Category: <span>Data Analysis</span>

Data Analysis

Is There a Connection Between Opioid-Related Deaths and Opioid…

  • August 21, 2020October 29, 2022
  • by mikelabun

Statistics Canada has suggested that Canada’s present drug crisis is partly fueled by prescriptions (2019). In January of 2020, we were not able to find data to substantiate this statement, so we turned to US data to look for evidence of a relationship between opioid deaths and related prescriptions. As prescriptions of oxycodone and hydrocodone have increased over time, so have opioid-related deaths, but the relationship is unclear. While New York State has reduced prescriptions of the more addictive of the two substances, hydrocodone, this has unfortunately had no observable impact on opioid-related deaths.

Background

Deaths due to opioid overdose have spiked in Canada, doubling between 2015 and 2017. Statistics Canada has posited that while illegal drugs with fentanyl are part of the problem, “…prescription opioids have also contributed to the crisis… High rates of opioid prescribing mean that more people are being exposed to the risks of opioids” (2019).
We were not able to locate data to back up Statistics Canada’s statement, but we were able to get informative data from our neighbours to the south: The Washington Post (2020) has obtained and published detailed information about supplier-retail sales of the US’s two most prevalent prescribed opioids, oxycodone and hydrocodone, which we combined with the US’s National Centre for Health Statistics’ (NCHS) annual estimates for deaths due to opioid poisoning (Hedegaard et al., 2020). We then used this data to explore the relationship between the prescription sales and opioid-related deaths. Due to the massive size of the dataset, we decided to focus on two of the United State’s best-known states: California and New York.

Prescriptions and Income

The maps below allow the reader to compare prescriptions with per capita income. Counties are coloured according to per capita income, and each county has a dot that represents total amount of opioid prescriptions, in kilograms. In general, income and opioid colours darken together, indicating that as income increases, so do prescriptions.

California and New York per capita income (shading) and prescriptions (circles).

We adopted per capita as a measure because there is a significant difference in the population of California (38.8 million) and New York (12.9 million).

Comparing Death Rates

Below are the number of opioid-related deaths per 1000 people per year in New York and California as assessed by the NCHS. As you can see, New York began with a lower opioid-related death rate, but beginning in 2010, NY’s rate started to climb relative to CA’s, becoming equal to CA’s in 2012.

Death rates in New York and California over time. Note that to focus on changes, all the “Death Rate” charts in this report have a shortened axis that ranges from eight to 13.5 deaths per 1000 people per year.

The Relationship Between Prescription and Death

So, then, what is the relationship between opioid prescriptions and opioid-related deaths? Our attempt to grapple with this question in California’s case is below. The green bars show death rates, while the lines show hydrocodone and oxycodone prescriptions, respectively. While the two ratings both show a subtle rise from 2006-2011, they go in opposite directions after that. In other words, even though something happened in California to curb the number of prescriptions, the death rate continued to climb.

Graph showing death rates (bars) as compared with amount of prescribed opioids (lines).

The chart below indicates the same information for New York State, and it’s more complicated but quite interesting in its own right. Beginning in 2014, due to the fact that hydrocodone is generally thought to be more addictive than oxycodone, New York restricted the distribution of hydrocodone without a doctor’s visit, and at the same time, all products containing hydrocodone in any amount and in any combination were added to Schedule II – Controlled/Restricted Substance (Drug Enforcement Administration, 2014). Buying patterns followed suit, and sales of hydrocodone decreased, while oxycodone significantly increased.

Sadly, the problem of opioid-related deaths skyrocketed regardless, as New York’s death rate began marginally lower than California’s (New York was 9.0/1000 while California was 10.3/1000) but actually ended up marginally higher (with New York at 12.9/1000 and California at 12.6/1000).

Conclusion

We know that both opioid-related deaths and opioid prescriptions in the United States have climbed from 2006- 2014, and so a correlation is suggested (Centers, 2020). However, New York’s attempt to reduce harm by limiting hydrocodone has been unsuccessful, suggesting that hydrocodone, at least by itself, is not a driver. More work will need to be done to compare other communities’ approaches in an effort to discover best practices.

References

Centres for Disease Control and Prevention (Ed.). (2020, March 19). Drug Overdose Deaths. Retrieved from https://www.cdc.gov/drugoverdose/data/statedeaths.html

Drug Enforcement Administration. (2014, August 22). Rules – 2014 – inal Rule: Rescheduling of Hydrocodone Combination Products From Schedule III to Schedule II. Retrieved from https://www.deadiversion.usdoj.gov/fed_regs/rules/2014/fr0822.htm

Hedegaard, H., Miniño, A. M., & Warner, M. (2020, January 30). Drug Overdose Deaths in the United States, 1999–2018. Retrieved from https://www.cdc.gov/nchs/products/databriefs/db356.htm

Public Health Agency of Canada. (2018, March 17). Government of Canada. Retrieved from https://www.canada.ca/en/public-health/services/publications/healthy-living/national- report-apparent-opioid-related-deaths-released-march-2018.html

Statistics Canada. (2019, April 09). Government of Canada. Retrieved from https://www.canada.ca/en/health-canada/services/publications/healthy-living/canada- opioid-crisis-fact-sheet.html

Washington Post Staff. (2020, January 24). The Opioid Files: Follow The Post’s investigation of the opioid epidemic. Retrieved from https://www.washingtonpost.com/national/2019/07/20/opioid-files/?arc404=true

Drug testing using FTIR spectroscophy Data Analysis

Harm Reduction – Executive Summary

  • August 20, 2020October 29, 2022
  • by mikelabun

Since 2003, the AIDS Network Kootenay Outreach and Support Society (ANKORS), now partnered with Interior Health and the British Columbia Centre on Substance Use, has been offering free drug checks to attendees at music festivals like the Shambhala and Bass Coast held in British Columbia, Canada. Drug checking is a free harm reduction service that allows service users to bring drugs and have their compositions checked using technologies such as Fourier-transform infrared spectroscopy (FTIR). Service users are then shown the results, allowing them to make informed decisions regarding the drug’s use. To understand the impact of such services, ANKORS created survey sheets for service users, technicians, and advisors to record information during their transactions. Data has been collected regarding where drugs are obtained, what service users think they are, whether or not the FTIR results concord, and what the service user then determines to do with the drug.
Based on the studied dataset, some of the key findings of the data analysis are:
● It was discovered that the percentage of new service users accessing the drug checking services increased from 52% in 2018 to 70% in 2019.
● The two drugs most commonly brought by the same service user during the same visit at Shambhala are MDMA and LDS, and the most common drug additive is one of many possible forms of sugar.
● There is no statistically significant difference between what a service user believes a drug to be and the spectroscopy analysis.

Researchers also conducted unsupervised learning to look for patterns, specifically utilizing various clustering algorithms to draw inferences from the data. Additionally, various ensemble and boosting algorithms were conducted to predict whether or not a service user will discard their drug as well as to predict what service users will decide to do with a drug after getting the results of the drug check.
The results of the exploratory data analysis were clear enough, but the results of both the supervised and unsupervised learning approaches were inconclusive. Further research could involve experimenting more with variable selection, aggregating multiple datasets over many years, or asking more specific questions on the survey sheets, and suggestions are made for data collection that would facilitate this future research. An appendix is included that explains, to the general reader, the supervised and unsupervised techniques that were utilized.

To obtain a copy of the full report, email mikelabun@gmail.com.

See the code

Read the report:

Drug-Checking-at-Music-Festivals-in-British-Columbia