This retrospective observational study presents a high-level overview of the comparative overdose mortality, morbidity and abuse of buprenorphine, fentanyl, hydromorphone, morphine, oxycodone, tapentadol and tramadol in Australia and Western Australia. This study was restricted to data sources which were accessible and allowed for differentiation between opioids. This has resulted in a mixed collection of study populations, time periods and geographical locations. WA specific data cannot be assumed to reflect the Australia-wide data. Nevertheless, this paper which combines Australian data on the comparative overdose mortality, morbidity and abuse of seven common prescription opioids is the largest comparison to date.
The comparative overdose mortality, morbidity and abuse of common prescription opioids shows a mixed picture in Australia. Overall, there was a statistically significant increase of 7% in overdose mortality each year for all analysed opioids combined. This result is similar to that reported by Roxburgh et al [6] at 6%, who analysed a smaller number of prescription opioids and for an earlier time period. Of the opioids, morphine recorded the highest absolute overdose mortality, highest comparative overdose mortality per 100,000 grams OME dispensed, and the fourth highest increase in overdose mortality each year, at 5%. Important to note is that mortalities attributed to morphine may be an overestimation (Section 5) and some may be due to heroin toxicity [25]. Oxycodone was the second highest overdose mortality, at 69% lower than morphine, with a yearly increase of 4%. This was followed by tramadol at 78% lower with a yearly increase of 6%, fentanyl at 84% lower with a yearly increase of 33%, buprenorphine 87% lower, hydromorphone 91% lower with 37% yearly increase and finally tapentadol with the lowest overdose mortality at 96% lower than morphine. Statistical significance was achieved for each of the above results. The large swings in overdose mortality per 100,000 grams OME for buprenorphine are due to one death in 2001 and three deaths in 2004 during which only small amounts of buprenorphine were sold (2001: 2414 grams per OME, 2004: 2058 grams per OME), approximately 0.002% of yearly sales in 2017.
The underlying reason for the difference in the rate of overdose mortality between common prescription opioids has not been thoroughly researched. A 2018 paper [14] proposes that the potency of the opioid accounts for 96% of the difference in serious adverse side effects. However, considering the variation in rate of overdose mortality per 100,000 grams OME dispensed, which corrects for potency, and the availability of the opioid in the community, it would follow that other causes of the difference exist. Assuming the overdose mortality and dispensation data is a true reflection of reality, a possible explanation could be the inherent difference in the pharmacokinetics of each opioid and their unique effect upon the individual in terms of risk of overdose mortality, morbidity and abuse potential. Illegally produced opioids, not captured in the dispensation data could also account for the difference in rates of overdose mortality. However, this is unlikely as the difference between the opioids are generally consistent over time. This points to an inherent difference in the pharmacokinetics rather than a difference in availability, which would be expected to result in a frequent change of overdose mortality which is not the case. Anecdotally, illegal Fentanyl, or riskier use, may feature in the increase in overdose mortality rate year on year from 2009, as has also been reported in the USA. These are areas which require further research.
The number of moderate to severe cases of opioid poisoning, as handled by WAPIC, 2007–2017, which serves as a measure of morbidity, differs markedly in ranking compared with overdose mortality. Tramadol was the highest, with a rate per 100,000 grams OME dispensed of 3.9, followed by buprenorphine, tapentadol, oxycodone, morphine, fentanyl and hydromorphone. A statistically significant decline in the rate of cases per 100,000 grams OME dispensed was detected for buprenorphine at 17%, on the other hand a significant 11% increase was detected for oxycodone. Similarly, oxycodone had a 34% greater change over time in moderate and severe poisonings than buprenorphine.
Data from the US records a lower risk of major medical effect or hospitalisation, with the exception of hydromorphone, than that of Western Australia [14]. The rate of major medical effect or hospitalisation per 100,000 grams OME dispensed in the US is as follows; hydromorphone (1.28, 7.67), oxycodone (0.28, 1.66), morphine (0.21, 1.01), tramadol (0.06, 0.32) and tapentadol (0.04, 0.26) [14]. This is compared with the average moderate to severe toxicity for Western Australia; tramadol (3.9), buprenorphine (3.5), tapentadol (3.3), oxycodone (2.6), morphine (2.0), hydromorphone (1.6) and fentanyl (1.3).
New treatment episodes commenced for opioid addiction in Western Australia between 2015 and 2017, a measure of abuse, was highest for morphine (418), followed by oxycodone (380) and lowest for hydromorphone and tapentadol (“nil or extremely low”). The average rate of commencement of a treatment episode, in Western Australia, per 100,000 grams OME dispensed was highest for morphine (160.1) followed by buprenorphine (122.6), oxycodone (60.2), fentanyl (26.7), tramadol (9.6) and lowest for tapentadol and hydromorphone. This rate remains consistent over the study period, with clear linear trends observed.
Calls received by WAIPC between 2007 and 2017, for WA calls only, where the intent of the opioid poisoning was classified as intentional abuse numbered 74 over an 11-year period. This data needs to be approached with a degree of scepticism given the small number of calls. The Western Australian data records a higher average rate of intentional abuse, than that of the US data [27] with the exception of tapentadol and hydromorphone. The US data, from October 2011 to June 2016 records the following number of calls to a poison information centre per 100,000 units dispensed, where the intent was recorded as abuse; hydromorphone (0.062) morphine (0.052), oxycodone (0.036), tapentadol (0.028) and tramadol (0.022). In comparison, for Western Australia the following average number of calls per 100,000g OME were calculated; fentanyl (0.8), morphine (0.7) oxycodone (0.5), buprenorphine (0.4), tramadol (0.3), hydromorphone (0.0) and tapentadol (0.0). The ranking is broadly similar, with the exception of hydromorphone which is the highest for the US, and equal lowest for Australia.