Optimal Medication Dispensing for People Living with HIV with and without Other Chronic Diseases in Zambia: A Mathematical Model

PROJECT SUMMARY In the current era of universal antiretroviral treatment (ART), health systems have the challenge of optimizing care for a growing number of diverse groups of people living with HIV and on ART who are also on chronic, life-long treatment for non-communicable diseases (NCDs). Differentiated service delivery (DSD) was

developed as a means to combat suboptimal long-term retention on ART and increase the efficiency of resource utilization by simplifying and adapting care along the cascade, guided by patient preferences and needs, while reducing unnecessary burdens on individuals and the health system. Much remains to be

understood, however, about whether DSD models will work for integrated service delivery for multiple conditions and how to find the optimal DSD model for specific populations and health systems. Using a large patient dataset of routinely collected medical records in Zambia, I propose to create a dynamic microsimulation

model of disease and service delivery – focusing on HIV with and without other chronic diseases such as diabetes – to identify the potential benefits and costs of an optimal medication dispensing strategy. I expect that the cost-effectiveness of alternative drug dispensing strategies will be a function of service volume, patient

outcomes, unit costs, and probability of congestion within clinics. This research will directly inform policy questions relating to how DSD models should be implemented to improve population health outcomes and program cost-effectiveness. Aims 1 and 2 will construct HIV service delivery indices of intensity, capacity,

density, and frequency, assess distributions by patient/facility groups, and evaluate economies of scale of drug dispensing strategies for HIV with and without other chronic diseases such as diabetes. I will build a model to quantify the extent of service capacity and utilization increase using an integrated and multi-month drug

dispensing strategy for HIV and diabetes, based on the proportion of stable/comorbid patients, unit time/cost per service, and clinic capacity. Aim 3 will project the long-term impact and cost-effectiveness of an integrated drug dispensing strategy for HIV with and without other chronic diseases. My model will estimate whether and

to what extent service integration and multi-month drug dispensing can change total treatment initiation and retention in a clinic and how that may in turn change the probability of clinic congestion, unit costs, treatment retention/service quality, and cost effectiveness. The primary research environment, Boston University Medical

Campus, will contribute to the success of the proposed project, as it provides rich, multidisciplinary institutional resources, including a strong commitment to research and training and ample departmental facilities and resources. I have access to the necessary resources and data to complete the proposed research. Through

this mentored research training award, I will gain expertise in infectious diseases transmission modeling and cost-effectiveness, and experience of successful grant writing.