Search for: All records

The soaring drug overdose crisis in the United States has claimed more than half a million lives in the past decade and remains a major public health threat. The ability to predict drug overdose deaths at the county level can help local communities develop action plans in response to emerging changes. Applying off-the-shelf machine learning algorithms for prediction can be challenging due to the heterogeneous risk profiles of the counties and suppressed data in common publicly available data sources. To fill these gaps, we develop a cluster-aware supervised learning (CASL) framework to enhance the prediction of county-level drug overdose deaths. This CASL model simultaneously clusters counties into groups based on geographical and socioeconomic characteristics and minimizes the loss function that accounts for suppressed values and cluster-specific regularization. Our computational study uses real-world data from 2010 to 2021, focusing on the ten states most severely impacted by the drug overdose crisis. The results demonstrate that our proposed CASL framework significantly outperforms state-of-the-art methods by achieving a superior balance in prediction accuracy for both unsuppressed and suppressed observations. The proposed model also identifies different clusters of counties, capturing heterogeneous patterns of overdose mortality among counties of diverse characteristics. 

The ongoing opioid epidemic has been met with the inadequate use of data-informed approaches to respond to the crisis. Although data relevant to opioid and substance use do exist and have been utilized for research in the literature and practice, they have not been prepared for cross-sector coordination and for providing practical intelligence to inform policy planning directly. In this article, we share our views on how data can better serve the purposes of informing policy and planning to maximize population health and safety benefits. Based on our experience in advising state policymakers on developing settlement allocation strategies based on empirical data, we discuss several issues in the data, including coverage, specificity in drug types, time relevance, geographic units, and access, which may hinder data-informed policymaking. Following these discussions, we envision a coordinated data and policy framework as an ideal case to ensure access to meaningful and timely data and harness the full potential of the data to inform policy to combat the continuing epidemic. 

Objectives Oral direct-acting antivirals (DAAs) for hepatitis C virus (HCV) have dramatically changed the treatment paradigm. Our aim was to project temporal trends in HCV diagnosis, treatment and disease burden in France, Germany, Italy, Spain and the UK. Design A mathematical simulation model of natural history of HCV infection. Participants HCV-infected patients defined based on country-specific age, fibrosis and genotype distributions. Interventions HCV screening practice and availability of different waves of DAA treatment in each country. Outcome measures Temporal trends in the number of patients who achieve sustained virological response (SVR), fail treatment (by drug regimen) and develop advanced sequelae from 2014 to 2030 in each country. Results We projected that 1 324 000 individuals would receive treatment from 2014 to 2030 in the five European countries and 12 000–37 000 of them would fail to achieve SVR. By 2021, the number of individuals cured of HCV would supersede the number of actively infected individuals in France, Germany, Spain and the UK. Under status quo, the diagnosis rate would reach between 65% and 75% and treatment coverage between 65% and 74% by 2030 in these countries. The number of patients who fail treatment would decrease over time, with the majority of those who fail treatment having been exposed to non-structural protein 5A inhibitors. Conclusions In the era of DAAs, the number of people with HCV who achieved a cure will exceed the number of viraemic patients, but many patients will remain undiagnosed, untreated, fail multiple treatments and develop advanced sequelae. Scaling-up screening and treatment capacity, and timely and effective retreatment are needed to avail the full benefits of DAAs and to meet HCV elimination targets set by WHO. 

Summary BackgroundThe hepatitis C virus (HCV) care cascade has changed dramatically following the introduction of direct‐acting anti‐virals (DAAs). Up‐to‐date estimates of the cascade are needed to monitor progress, identify key gaps and inform policy. AimTo estimate the current and future HCV care cascade in the United States, nationally and in select subpopulations of interest. MethodsWe used a previously validated mathematical model to simulate the landscape of HCV in the United States from 2011 onwards, accounting for HCV screening policy updates, newer HCV treatments and rising HCV incidence. ResultsBy the end of 2018, of 4.29 million HCV persons alive, 2.71 million (63%) were actively viremic, 2.24 million (52%) aware and 1.58 million (37%) cured. By 2030, under the status quo, of 3.65 million HCV persons alive, 1.88 million (51%) would be viremic, 2.25 million (62%) aware and 1.77 million (49%) cured. The HCV care cascade in 2018 differed substantially by subpopulation: of 1.34 million incarcerated HCV persons, 96% were viremic, 36% aware and 4% cured; of 0.87 million HCV persons in Medicare, 31% were viremic, 72% aware and 69% cured; and of 0.37 million HCV persons in Medicaid, 49% were viremic, 54% aware and 51% cured. Implementing universal screening, providing unrestricted treatment and controlling HCV incidence were factors found to have the largest effect on improving the HCV care cascade. ConclusionsSince the launch of DAAs, the HCV care cascade has shifted towards higher awareness and treatment rates; however, additional interventions are needed to move towards HCV elimination.