Data Science Solutions to Characterize Polysubstance Use Behavior from Online Sources

Phase I SBIR proposals will be accepted. Fast-track proposals will not be accepted. Phase I clinical trials will not be accepted. Number of anticipated awards: 1 Budget (total costs): Phase I up to $243,500 for up to 6 months; Phase II of up to $2,045,816 and a Phase II duration of up to 2 years PROPOSALS THAT EXCEED THE BUDGET OR PROJECT DURATION LISTED ABOVE MAY NOT BE FUNDED. Background Polysubstance use occurs when two or more drugs or substances are taken together, either intentionally or unintentionally, and is common. Whether intentional or unintentional, mixing drugs is an unsafe practice and can lead to fatal and nonfatal drug overdose. Despite the clear public safety risks, patterns of polysubstance use are not well understood, and behaviors are constantly evolving. In addition, the drug supply now includes high-concentration tetrahydrocannabinol (THC) products and other new and emerging drugs (e.g., novel psychoactive substances, psychedelics, and adulterants and contaminants like xylazine). Particularly concerning are derived psychoactive cannabis products (DPCP; e.g., hexahydrocannabinol [HHC], tetrahydrocannabinol-O-acetate [THC-O], 8 THC, 10 THC, tetrahydrocannabivarin [THCV]), which can be synthesized through a relatively easy series of chemical reactions by treating cannabidiol (CBD) oil with solvents, acids, and heat. The resulting mixtures of THC (primarily 8 THC) can be contaminated with harmful chemicals such as solvent residues, toxic metals, and unintended reaction byproducts. In addition, these products are often manufactured in at-home or do-it-yourself labs, where there is little control and limited safeguards. Despite the widespread use of these products, quality and safety standards are lacking. Further concerning is the emergence of DPCP with great risk for intoxication. In 1995, the average concentration of 9 THC in U.S. seized cannabis was 4%; in 2017 it was 17%. In a recent study examining THC concentrations in 10 states, most products averaged 70-91% THC. Moreover, manufacturers are distributing cannabis products that are essentially pure THC in different forms (e.g., oils, was, crystals, shatter, edibles), where concentrations exceed 95%. Use of high-concentration THC products include increased risk of addiction, psychosis, and injury. However, traditional methods of public health surveillance may not be responsive enough to this changing risk environment especially with evolving drug control laws (e.g., state legalization of cannabis and other illicit substances). Hence, leveraging the potential of different data sources and data science approaches to conduct data mining, analysis, and visualization of publicly available data from online information sources is now a critical tool in public health surveillance.