America’s Overdose Crises: A Historical and Data-Driven Perspective

Drug abuse remains a major problem in the United States, with overdose deaths increasing fivefold since 1999 and surpassing 100,000 annual fatalities during the COVID 19 pandemic. Peak mortality occurred in 2022 when more than 110,000 persons died from a drug overdose, mostly synthetic fentanyl and methamphetamines. Although it often receives less attention, alcohol abuse also contributes substantially to preventable mortality, through liver diseases, the development of some cancers, and accidental deaths. In this talk, we briefly discuss the history of drug consumption in the United States over the past 150 years. We then quantify recent drug and alcohol related mortality trends by stratifying data by age, sex, race, geographical location, drug type. We also show how data assimilation methods can be integrated with mathematical models to forecast future trends. Finally, we present a mathematical method to describe the neurobiology of drug addiction, whereby substances of abuse activate and disrupt neuronal circuits in the brain reward system. Our model incorporates known psychiatric concepts such as incentive salience and reward prediction error within a simple and easily interpretable dynamical system framework.