Return on investment 2: evaluating the cost-effectiveness of needle and syringe programs in Australia 2009

Appendix A: Development of epidemic mathematical transmission model

Page last updated: 2009

An epidemic mathematical transmission model of HIV and HCV among Australian IDUs was formulated to dynamically describe the change in the number of people in the population over time according to disease states. The model considered heterogeneity in injecting behaviour reported by the Illicit Drug Reporting System (IDRS)56-62: IDUs who did not inject in the last month, injected weekly or less, injected more than weekly, injected once daily, injected two to three times per day, and injected more than three times per day. The frequency of sharing injecting equipment, number of people with whom equipment is shared, number of times each syringe is used before it is disposed, and frequency with which syringes and other injecting equipment (e.g. spoons, tourniquets, etc) are cleaned before reuse, and the efficacy of cleaning equipment contaminated with HIV or HCV were all factored into the model's calculation of the per capita rate of IDUs becoming infected. The model also tracked the entry of new injectors into the population and the rate of ceasing injecting behaviour, while also matching the assumed dynamic number of IDUs in the population over time. Drug-related, disease-related, and background death rates were also included. All parameter values were estimated based on exhaustive searching of the relevant literature and available data from Australian reports and databases (see Table B.1).

Data were also stratified by each Australian state and territory as well as Aboriginal and Torres Strait Islander populations. The numbers of IDUs in each jurisdiction were included, based on various indicators, along with the dynamic number of sterile syringes distributed by NSPs to these populations over time. We considered different syringe coverage rates within the IDU populations.

Force of infection and analysis of 'static' incidence
Dynamic transmission model
Assumptions for modelling secondary transmissions