Commentary on “Cost-effectiveness of expanding the capacity of opioid agonist treatment in Ukraine: Dynamic modeling analysis.”

Oct 19,2019 | hmclare Commentary

The cost-effectiveness modeling project recently reported by Morozova and colleagues addresses the implementation science question of how we select capacity building strategies, while considering the collective knowledge about the environmental context. This research sits at the intersection of decision modeling and implementation science, offering further support for their complementary applications.

Like many countries, Ukraine is currently grappling with both an opioid epidemic and an increase in transmission of infectious disease among persons injecting drugs. To inform federal funding decisions about how to scale-up treatment of opioid use disorder (OUD) with medication that reduces opioid cravings, the authors developed a model to analyze the cost-effectiveness of various capacity scale-up strategies. This medication treatment can be provided in both specialty and primary care settings but was limited to specialty care settings when the authors developed the model. In contrast to its use in implementation science, a “strategy” is defined by the authors as “the allocation of a specific number of treatment slots to specialty and primary care clinics.” Although several countries already increased treatment capacity for OUD patients by training and supporting primary care providers to begin offering the treatment, this model seeks to determine the most cost-effective approach to capacity building for this setting, using a ten-year time horizon (2016-2025). Two unique contributions of the model parameters were (1) the incorporation of “peer effects” on drug use initiation and relapse and (2) waitlist dynamics, which allowed treatment demand and treatment retention to be capacity dependent instead of assuming that the demand for treatment will always exceed the capacity. The study design benefits from a quantifiable target of increasing treatment capacity slots, in accordance with recommendations from the World Health Organization. Other implementation projects may struggle to define and select one outcome measure to analyze for cost-effectiveness; however, overcoming this obstacle could benefit both short and long-term resource allocation decisions.

Where does this modeling approach fit within the implementation research continuum? After completing a successful pilot of OAT in Ukrainian primary care, Morozova et al. modeled various capacity building strategies to inform next steps in the implementation continuum instead of assuming that the success of their pilot program inherently supported the scale-up of their chosen approach. By conducting cost-effectiveness modeling, the research team reduced their bias towards promoting their own solution. The team benefited from having already developed a substance use treatment model, presumably reducing the time and resources needed to adapt this model to OUD and to the Ukraine setting. The researchers were also able to incorporate context-specific barriers and facilitators into the decision-making process, such as stigma around seeking treatment and national treatment registries that require patient names. Lastly, the model incorporates city-specific estimates of demand, which in turn allows for tailored regional scale-up strategies. This approach to modeling treatment demand could be applied beyond substance use disorders. The completed model then becomes a tool in the policymaker-researcher relationship, allowing both parties to adjust parameters and assess short and long-term predicted impacts.

Read the full abstract