Development of a resource allocation index model for allocating resources for HIV/STI prevention

Presenter: Frida Hansdotter, Swedish Institute for Infectious Disease Control

Abstract

Background: HIV continues to pose a serious threat to high risk groups in Europe. In addition, Sexually Transmitted Infections (STI) cause significant harm among the young population in particular. In Sweden, national resources have been set aside for HIV/STI prevention since 1986. A major part has been distributed to regional health authorities on the basis of applications for project support. This application-based resource allocation system has shown disadvantages; resources have not necessarily been allocated according to distribution of risk groups and hence need of preventive measures. Further, the system has become increasingly cumbersome and time-consuming for all parts involved in the application process. Consequently, stakeholders have expressed interest in a more efficient and rational resource allocation mechanism.

Objective: The purpose of this study was to develop a need-based index for allocation of resources for HIV/STI prevention at regional level in Sweden.

Methods: Through literature review, key informant interviews and epidemiological data analysis adequate key variables for influencing resource allocation for HIV/STI prevention were identified. Each of the variables were assessed regarding the availability of valid data at regional level. Variables for which such data were not available were excluded from the index. An index model was developed providing possibility of weighting the variables depending on priorities given by decision makers.

Results: The study identified the following key variables: population – general and target age groups, foreign-born from countries with an HIV prevalence >1 (high-endemic countries), immigrants from high-endemic countries, asylum seekers from high-endemic countries, travelers to high-endemic countries, injecting drug users (IDU), men who have sex with men (MSM), commercial sex workers (CSW), HIV-incidence, HIV-prevalence, chlamydia incidence. The variables IDU, MSM, CSW and travelers to high-endemic countries were discarded because of lack of valid data at regional level. Following variables were included in the index; population 15-54, foreign-born and asylum seekers from high-endemic countries, HIV-incidence and prevalence and chlamydia incidence. Three options for resource allocation were then developed using the index model, with different weights given to the variables. Generally, the more weight given in the index to HIV-incidence and prevalence the more of resources would be allocated to the three metropolitan areas in the country. Such an allocation was considered relevant by decision makers.

Discussion: This study showed that it is possible to develop a rational index for allocation of resources to HIV/STI prevention. However, it also showed that even in a country with rich information valid data at regional level on the distribution of the risk groups often are missing. Proxy variables, such as incidence and prevalence data, may be used to substitute for the absence of data on the variables related to size and distribution of risk groups. The index model for resource allocation developed in the study proved to be flexible and sensitive to policy makers prioritization regarding the key variables. Using this index model facilitates a rational and transparent allocation process. It also enables a more efficient distribution mechanism than an application based allocation system.