Prescribing of opioid analgesics and related mortality before and after the introduction of long-acting oxycodone

In medical school I was given the textbook "Managing Pain: The Canadian Healthcare Professionals Guide".(1) The first chapter decries "opioiphobia", an irrational fear of prescribing opioids. The book recommends prescribing oxycodone over morphine and endorses continuous release opioids because of a “lower abuse potential”. The production costs of the manual, owned by many family physicians I know, were paid for by the manufacturer of OxyContin.

Even though opioid misuse is a very complicated issue, some simple changes may be helpful.

Eighty milligram OxyContin pills remain on public formulary in Ontario despite the fact that this dose is not supported by current pain management guidelines.(2) A few recent studies make me question the value of this coverage to tax payers. Detoxifications from OxyContin at the Centre for Addiction and Mental Health increased sharply after 1999 (3) and OxyContin is now a drug of choice on the streets of Toronto.(4) The study by Dhalla et al suggests that the rate of oxycodone related deaths per prescription doubled after OxyContin was added to the public formulary.

The Committee to Evaluate Drugs recently recommended that 5 mg OxyContin pills not be added to the formulary in Ontario, noting the excessive cost – twice that of morphine – and that “oxycodone has not been demonstrated to be therapeutically superior to morphine” (5) Similar reasoning might lead them to remove OxyContin from the public formulary altogether. The tens of millions of dollars saved could be spent on safer pain management modalities.(6) We would also see if prescription and death rates fall as quickly as they rose when it was added.

The unpleasantness of denying patients opioids sometimes makes it difficult for me to remember that doing so is safe and often appropriate. Indeed, while there is virtually no evidence that long term opioids are helpful in non-cancer pain,(7) the study by Dhalla et al shows that writing these prescriptions can be deadly.

I will think of this every time I look at the new gap on my bookshelf.

References

1. Jovey R (ed.). Managing Pain: The Canadian Health Professional’s Reference. Toronto: Healthcare & Financial Publishing; 2002.

2. Chou R, Fanciullo GJ, Fine PG, Adler JA, Ballantyne JC, Davies P, Donovan MI, Fishbain DA, Foley KM, Fudin J, Gilson AM, Kelter A, Mauskop A, O'Connor PG, Passik SD, Pasternak GW, Portenoy RK, Rich BA, Roberts RG, Todd KH, Miaskowski C; American Pain Society-American Academy of Pain Medicine Opioids Guidelines Panel. Clinical guidelines for the use of chronic opioid therapy in chronic noncancer pain. J Pain. 2009 Feb;10(2):113-30.

3. Sproule B, Brands B, Li S, Catz-Biro L. Changing patterns in opioid addiction: characterizing users of oxycodone and other opioids. Can Fam Physician. 2009 Jan;55(1):68-9, 69.e1-5 4. Firestone M, Fischer B. A qualitative exploration of prescription opioid injection among street-based drug users in Toronto: behaviours, preferences and drug availability. Harm Reduct J. 2008 Oct 17;5:30

5. Committee to Evaluate Drugs, Onatrio. Oxycodone. July 2007. http://health.gov.on.ca/english/providers/program/drugs/ced/ pdf/oxycodone.pdf (accessed 13 December 2009).

6. Sapsford R, Delivering world class value for money in provincial drug system, July 2009. http://www.health.gov.on.ca/english/providers/program/drugs/ resources/drug_system_renewal_forum.pdf (accessed 22 December 2009).

7. Chou R, Ballantyne JC, Fanciullo GJ, Fine PG, Miaskowski C. Research gaps on use of opioids for chronic noncancer pain: findings from a review of the evidence for an American Pain Society and American Academy of Pain Medicine clinical practice guideline. J Pain. 2009 Feb;10(2):147-59

Conflict of Interest:

None declared

The analysis and conclusions by Dhalla et al and some of the comments in the editorial by Fischer and Rehm leave me with more questions than answers. If I understand correctly, the authors utilized at least 3 different databases, from 3 different time periods and somehow created a linkage resulting in some inferred conclusions on mortality causation.

My specific questions / observation for the authors are as follows:

1. Please confirm exactly how many different databases were used for this study and what exactly each of these databases measures in exactly what population of people over exactly what time period? Most importantly, explain in simple language how these different sources were linked and the validity and limitations of this exercise.

2. There was no information provided about expertise on the part of the authors in interpreting coroner's data. The fact that the authors agreed completely on 40 "test" files out of 7000 is not that reassuring and introduces a potential element of bias into the analysis. The clinical data surrounding a suspicious death is all important in deciding a cause of death since the methodology and interpretation of post-mortem toxicology is controversial. For example, there can be a large degree of overlap between the serum levels of an opioid naive person dying of an overdose and a patient on therapeutic opioids who is walking and talking. (Wallage 2006, Thompson 2008). The presence of clonazepam or alprazolam, two very commonly prescribed benzodiazepines, can be totally missed depending on which detection technology and which cut off level is utilized. (Wolf 2005) We are not given information on the type of toxicology performed and whether or not this changed over the years of the analysis. The definition of "unintentional opioid-related death" is not specified and can have a large impact (up to a two-fold difference) on reported death rates. (Jauncey 2005)

3. Did the authors make any attempt to further analyze "unintentional" deaths involving prescription opioids to assess whether the death was due to a therapeutic misadventure or due to intentional misuse of the medication? The authors note that 66% of decedents visited a physician in the 4 weeks prior to death and that 56% of decedents filled a prescription for an opioid in the month prior to death and suggest this could be an opportunity for prevention. What about the denominator? How many visits do people with chronic pain who do not die make to their physicians in a given month? Since it is likely only a very small fraction of those visiting a doctor who ended up dying, what tools can physicians utilize to recognize this small number in order to develop effective interventions. In a similar analysis of unintentional pharmaceutical overdose deaths in West Virginia in 2006, with better data linkage, Hall et al determined that indicators of substance abuse were present in 94% of decedents. (Hall 2006) If a patient purposely sets out to deceive a physician to obtain pain medication how easy is it to detect them and intervene? These distinctions are very important if one is going to propose appropriate solutions to the problem.

4. Putting some type of regulatory restrictions on the prescribing of opioids for pain will not materially affect those who choose to misuse these substances. In over 27 years of looking after the needs of patients suffering from substance use disorders, I have always been impressed by the ingenuity demonstrated by these individuals in finding what they want or need. Regulatory restrictions will mostly penalize people with pain who use opioids responsibly.

5. Do any of the authors or editorialists or did any of the reviewers of this paper actually have clinical experience in treating pain in order to put a clinical context to this data? The authors reported that between 2001 and 2007, the average dosage for CR-oxycodone prescriptions increased by 24% from 1830-2280mg per prescription. We have no information about the duration of these prescriptions. If one assumes that the average prescription was written to last 4 weeks, this translates to an average daily dosage increase of oxycodone from 65mg/day to 81mg/day. As a pain clinician, my comment would be that 80 mg per day of oxycodone is still a relatively low dosage for someone with severe pain. It would have been far more helpful to present this data in a clinical context by comparing the overall mortality related to opioids versus that from acetaminophen, NSAID and COXIB use.

6. It is interesting that the authors chose to report mortality rates in deaths per million population. All of the Canadian databases that I have searched seem to choose deaths per 100,000 population as a standard reporting method. Could this be because the numbers per 100,000 would have seemed too small to be newsworthy? One could translate Dhalla's reported death rates of 13.7 per million and 27.2 per million to a risk of death for the population which changed from 1 in 73,000 to 1 in 36,000. Whether one uses the EU descriptive scale, Calman verbal scale or the Paling perspective scale, this level of risk would be classed as "very rare" or "very low" and even doubling this risk does not change this interpretation. I would further suggest that if an individual chooses to use prescription opioids appropriately rather than misuse them, then the risk of death likely drops much further to "negligible" levels.

7. The unspoken assumption from this paper is that the pathway for illicit use of prescription opioids is a direct one from the doctor's prescription pad to the street. The "leakage" of opioids to the street may take many routes. Based on both Canadian and U.S. data, the most common pathway is via a friend's or family's medicine cabinet. (Paglia- Boak 2009, SAMSHA 2008) Other pathways can include wholesale and pharmacy thefts or diversion, cross-border smuggling, counterfeit pill manufacturing, internet purchases and others. (Forman 2006, Inciardi 2007) We really do not have good data on the relative contribution of these other factors to the availability of opioids on the street. I agree with Fischer and Rehm that leakage is likely proportional to the volume of prescriptions, but potency also plays a role. Dasgupta et al demonstrated that when one factors in the equianalgesic potency of various prescribed opioids, then the rate of related morbidity is actually very similar for all opioid molecules. (Dasgupta 2006) Since Canadians can buy codeine over the counter and since this is the most prescribed opioid in Canada, and since this was associated with the second largest number of single opioid deaths in Dhalla's analysis, any solution to the public health problem caused by the availability of opioids has to involve codeine. In some triplicate provinces, codeine is not even included as a monitored substance!

8. In North America we have a long history of repeated failed attempts to use supply side restrictions and law enforcement strategies in the "War on Drugs." Perhaps the solution needs to focus more on demand reduction by providing better access to addiction services. For example, opioid agonist therapy (OAT) in Canada has largely been restricted to the provision of structured methadone programs. Although these programs are a proven evidence-based approach to harm reduction in opioid addicted people, there is still a growing wait list and a great deal of stigma among the general population in attending community "meth clinics." If Fisher and Rehm's suggestion is true that the abuse of prescription opioids is moving into middle class society, this stigma, and the associated reluctance, will be even greater. One solution to this would be for provinces to fund the availability of buprenorphine as a substitute in order to move OAT into the GP's office. The experience in France demonstrated that this strategy can save many lives. (Fatseas 2007)

9. Dhalla et al note that the prescribing of oxycodone increased by 850%, such that by 2006, oxycodone accounted for 32% of opioid prescriptions. Their speculation that this may have been due to pharmaceutical marketing hints at some personal biases among the authors. If one were to ask a Canadian pain clinician, other equally plausible reasons for the increase may include better oral absorption, less first- pass effect, less drowsiness, and a slightly different opioid receptor profile that has more published evidence in neuropathic pain than morphine. In short, perhaps the growing number of prescriptions has something to do with better effectiveness. In addition IR oxycodone with acetaminophen was already a very commonly prescribed opioid prior to the availability of CR-oxycodone, so as physicians were taught to use CR opioids for chronic pain it seemed a more natural transition to switch to the CR equivalent rather than go to morphine.

10. Another potential reason for the growing popularity of oxycodone on the street may be the increased attention to prescription opioid misuse in the media. Dasgupta, compared the frequency of reports on oxycodone misuse in the U.S. media with the prevalence of reports of oxycodone- related opioid deaths. Their analysis suggests that an increase in media reports predicted an increase in opioid deaths by 2-6 months, accounting for 88% of the variation in mortality. (Dasgupta 2009) Could it be that sensationalistic media reports, rather than dissuading people actually increase the desire of certain people to misuse certain drugs? I can remember watching a CBC broadcast some years ago demonstrating how to crush and abuse CR-oxycodone on national television!

11. Dhalla chose to focus on the fact that by 2006, 28% of all opioid prescriptions in Ontario were for controlled-release (CR) oxycodone, rather than focus on the fact that 72% of scripts were for (IR) immediate release oxycodone. If one looks at the OPICAN study (Fischer 2006) , the abuse of IR oxycodone and codeine products were much more commonly reported from the Toronto area than CR oxycodone. Unless a decedent is found clutching a bottle of CR-oxycodone or a number of CR-oxycodone pills are found in the stomach contents or nearby, I would suggest that it is very difficult to tell which oxycodone formulation contributed to the person's demise.

12. From 1991-2004 Dhalla reported 3406 deaths where opioids were implicated. Of these deaths, 62% were deemed to be single opioid-related deaths. The molecules involved were: morphine or heroin 761 (36%), codeine 431(20%), methadone 304 (14%) and oxycodone 234 (11%). Since CR- oxycodone only became available in Canada in 1996, it would have been more meaningful to present this data from the time CR-oxycodone became available. By eyeballing the graphs in Figure 2 of the paper, it looks like total opioid-related deaths went from approximately 19 per million to 27 per million, an apparent increase of 42%. When one looks at the lower graph in Figure 2, the number of oxycodone-related deaths went from 2 per million to 8 per million in 2003. Yes, this is a 4-fold increase, but both are still very small numbers.

13. Dhalla's report noted that the majority of oxycodone-related deaths involved at least 1 other CNS drug - but did not comment further on what impact these other drugs may have had on the actual assigned cause of death. In the case of oxycodone-related deaths, 91% were found to have other substances in their body: benzodiazepines (59%), alcohol (43%), and tricyclics (26%). The coroner's data listed suicide as a cause of death in 23% of all opioid-related deaths and "undetermined" in 21 %. Combining this knowledge with the 26% of oxycodone-related deaths that had a tricyclic on board invites the question: How many of these undetermined deaths may have been due to suicide? Tricyclics are rarely prescribed today for depression but are more commonly prescribed for chronic pain management. The suicide rate in people with pain is known to be twice that of the general population. (Tang 2006) Putting some type of restriction on any given opioid molecule would not be expected to change this suicide rate much. Providing other evidence-based treatment options might.

14. In their editorial, Fischer and Rehm comment that the United Kingdom is a less "opioid rich environment" due to less prescribing of opioids for pain. It is interesting to note that the largest cause of opioid-related death in the U.K. in 2004 was heroin/morphine at a very similar rate (25-30 per million) as the total opioid related deaths noted by Dhalla in 2004 in Ontario. (Office for National Statistics 2007). It suggests that people who misuse opioids will utilize whatever is available.

Summary

There are other examples where the use and consequences of psychoactive substance misuse increases with the increased availability of a drug. Over time, heath care professionals, the public and, most importantly, those who choose to misuse drugs become more educated, the risks become better known and the consequences start to fall. The introduction and increased availability of methadone for opioid maintenance treatment in many countries is a good example.

If one looks at the street popularity of various drugs of abuse over many decades, one sees a waxing and waning in the reported use of cocaine and cannabis. We have just come through a period of increased popularity of prescription medications generally and opioid analgesics specifically. If one looks at recent survey data from both the United States and Canada, the percentage of the overall population who admitted misusing a prescription opioid in the past month, has been level for the past 5 years and actually decreased in the 12-17 year old age group. (SAMSHA 2009, Paglia-Boak 2009 )

Those who choose to misuse drugs of any type make their own assessment of the risks vs. benefits of a particular drug and mode of delivery. A common perception, especially among young people, is that prescription drugs are safer than street drugs of unknown origin. There is some "street logic" to this. Fischer et al have even suggested the "unthinkable": that the trend to increasing misuse of prescription opioids may have some net social benefit by reducing the abuse of heroin injection and therefore constituting a "quasi-medical opioid substitution" program. (Fischer 2009).

As a pain clinician I would like nothing better than to reduce the harms associated with prescription opioid misuse so that I do not have to play the role of policeman as well as physician. Unfortunately, in Ontario, the evidence-based options for treating chronic pain are largely limited to pharmacotherapy. Even then it is a battle to obtain public coverage for some evidence-based non-opioid pain treatments. It is embarrassing that in the largest province in the country, there is no public access to even one fully funded interdisciplinary pain management program. In an attempt to relieve suffering in their patients, clinicians turn to what is available. The fact is that opioids are an evidence-based treatment for chronic pain and work very well for some people for some types of pain for some period of time. Taken by the wrong person for the wrong reason, opioids can do harm - up to and including death. Ontario is one of the few provinces that does not have a controlled substance prescription electronic database available so that a physician can quickly find out what other prescriptions from what other doctor the patient has received recently.

The problem of misuse of pharmaceuticals is a multidimensional problem which requires multi dimensional solutions. If we do not ask the right questions and collect the right data, we will not really understand the problem and will therefore not develop the right solutions. We need less heat and more light on this issue.

There is risk in every treatment we provide and in chronic pain there is risk in not treating patients. The art of medicine is learning to balance the risks and benefits in each patient we see. This requires ongoing education, the proper tools to do the job and thoughtful public policy. Any measures that attempt to reduce the risk of harm to the small percentage of people who choose to misuse these valuable medications should not be at the cost of punishing the vast majority of people with pain who use these medications responsibly.

Roman D. Jovey, M.D. Past-President, Canadian Pain Society

CPM Centres for Pain Management

References: Dasgupta N, Kramer ED, Zalman MA, Carino S Jr, Smith MY, Haddox JD, Wright C 4th. Association between non-medical and prescriptive usage of opioids. Drug Alcohol Depend. 2006 Apr 28;82(2):135-42. Dasgupta N, Mandl KD, Brownstein JS. Breaking the News or Fueling the Epidemic? Temporal Association between News Media Report Volume and Opioid-Related Mortality. PLoS One 2009; 4(11): e7758 p1-7.

Fatseas M, Auriacombe M. Why buprenorphine is so successful in treating opiate addiction in France. Curr Psychiatry Rep. 2007 Oct;9(5):358-64.

Fischer B, Gittens J, Kendall P, Rehm J. Thinking the unthinkable: Could the increasing misuse of prescription opioids among street drug users offer benefits for public health? Public Health 123 (2009) 145–146

Fischer B, Rehm J, Patra J, Cruz M. Changes in illicit opioid use across Canada. CMAJ 2006; 175(11):1385-87

Forman RF, Woody GE, McLellan T, Lynch KG. The availability of Web sites offering to sell opioid medications without prescriptions. Am J Psychiatry. 2006; 163:1233-1238.

Hall AJ, Logan JE, Toblin RL, et al. Patterns of Abuse Among Unintentional Pharmaceutical Overdose Fatalities JAMA. 2008;300(22):2613- 2620

Inciardi JA, Surratt HL, Kurtz SP, Cicero TJ. Mechanisms of prescription drug diversion among drug-involved club- and street-based populations. Pain Med. 2007 Mar;8(2):171-83. Jauncey ME, Taylor LK, Degenhardt LJ.The definition of opioid-related deaths in Australia: implications for surveillance and policy. Drug Alcohol Rev. 2005 Sep;24(5):401-9.

Office for National Statistics. Health Statistics Quarterly. Spring 2007, No. 33. p88. Available from: http://www.statistics.gov.uk/downloads/theme_health/hsq33web.pdf

Paglia-Boak, A., Mann, R.E., Adlaf, E.M., & Rehm, J. (2009). Drug use among Ontario students, 1977-2009: Detailed OSDUHS fi ndings. (CAMH Research Document Series No. 27). Toronto, ON: Centre for Addiction and Mental Health Substance Abuse and Mental Health Services Administration. (2009). Results from the 2008 National Survey on Drug Use and Health: National Findings (Office of Applied Studies, NSDUH Series H-36, HHS Publication No. SMA 09- 4434). Rockville, MD. Tang NK, Crane C. Suicidality in chronic pain: a review of the prevalence, risk factors and psychological links. Psychol Med. 2006 May;36(5):575-86.

Thompson JG, Vanderwerf S, Seningen J, Carr M, Kloss J, Apple FS. Free oxycodone concentrations in 67 postmortem cases from the Hennepin County medical examiner's office. J Anal Toxicol. 2008 Oct;32(8):673-9.

Wallage HR, Palmentier JP.Hydromorphone-related fatalities in Ontario. J Anal Toxicol. 2006 Apr;30(3):202-9.

Wolf BC, Lavezzi WA, Sullivan LM, Flannagan LM. One hundred seventy two deaths involving the use of oxycodone in Palm Beach County. J Forensic Sci. 2005 Jan;50(1):192-5.

Disclosures: Dr. Jovey has consulted for or been a member of the speakers’ bureau for: AstraZeneca, Bayer Canada, Biovail Canada, Boehringer-Ingelhiem, Lilly Canada, Janssen-Ortho, Glaxo-Smith-Kline, King Pharmaceuticals, Merck-Frosst Canada, Nycomed, Pfizer, Palladin, Purdue Pharma, Sanofi-Aventis, Valeant, Wyeth

Conflict of Interest:

None declared

The oxycontin problem is significant, especially in small northern communities, where a little bit of addiction goes a long in disabling a whole community. I read Dr Dhalla's article with keen interest. My impression is that most patients get their narcotics from illegal sources. This is reinforced by stories of authorities confiscating large containers with false bottoms used for importing bulk amounts of 'oxy's' into communities. Clinicians in our region are aware of the devastation opiod addiction can cause and are careful with prescribing. The authors note that many fatal overdoses have seen an md for a perscription in a preceeding time frame. What is not clear is whether that is their main source of narcotics. Do the authors have any information on non-prescribed narcotic abuse/access? Is the supply problem from individual prescriptions being misused or from access to illicit bulk narcotic drug trade. Do we know how much of the narcotic problem is a prescribing problem?

Conflict of Interest:

None declared

I would like to thank the editors and the authors of these articles for placing the problem of chronic pain, opioid prescribing, and addiction on the front cover of CMAJ.

The authors of both articles have pointed out numerous factors that have contributed to the increase in opioid prescriptions and deaths in Canada. Blaming the “drug” is not the answer and Dr’s Fischer and Rehm have eloquently made that clear.

Availability is what makes oxycodone the current leader of the pack. The most dangerous drug with respect to deaths (out of proportion to its availability) is methadone, due to its long half life and interactions with so many other substances, although methadone is considered very effective and safe if used as recommended.1

Other reasons for our opioid-overloaded culture include the following:

1. The majority of chronic pain patients in Canada are managed by their primary care physicians who are very poorly trained in pain treatment. 2,3

2. Chronic pain patients require more time per visit, for which doctors are not generally compensated. The patients suffer from more psychosocial distress, and many g.p.’s consider them difficult to manage. 4,5,6

3. The provincial insurance program (at least in Ontario) does not cover any of the non drug-based programs for dealing with chronic pain, including physiotherapy, psychological counseling, occupational therapy or group self management and exercise programs.

4. Opiates are covered on the Ontario Drug Benefits program, while excellent alternatives to opiates for neuropathic pain such as pregabalin, gabapentin, or topical lidocaine, are not.

5. The easiest action for the stressed and time-pressured family doctor is to reach for the prescription pad.

We know that application of universal precautions to opiate prescribing, with narcotic contracts, urine drug testing, and tight boundaries for patients exhibiting aberrant or ambiguous behaviours will reduce these behaviours and potentially also diversion.7,8

One can hope that some of the new tamper-resistant pharmaceutical agents that are coming our way will improve our track record.

However informed prescribers remain the linchpin of any intervention. Jurisdictions in the U.S. require that all physicians take an opiate prescribing course on a regular basis to maintain their licenses. Education would go a long way to preventing the tragedies that Dr’s Dhalla et al described.

Sincerely,

Ruth Dubin MD,PhD,FCFP Kingston Family Health Team

REFERENCES

1. SAMSHA 2009 www.kap.samhsa.gov/

2. Jovey et al, Canadian Pain Society Nanos Survey 2007-2008

3. Watt-Watson et al, 2008. Canadian Pain Society

4. Tripp et al, Pain. Res. Man. 2006

5. Ponte et al, Fam. Med. 2005

6. Goubert et al, Pain, 2005

7. Gourlay, Pain Med. 2005

8. Manchikanti et al, Pain Med. 2006

Conflict of Interest:

None declared