses will guide identification of appropriate intervention targets for education campaigns by Gilead or interested third parties and identification of possible physician ?change agents? for peer-based interventions based on social learning and social influence processes. Physician targeting tools will be designed and tested based on these findings via simulations on the network we create and via direct contact with a small group of physicians we have identified as targets and change agents.
Pre-Exposure Prophylaxis (PrEP) against HIV infection has been proven effective, but it is underutilized as an HIV prevention strategy in part, due to a lack of familiarity with PrEP treatment regimes, particularly among healthcare providers that serve the most at risk communities. We propose to develop a novel tool that uses medical claims data to infer physician networks and then use those networks to determine how to best implement targeted interventions to encourage primary care providers to offer Pre-Exposure Prophylaxis to their patients at risk for HIV. This will increase access and availability of the treatment, thereby increasing the likelihood of uptake and adherence, critical for strengthening the clinical impact of this highly effective treatment modality.
