Date Published: April 25, 2019
Publisher: Public Library of Science
Author(s): Elizabeth R. Stevens, Kimberly A. Nucifora, Mary K. Irvine, Katherine Penrose, McKaylee Robertson, Sarah Kulkarni, Rebekkah Robbins, Bisrat Abraham, Denis Nash, R. Scott Braithwaite, Evelyn Byrd Quinlivan.
A study of a comprehensive HIV Care Coordination Program (CCP) showed effectiveness in increasing viral load suppression (VLS) among PLWH in New York City (NYC). We evaluated the cost-effectiveness of a scale-up of the CCP in NYC.
We incorporated observed effects and costs of the CCP into a computer simulation of HIV in NYC, comparing strategy scale-up with no implementation. The simulation combined a deterministic compartmental model of HIV transmission with a stochastic microsimulation of HIV progression, and was calibrated to NYC HIV epidemiological data from 1997 to 2009. We assessed incremental cost-effectiveness from a health sector perspective using 2017 $US, a 20-year time horizon, and a 3% annual discount rate. We explored two scenarios: (1) two-year average enrollment and (2) continuous enrollment.
In scenario 1, scale-up resulted in a cost-per-infection-averted of $898,104 and a cost-per-QALY-gained of $423,721. In sensitivity analyses, scale-up achieved cost-effectiveness if effectiveness increased from RR1.11 to RR1.37 or costs decreased by 41.7%. Limiting the intervention to persons with unsuppressed viral load prior to enrollment (RR1.32) attenuated the cost reduction necessary to 11.5%. In scenario 2, scale-up resulted in a cost-per-infection-averted of $705,171 and cost-per-QALY-gained of $720,970. In sensitivity analyses, scale-up achieved cost-effectiveness if effectiveness increased from RR1.11 to RR1.46 or program costs decreased by 71.3%. Limiting the intervention to persons with unsuppressed viral load attenuated the cost reduction necessary to 38.7%.
Cost-effective CCP scale-up would require reduced costs and/or focused enrollment within NYC, but may be more readily achieved in cities with lower background VLS levels.
Treatment advances have improved health and survival for persons living with HIV (PLWH), as well as opportunities to prevent transmission.[1–4] However, along the care continuum there are many challenges to maximizing the individual and public health benefits of treatment,[5–11] and outcomes remain persistently suboptimal throughout the US, with 15% of the estimated 1.11 million PLWH being undiagnosed and only 57.9% experiencing viral load suppression within six months of diagnosis (VLS).[12, 13] Although achieving better care continuum outcomes than other large US cities, New York City (NYC) has to make further progress on VLS in order to meet UNAIDS 90-90-90 targets. In 2016, of an estimated 87,700 PLWH in New York City (NYC) 4.2% were undiagnosed, and 80% of diagnosed PLWH achieved VLS.
A previously validated simulation of HIV progression and transmission[28, 29] in NYC was modified to incorporate the observed effects and costs of the CCP intervention. The simulation estimated the impact and cost-effectiveness of a scale-up of the CCP intervention among all persons at apparent risk for sub-optimal HIV care outcomes in NYC (approximately 35% of PLWH) compared to no implementation of CCP. All study methods were approved by the NYC DOHMH IRB.
We provide estimates of the impact and cost-effectiveness of a hypothetical scale-up of a comprehensive HIV care coordination intervention for promoting VLS among persons in NYC with documented barriers to care and treatment. Our analyses suggest that, from a health sector perspective, a broad scale-up of the CCP was not likely to be cost-effective at current costs and observed levels of effectiveness. Our findings are robust over a range of assumptions regarding cost and effectiveness, with the scaled-up CCP in both the time-unlimited enrollment scenario and the 2-year average enrollment scenario becoming cost-effective only after a decrease in programmatic costs or an increase in effectiveness, while the targeted CCP became cost effective with a smaller decrease in programmatic costs. Thus, CCP scale-up could achieve cost-effectiveness through an increased focus on populations for whom the existing program is most effective, and/or by determining and applying an optimal enrollment period.
Our results suggest that CCP implementation within NYC would require a more targeted approach and reduced costs to achieve cost-effectiveness. However, the CCP with a 2-year average enrollment period has the potential to be cost-effective in environments with a lower background of VLS, and/or when restricted to those who have been consistently unsuppressed in the year prior to CCP enrollment, and/or where substantial reductions in programmatic costs could be achieved.