Economic Evaluation of Rapid HIV Testing Approaches

Abstract

Introduction HIV testing is an important step in controlling the spread of HIV worldwide. Data from the UNAIDS suggests that about 50% of people living with HIV are unaware of their diagnosis. Canadian estimates suggest that about 25% of HIV infected persons are unaware of their status. Ontario specific estimates suggests that there may be close to 40,000 persons living with HIV in Ontario; however, 27,000 persons are known and documented to be living with HIV in Ontario. The importance of testing and early diagnosis is well documented with evidence demonstrating that early treatment improves outcomes both for infected individuals and the communities they live in. This premise forms the foundation for the highly effective “treatment as prevention” approach. There are currently many HIV testing approaches including serum and saliva-based technologies. Testing can also be categorized based on duration to receipt of test result. With conventional HIV testing, results become available after greater than 24 hours. Rapid testing approaches such as rapid - facility based testing; rapid - location based testing; and rapid - mobile testing ensure results are available within 24 hours and in many cases, within a few minutes. Effectiveness studies of individual approaches have been conducted, however, studies on the cost-effectiveness of the various approaches in Canada are presently lacking. Most of the available economic studies have been conducted in high prevalence low-income countries, and as such, are not entirely applicable to the Canadian experience. Research objectives The overall aim of the thesis is to investigate the program and performance effectiveness, as well as the economic evidence for the use of rapid HIV testing options compared with existing conventional testing options. To achieve this there were three specific objectives, namely: i. To perform a systematic review of the available cost effectiveness evidence of rapid HIV testing approaches. ii. To perform a second systematic review and determine relative effect estimates for the use of available rapid HIV testing approaches versus conventional approaches iii. To determine the cost effectiveness estimates of the use of these testing approaches as they apply to a Canadian context. Results Rapid HIV Testing for improving uptake of HIV/AIDS services in people with HIV Infection - A Systematic Review and Network Meta-analysis HIV testing has evolved to include rapid testing done in hospitals, non-clinical environments such as bathhouses, places of worship, learning environments and in other cases mobile HIV testing options have also been offered using mobile vans or other motorized vehicles. While there is overall evidence from systematic reviews and meta-analysis of the head-to-head benefits of rapid HIV testing compared to conventional HIV testing, there is insufficient evidence of the indirect benefits of the various types of rapid HIV testing based on location of the tests to identify if there is a differential benefit based on the location where the testing was conducted. This study addressed this by conducting a systematic review and a network-meta-analysis of the HIV testing options. From 3329 articles, we included 4 randomised controlled trials and 1 cluster randomised controlled trial that compared conventional hospital-based HIV testing with rapid HIV testing options stratified by testing location and conducted both head-to-head and indirect testing comparison of these approaches. For our analysis we used RevMan and NetMetaXL respectively. We present our effect estimates as relative risks. Our analysis showed that with direct head-to-head comparison, both facility and location based rapid voluntary counselling and testing (VCT) were associated with improvements in receipt of results (relative risk (RR) = 2.52; (95% CI: 1.33 to 4.75); and (RR = 1.76; 95% CI: 1.46 to 2.12) respectively. Of note, head to head analysis from two studies including 83,825 subjects showed that rapid facility-based HIV testing was associated with increased HIV case finding among participants (RR = 1.90; 95% CI: 1.05 to 3.43). Heterogeneity between the two studies was low (I2 = 0%; p = <0.0001). Finally, we note that the estimates derived from these indirect testing comparisons have wide confidence intervals. The evidence from this review will be of interest to practitioners, researchers as well as policy makers reinforcing not only the clinical effectiveness of rapid HIV testing approaches but also the benefit of rapid location-based testing where this may result in its increased use in population-based HIV programming Rapid HIV testing options versus conventional HIV testing: A systematic review of economic evaluations A systematic review of economic evaluations of rapid HIV testing approaches versus conventional HIV testing in high income, low HIV prevalence settings was conducted. This review also assessed the methodologic quality of the included studies using the Drummond criteria and Consolidated Health Economic Evaluation Reporting Standards (CHEERS) checklist. 1524 records of English language studies were identified from which five articles satisfied the inclusion criteria and included in the review. The review showed that there was economic evaluation evidence to support the use of rapid HIV testing approaches. Estimates showed that rapid HIV testing options were associated with cost per quality adjusted life year (QALY) gained ranging from $42,768 to $90,498. Regardless of HIV prevalence, rapid HIV testing approaches continued to be the most cost-effective option. Economic evaluation of HIV testing - A discrete event simulation The third study was an economic evaluation to examine using Canadian cost estimates to estimate the clinical and economic benefits of the use of rapid HIV testing approaches as a population programming tool in low prevalence high income countries. To achieve this, we developed a modified Centers for Disease Control staging for HIV progression based on CD 4 cell counts and clinical symptoms and applied the effect estimates obtained from the network meta-analysis conducted to meet the second objective. These estimates considered the relative effects of the various testing options. I then applied disease stage costs and resource utilization estimates for a Canadian reference population. The analysis was conducted from a public payer perspective and employed a discrete event simulation (DES) modelling approach model developed in Microsoft Excel to describe clinical progression of HIV patients diagnosed using different testing approaches over time. DES allows for modelling of the various patient populations that exists and accounting for the events (change in disease stages) that can happen differently between various patients. Consistent with the recommendation of the Canadian Agency for Drug Treatments and Health (CADTH), the reference case analysis was conducted using a probabilistic analysis assigning probability distributions to input parameters and randomly selecting values for each parameter from their distribution. The results are presented as cost per person tested, and cost per quality adjusted life years (QALY) gained. Our analysis showed that conventional HIV testing was the most expensive option at $879,019.67 and produced 29.49 QALY’s, while the least expensive option was associated with the use of rapid hospital-based HIV testing at $878,977.47 also producing 29.49 QALY’s. Both rapid HIV testing approaches were less expensive and produced more QALY’s compared to the conventional hospital-based testing. We also found that a willingness to pay (WTP) threshold of $50,000, rapid hospital-based HIV testing was the most likely cost-effective testing option in 80% of replications and the cost effectiveness of rapid hospital-based testing continued at higher thresholds of willingness to pay.