When it comes to researching the effects of cannabis, scientists have many challenges. Despite this, researchers have come up with clever workarounds for gathering the data necessary to evaluate their hypotheses.
During the 2018 Cannabis Science Conference — held Tuesday, Aug. 28, and Wednesday, Aug. 29, in Portland, Oregon — Jeff Chen, director of the University of California, Los Angeles Cannabis Research Initiative, discussed how crowdsourcing data could provide a route to a better understanding of cannabis.
“Some of the advantages of crowdsourcing are that you can gather data very rapidly, you can do it at a very low cost — again, you’re often using a voluntary workforce, and you can gather data sets that would otherwise be hostile to gather,” Chen told those attending his lecture.
Crowdsourcing is done in many industries, from creative production to research and analysis. The company 99Designs is one whose business model is structured around crowdsourcing, allowing clients to hire from a network of freelance designers.
When it comes to sourcing information for cannabis research, there are many pathways researchers can explore, including gathering objective biosensor data from a device such as a Fitbit or smartphone, chart review, and analyzing subjective survey data from applications that monitor consumer behavior.
In Canada, which is set to launch its recreational cannabis program nationwide on Oct. 17, 2018, the government has been crowdsourcing pricing information and purchase locations from citizens to create heat maps since 2010.
“[This is] something that is not quite feasible if you’re just trying to send government officials all across the country to understand this pricing data,” Chen said.
Exploring The Key Types of Crowdsourcing Data
Through crowdsourcing, researchers are able to study people who are purchasing and using cannabis. Unlike research facilities, consumers, in states where cannabis is legal, may purchase and use whichever product works best for them. By gathering data from these consumers, researchers can better understand the cause and effect relationship of a particular product for a particular consumer. When scaled, the findings could provide greater insight into consumer behavior and potential medical applications.
Researchers can gather consumer data from biosensor data or subjective survey data via fitness trackers and smartphone applications. In Canada, the government has been crowdsourcing data from cannabis consumers since 2010. (Unsplash/Photo by Jens Johnsson and used under the terms of Unsplash’s license)
Gathering biosensor data from device such as fitness trackers and heart rate monitors can help researchers in a number of ways.
For example: A consumer wanted to know if a particular cannabis product helps him or her get a better night of sleep. Sleep monitors collect data based on timestamps and movement. If a consumer used a sleep monitor both before and after implementing a cannabis regimen to help get a better night of sleep, researchers can evaluate if the cannabis does, in fact, help or hinder sleep by comparing biosensor data from before and after implementation. This is an example of chart review, which uses an objective comparison of biosensor data to draw a conclusion about a product’s efficacy.
Subjective survey data relies on self-reported actions and effects. Applications such as Strainprint and Releaf allow cannabis users to track therapeutic sessions and the experience of symptoms before and after use. Gathering this data depends on the continued interaction and opinions of the end-user, who tends to drop off if he or she has a negative experience with the product.
“I think it’s a really good first stab at starting to cut through some of this data,” Chen said, but he admits these data aren’t absolute, and comes with some stipulations. “If you’re someone who isn’t finding relief from a cannabis product, you’re gonna stop interacting with this app. So there’s a heavy bias in terms of favorable responses.”
Simply put, people tend to stop using the tool if they don’t like the cannabis product, leaving the data somewhat incomplete.
But when taken together, biosensor data, chart reviews, and subjective survey data can provide researchers with a more complete look at a particular problem and potential solutions.
Sample Size Matters
Chen believes that crowdsourcing has a lot of potential in the U.S. because the consumer sample size is so large.
“First off, obviously we have the largest market in the world,” he said. “One of the tenets of crowdsourcing is that you typically need large data sets. Especially the messier the data, the larger dataset you need to find signal amongst the noise.”
The larger sample sizes that crowdsourcing provides can make for large datasets for researchers. “… [T]he messier the data, the larger dataset you need to find signal amongst the noise,” said Jeff Chen, director of the University of California, Los Angeles Cannabis Research Initiative. (Best Running / Photo by James Cridland and used with a Creative Commons Attribution 2.0 Generic license)
While Israel is leading when it comes to the quality of cannabis research Israeli scientists are producing, there are only tens of thousands of medical cannabis patients from whom to crowdsource data. Compare that with the U.S., wherein the state of California alone more than 900,000 people hold medical cannabis licenses.
The More Product, The Better
Another factor to consider when crowdsourcing data is the vast number of variables when it comes to product availability. In the U.S., each state has a different set of rules and regulations that govern what products are available in its state-specific market.
Many states prohibit the sale of dry flower for consumption. The lack of this availability also means a lack of data that can be gathered from that particular subset of the population.
The more product availability, the more data scientists can gather to evaluate what product types and delivery methods are reported to be most beneficial to consumers.
“With California as an example, you can have suppositories, you can have transdermal patches, sublingual strips, but the diversity of delivery routes might actually be beneficial in that as you start to understand who seems to benefit from what delivery route, which delivery route seems more effective, it’s good to have a wide breadth of options to gather data on,” Chen said.
Lack of Federal Oversight Has Benefits and Drawbacks
In an industry without federal regulations, manufacturers have fewer barriers to get their products to market, which provides a faster pathway to evaluating how the product affects consumers.
“So, again, using California as an example, you don’t need any of those rigorous clinical trials,” Chen said. “You can get your product out there and start learning how your product is working if the cannabinoids [and the terpenes]are being delivered to the blood.”
But the scientific gold standard remains the clinical trial, which involves administration of a product to a person in a highly controlled environment. The clinical trial process is long, arduous and very expensive. Pharmaceutical companies spend millions on research and development to get products approved by the federal government for human consumption.
In the U.S., clinical trials on cannabis are even slower because of cannabis’s classification as a Schedule I drug. This classification requires research institutions to obtain a license to do research on a Schedule I substance. That license comes with stringent requirements on storage, handling, and sourcing.
Only one source is approved to provide cannabis for clinical medical research: the government — and the government, apparently, is incapable of growing good cannabis. In recent years, cannabis provided by the feds for research has come under fire for being unusable and not representative of the cannabis on the market today. The DEA even called for a 157 percent increase in the available supply of quality cannabis to be used for federally approved research.
Despite the challenges the designation of cannabis as a Schedule I substance creates, researchers continue to find creative ways, like crowdsourcing, to gather intelligence to help inform manufacturers and producers to create better, safer, more effective cannabis products.
“How can we let the people inform the scientists about what we should be looking at, what we should be focusing on?” Chen said. “Here in America alone, 30 million Americans are using cannabis. Nobody’s systematically collecting that data. What can we learn from it?”