Sparing opioid prescription to opioid naïve patients

02 Jun 2021 Sparing opioid prescription to opioid naïve patients

Register Download
Share
Posted at 23:41h in Portrait by

An individual Portrait on prescribing opioids to opioid naïve patients was sent to approximately 5200 family doctors in BC, randomized to receive the Portrait in two groups. The Portraits were mailed to the early group in November 2020 and to the delayed group in June 2021. This Portrait was produced by the Therapeutics Initiative in collaboration with the British Columbia Centre on Substance Use (BCCSU).

Below is a sample Portrait containing fictional individual physician data. View this sample Portrait in PDF format, or click on the DOWNLOAD button above. If you are a BC family physician and wish to sign up for (or opt out of) receiving Portrait, click on the REGISTER button above.

Abbreviations Used

BC: British Columbia
BCCSU: British Columbia Centre on Substance Use
CCI: Canadian Classification of Health Interventions
COPD: chronic obstruction pulmonary disease
CPSBC: College of Physicians and Surgeons of British Columbia
ED: emergency department
EMR: electronic medical record
FP: family practitioner
GP: general practice
ICD: International Classification of Diseases
MED: morphine equivalent daily
MSP: Medical Services Plan
NSAID: non-steroidal anti-inflammatory drug
OAT: opioid antagonist therapy
PSP: Practice Support Program
RCMP: Royal Canadian Mounted Police
RCT: randomized clinical trial

Sample Portrait

References and additional information

References

  1. Chang AK, Bijur PE, Esses D, et al. Effect of a single dose of oral opioid and nonopioid analgesics on acute extremity pain in the emergency department: a randomized clinical trial. JAMA. 2017 Nov 7;318(17):1661-7. doi:10.1001/jama.2017.16190
  2. Busse JW, Wang L, Kamaleldin M, et al. Opioids for chronic noncancer pain: a systematic review and meta-analysis. JAMA. 2018;320(23):2248-60. doi:10.1001/jama.2018.18472
  3. Krebs EE, Gravely A, Nugent S, et al. Effect of opioid vs nonopioid medications on pain-related function in patients with chronic back pain or hip or knee osteoarthritis pain: the SPACE randomized clinical trial. JAMA. 2018 Mar 6;319(9):872-82. doi:10.1001/jama.2018.0899
  4. Daoust R, Paquet J, Cournoyer A, et al. Side effects from opioids used for acute pain after emergency department discharge. Am J Emerg Med. 2020 Apr 1;38(4):695-701. Epub 2019 Jun 3. doi:10.1016/j.ajem.2019.06.001
  5. Fischer B, Jones W, Rehm J. High correlations between levels of consumption and mortality related to strong prescription opioid analgesics in British Columbia and Ontario, 2005-2009. Pharmacoepidemiol Drug Saf. 2013 Apr 1;22(14):438–42. doi:10.1002/pds.3404
  6. Chou R, Turner JA, Devine EB, et al. The effectiveness and risks of long-term opioid therapy for chronic pain: a systematic review for a National Institutes of Health Pathways to Prevention Workshop. Ann Intern Med. 2015 Feb 17;162(4):276–86. doi:10.7326/M14-2559
  7. Nielsen S, Van Hout MC. Over-the-counter codeine—from therapeutic use to dependence, and the grey areas in between. Curr Top Behav Neurosci. 2016;34:59–75. doi:10.1007/7854_2015_422
  8. Das M, Jain R, Dhawan A, et al. Assessment of abuse liability of tramadol among experienced drug users: double-blind crossover randomized controlled trial. J Opioid Manag. 2016 Nov;12(6):421–30. doi:10.5055/jom.2016.0361
  9. Adams EH, Breiner S, Cicero TJ, et al. A comparison of the abuse liability of tramadol, NSAIDs, and hydrocodone in patients with chronic pain. J Pain Symptom Manage. 2006 May;31(5):465–76. doi:10.1016/j.jpainsymman.2005.10.006
  10. Smolina K, Crabtree A, Chong M, et al. Patterns and history of prescription drug use among opioid-related drug overdose cases in British Columbia, Canada, 2015-2016. Drug Alcohol Depend. 2018 Jan 1;194:151-8. doi:10.1016/j.drugalcdep.2018.09.019
  11. Compton WM, Jones CM, Baldwin GT. Relationship between nonmedical prescription-opioid use and heroin use. N Engl J Med. 2016 Jan 14;374:154–63. doi:10.1056/NEJMra1508490
  12. Ray WA, Chung CP, Murray KT, et al. Prescription of long-acting opioids and mortality in patients with chronic noncancer pain. JAMA. 2016 Jun 14;315(22):2415–23. doi:10.1001/jama.2016.7789
  13. Klimas J, Gorfinkel L, Fairbairn N, et al. Strategies to identify patient risks of prescription opioid addiction when initiating opioids for pain: a systematic review. JAMA Netw Open. 2019 May 3;2(5):e193365. doi:10.1001/jamanetworkopen.2019.3365
  14. Chenot JF, Becker A, Leonhardt C, et al. Use of complementary alternative medicine for low back pain consulting in general practice: a cohort study. BMC Complement Altern Med. 2007 Dec 18;7(1):42. doi:10.1186/1472-6882-7-42
  15. Tick H, Nielsen A, Pelletier KR, et al. Evidence-based nonpharmacologic strategies for comprehensive pain care: the consortium pain task force white paper. Explore (NY). 2018 May 1;14(3):177–211. doi:10.1016/j.explore.2018.02.001
  16. Choosing Wisely Canada. Opioids –When you need them and when you don’t [Internet]. Toronto (ON): Choosing Wisely Canada [cited 2020 Nov 24]. Available from: https://choosingwiselycanada.org/patient-pamphlet-opioids/
  17. Institute for Safe Medication Practices Canada. Opioids for short-term pain [Internet]. Toronto (ON): Institute for Safe Medication Practices Canada [cited 2020 Nov 24]. Available from: https://www.ismp-canada.org/download/OpioidStewardship/Opioids-ShortTermPain-EN.pdf
  18. Wood E, Simel DL, Klimas J. Pain management with opioids in 2019-2020. JAMA. 2019 Oct 10;322(19):1912-13. doi:10.1001/jama.2019.15802
  19. Howard R, Fry B, Gunaseelan V, et al. Association of opioid prescribing with opioid consumption after surgery in Michigan. JAMA Surg. 2019;154(1):e184234. doi:10.1001/jamasurg.2018.4234
  20. Hill MV, McMahon ML, Stucke RS, et al. Wide variation and excessive dosage of opioid prescriptions for common general surgical procedures. Ann Surg. 2017 Apr 1;265(4):709–14. doi:10.1097/SLA.0000000000001993
  21. Hill MV, Stucke RS, McMahon ML, et al. An educational intervention decreases opioid prescribing after general surgical operations. Ann Surg. 2018 Mar 1;267(3):468-72. doi:10.1097/SLA.0000000000002198
  22. Nielsen S, Degenhardt L, Hoban B, et al. A synthesis of oral morphine equivalents (OME) for opioid utilisation studies. Pharmacoepidemiol Drug Saf. 2016 Jun 1;25(6):733–7. doi:10.1002/pds.3945

Resources

Opioids – When you need them and when you don’t. Choosing Wisely Canada. https://choosingwiselycanada.org/patient-pamphlet-opioids/

Opioids for short-term pain. Institute for Safe Medication Practices Canada. https://www.ismp-canada.org/download/OpioidStewardship/Opioids-ShortTermPain-EN.pdf

Data definitions

Who received this Portrait?

BC physicians meeting all of the following criteria received an individual opioid prescribing Portrait:

  • general practice (GP) physicians, including family practitioners (FP), who were registered by the BC Medical Services Plan (MSP) as a private practice, or physicians registered primarily as GP-emergency medicine and FP-emergency medicine physicians; and
  • had a valid mailing address in BC according to the College of Physicians and Surgeons of British Columbia’s public physician information; and
  • had ≥100 prescriptions (for any drug) filled at a community pharmacy in 2019 (early group) or 2020 (delay group) according to PharmaNet claims data; and
  • prescribed an oral opioid to at least one opioid naïve patient in 2017-2019 (early group) or 2018-2020 (delay group) according to PharmaNet claims data.

Physicians might have received a Portrait with masked or missing elements (i.e. no red bars are shown) because they met the above requirements but prescribed between 0 and 5 opioid prescriptions for that section of the Portrait. Portrait’s data access agreement requires the masking of data elements when <6 patients are included.

How were patients assigned to this Portrait?

Patients were included in a physician’s Portrait if they met all of the following criteria:

  • were continuously registered with the MSP during the Portrait period (early group: 2017-2019; delay group: 2018-2020) and the respective preceding six months (washout period); and
  • filled a prescription for an opioid at a community pharmacy with that physician’s prescribing number in PharmaNet during the Portrait period (early group: 2017-2019; delay group: 2018-2020); and
  • were opioid-naïve (i.e. no opioids dispensed in the previous six months per PharmaNet); and
  • were age ≥19 years at the time of opioid dispensation; and
  • were not on PharmaCare plan P (palliative) or plan B (long-term care), and
  • did not have a cancer diagnosis, chemotherapy, or radiotherapy recorded in MSP fee-for-service claims, the hospital discharge abstract database, or the National Ambulatory Care Reporting System database during the Portrait analytic period or the preceding six months. Patients with non-melanoma skin cancers (basal cell carcinoma or squamous cell carcinoma) were included in this Portrait.
    • Cancer diagnosis codes: (ICD = International Classification of Diseases) ICD-10 C00.xx-C43.xx, C45.xx-C97.xx or ICD-9 140.xx-172.xx, 174.xx-209.xx
    • Radiation and chemotherapy codes: (CCI = Canadian Classification of Health Interventions) CCI 1.xx.26.xx, 1.xx.26.xx; ICD-9-CM 99.25, V58.1, V66.2, V67.2; ICD-10 Z51.11; or MSP fee item 33581-33583

Prescription data for patients who are federally insured (e.g. Veterans, RCMP, Armed Forces and beneficiaries of Non-Insured Health Benefits) or First Nations Health Benefits – PharmaCare Plan W were not included.

What prescriptions were included in this Portrait?

Filling events of opioid prescriptions were extracted from PharmaNet claims data and included all prescriptions filled at a community pharmacy in BC with a physician’s prescribing number. Reversed prescription claims, out-of-province prescriptions, or drugs dispensed in hospital were not included.

All oral formulations of prescription opioids available in BC were included, including codeine, tramadol, hydromorphone, oxycodone, morphine, meperidine, tapentadol, pentazocine, and fentanyl. Opioid combination products (e.g. codeine-acetaminophen-caffeine) were included. Codeine formulations to control cough were not included. Methadone, buprenorphine, and Kadian© were not included as these were assumed to be for opioid antagonist therapy (OAT).

We included only opioid prescription to opioid-naïve patients, defined as patients without an opioid dispensed in the previous six months (washout period).

How were morphine equivalent daily (MED) dosages calculated?

We calculated opioid dose equivalents from the standard reference1 used for recent opioid utilization studies and the 2017 Canadian opioid guideline.2 We recognize that drug interactions, kidney and liver disease, and a patient’s genotype for opioid-metabolizing enzymes can affect opioid metabolism. Both tramadol and codeine are prodrugs that are converted to active metabolites before exerting their main actions on opioid receptors. Inter-individual differences in intestinal and hepatic CYP450 genes cause some people to metabolize tramadol or codeine poorly or extensively. This may impact analgesic or adverse effects significantly in individual patients.3

Converting tramadol to ‘equivalent’ doses of opioid agonists is difficult because its non-opioid pharmacological actions likely affect its effect on pain.4

Research component

The research objective is to determine the impact of the Portrait on physician prescribing of opioids. This impact will be evaluated at an aggregated level over the coming year by comparing pooled prescribing data from physicians in the early mailing group to pooled prescribing data from physicians in the delayed group. All prescribing data analyzed for this evaluation will not contain names, only encrypted patient and physician numbers. No physician or patient will be identified in any reports or publications. As a reminder, Portrait’s data access agreement requires the masking of data elements when <6 individuals (patients or physicians) are included. Ethics approval for this evaluation was obtained from the University of British Columbia Clinical Ethics Review Board (H20-00656). This evaluation is being led by Dr. Rita McCracken, University of British Columbia, Faculty of Medicine, Department of Family Practice.

If you have any questions or would like further information with respect to this evaluation, you may contact the Portrait team at (604) 822-4887 or email Portrait@ti.ubc.ca

FAQ

Can I request the names of my patients included in this Portrait?

No. Portrait’s data access agreement only permits access to encrypted patient identifiers. There is no way for us to identify individual patients in the data or to provide you with a list.

However, other resources may be available for you. Contact yourPractice Support Program (PSP) regional team to discuss how they might help you use your EMR to identify these patients. This article in the BC Medical Journal provides a list of regional initiatives that support opioid prescribing and pain management5.

I understand the need to limit opioid prescribing to minimize any addiction potential, but I work as an emergency department primary care provider. How does this teaching point apply to the emergency department context?

We did include physicians registered primarily as GP-emergency medicine and FP-emergency medicine physicians in our mailing list, hoping it would provide an opportunity for reflection in this setting as well. We anticipate there will be warranted variation in practice given the wide differences in patients, conditions, and contexts in BC, even within outpatient practices. The purpose of Portrait is not to instruct individual doctors on how they should practice, but to promote reflection on whether one is approaching “optimal prescribing” based on the best available evidence. The patterns presented to you are for your information only and it is important to interpret them in the context of your practice. Portraits are never intended to overrule individual clinical judgement, for which all prescribers of course enjoy professional freedom as well as responsibility.

Our Portrait strives to offer several teaching points, including the evidence that it can be hard to predict which opioid naïve patients will go on to develop opioid-use disorders and that shorter, lower-dose regimes are likely the safest choice to prescribe when opioids are selected. There are some prescribing patterns that may be more regional/cultural rather than based on the best available evidence. We have not gone into this concept in detail as we are providing only individual provider-level and province-wide aggregate feedback; however, it is intriguing to think about.

While it would be silly to suggest there is an absolute “correct” number of opioids that should or shouldn’t be prescribed, it still seems there is room for change for all of us in our prescribing patterns. We did not look specifically at regional or temporal variation, but some papers (the quality of which we have not appraised) suggest that there is variation in prescribing in emergency departments and some of the variation is probably not necessitated by patient factors:

  • In an emergency department setting, there is wide variation between physicians in terms of opioid prescriptions6
  • Canadian emergency physicians may be prescribing more opioids than patients generally seem to use7
  • Over time, opioid prescribing in the emergency department can be reduced. This change is more pronounced with musculoskeletal/general pain rather than for fractures and nephrolithiasis; however, even post-fracture opioid use decreased substantially [58% from previous]8

Not specific to the emergency department, but some variation that happens in medicine may be cultural. We recently published a Therapeutics Letter on tramadol9, which includes the observation that approximately 70% of post-operative patients in Canada are prescribed opioids, of which 19% are for tramadol or tramadol-acetaminophen. In contrast, only about 11% of Swedish post-operative patients receive any opioids. It is possible that Swedish patients are under-treated for pain, but also possible that in Canada we may have learned to prescribe opioids more often than absolutely required.10

I have a unique practice. I prescribe high dose opioids to many patients for therapeutic purposes for opioid use disorder, and initiate people on prescription opioids who have previously been using them illicitly and as such are not “opioid naïve.I don’t feel that my inclusion in your opioid prescribing review is appropriate either for your data or for my information.

One of the inherent features of any Portrait, whether for dispensed prescriptions, diagnostic tests, or clinical services, is that it cannot possibly account for individual features of everyone. Right now, we do not have the ability to exclude physicians with the College of Family Physicians of Canada (Addiction Medicine) or other focused practices, but are working towards being able to do this more readily. We did exclude patients on methadone and suboxone to try to decrease the number of patients with opioid use disorder captured in our Portrait; the reality of safe supply and practice changes during COVID-19 (including increased use of hydromorphone as safe supply) presented a challenge in our analysis.

For the opioid Portrait, we identified that the distribution of number of prescriptions is not normal (Gaussian), unlike a typical “bell graph”.  We identified a long right-tail of the distribution, meaning that there were a few physicians who prescribed opioids to a very large number of patients. However, we could not identify a clear threshold between most physicians and those outliers. Hence, we also provided data to physicians with high prescription volume. We did, however, present the median (rather than the mean) as a “middle” value for all BC Family Practitioners. This gave less weight to values (number of patients) that were clustered toward one end of their range and/or if there were a few extreme values. So, while your clinic or others may have unique features, the effect on the median is minimal.

We recognize that prescribing does not happen in isolation and that many factors contribute to individual prescribing decisions. In the future, we hope to provide Portraits to specialist physicians, nurse practitioners, and other prescribers (e.g. dentists).

What is the evidence behind defining opioid naïve as patients who have not received an opioid prescription in the previous six months?

Opioid tolerance takes >1 week to develop, so patients who are not already on opioids would be considered to not have physiological tolerance (unless they are using, without the prescriber’s direction, non-prescribed opioids, or drugs left over from a long-prior prescription). We selected a 6-month washout period based on a project done in Ontario by Health Quality Ontario. Other studies using administrative health data also use the same length of washout period to define “opioid naïve”.11,12

Is there any evidence that people for whom the need for exceptional pain control occurred more than six months in the past are likely to have acceptable relief with non-opiates during their next pain episode?

Even in people who need exceptional pain control, there remains risk of prolonged or disordered use. Since there is no reliable way to predict patients who are likely to develop opioid use disorder or who are likely to benefit from long-term opioid therapy, we must use aggregate rather than individual data to look at overall benefits and harms. While there were no long-term clinical trials published until recently, emerging information suggests that chronic use of opioids is probably not beneficial in patients with non-cancer pain.13 Such studies typically report mean effects, so they reflect randomized clinical trial (RCT) evidence of net benefits/harms. Of course, there may be exceptions amongst individual patients, and the RCTs may not have enrolled people already taking opioids chronically. Still, this evidence encourages us to use these drugs infrequently, carefully, in the lowest possible doses, for the shortest possible durations, and with careful monitoring.

We are not aware of RCT studies of people in whom opioids had been used transiently or episodically who were then randomized to either an opioid or an alternative pain treatment. We do have evidence that serious adverse outcomes correlate to dose14, which is one of the reasons why we provided an ‘average’ dose comparison for the information of portrait recipients.

Is there any evidence whether the population who can go six months between opiate prescriptions exhibit opiate harms to a degree greater than pain relief benefits? These might be people whose previous prescriptions were being used up very slowly and in whom renewals could have been of rather small amounts, or those whose renal function would make nonsteroidal anti-inflammatory drugs (NSAIDs) a poor choice. Perhaps opiates are being under-prescribed in some populations and significant pain is being under-treated.

As mentioned, patients may very slowly use a prescription dispensed >6 months prior, and therefore are not truly “opioid naïve”. If this is the case, it may be worth considering that they could control their pain without opioids. In people with significant renal impairment, congestive heart failure, or evidence of upper intestinal bleeding, we agree that NSAIDs and other drugs used for pain are often contra-indicated or at least potentially dangerous. We are not aware of studies comparing opioid vs. non-opioid pain treatments in any of these diagnostic groups, or in people otherwise at particular risk of hypoventilation (e.g. chronic obstructive pulmonary disease [COPD] or restrictive lung disease).

We agree that under-treatment of pain may remain an issue. People receiving palliative care or who have been diagnosed with cancer are specifically excluded from Portrait. It is not practical for us to identify in databases other patients who are not treated with an explicit palliative care paradigm. You are doubtless aware of the significant connection (and now established legal responsibility) between pharmaceutical company promotion of opioid drugs and their campaigns based on the premise that pain was widely under-treated. This is well recognized as an important contributor to the global opioid overdose crisis, and a reason why some countries have reassessed the use of pain scores as the “fifth vital sign”.

Overall, Portraits exist to provide anonymized individual feedback on prescribing patterns and to recommend a direction for prescribing given the best available evidence. They are never intended to overrule individual clinical judgement, for which all prescribers of course enjoy professional freedom as well as responsibility.

I have a unique/specialized practice and do not believe that my data should be compared to GPs. Shouldn’t my comparison cohort be other physicians with practices similar to mine?

Our ability to accurately include physicians depends on how each physician is listed in the BC MSP data. For example, to be included in the program, a physician must be defined as a General Practitioner with a license status of Private Practice in the MSP data; beyond that, it is difficult to accurately identify any individual’s specific work environment, as many physicians work in multiple locations. We try to keep Portraits relevant to physicians by having a minimum prescribing criterion for each Portrait topic. Physicians who prescribe no or very little of the drug in question typically will not receive a Portrait.

Creating accurate comparison cohorts is difficult for the same reasons. We often include the BC median as a point of interest when it is feasible, but we know that many physicians have unique or specialized practices and this comparator is not ideal. Hopefully, as we continue to build and refine the Portrait program, we will be able to provide comparisons that are more relevant to a variety of practices.

Portraits are really intended to be a tool for you to reflect upon and consider your own prescribing. Rest assured that we are not directly comparing your prescribing to any particular cohort. You are the only one who ever sees your own Portrait data, and you can decide for yourself whether you think your prescribing is in line with the evidence given your unique clinical setting. You will only receive future Portraits for which you have prescribed more than the minimum inclusion criteria, so you shouldn’t receive anything that is irrelevant to your practice.

I don’t see my question here. Where can I find more information?

Please check out our program FAQ page. We also welcome your feedback. Questions can be directed to our Portrait team by phone (604) 822-4887 or email Portrait@ti.ubc.ca.

FAQ References:

  1. Nielsen S, Degenhardt L, Hoban B, et al. A synthesis of oral morphine equivalents (OME) for opioid utilisation studies. Pharmacoepidemiol Drug Saf. 2016;25(6):733-737. doi:10.1002/pds.3945
  2. Busse JW, Craigie S, Juurlink DN, et al. Guideline for opioid therapy and chronic noncancer pain. CMAJ. 2017;189:E659-66. doi:10.1503/cmaj.170363
  3. Smith H. Opioid metabolism. Mayo Clin Proc. 2009;84(7):613-24. doi:10.4065/84.7.613
  4. Grond S, Sablotzki A. Clinical pharmacology of tramadol. Clin Pharmacokinet. 2004;43:879-923. doi:10.2165/00003088-200443130-00004
  5. Ross S. Pain management and opioid-prescribing tools and resources. BCMJ. 2017;59(3):186-7. https://bcmj.org/gpsc/pain-management-and-opioid-prescribing-tools-and-resources
  6. Barnett ML, Olenski AR, Jena AB. Opioid-prescribing patterns of emergency physicians and risk of long-term use. N Engl J Med. 2017;376:663-673. doi:10.1056/NEJMsa1610524
  7. Daoust R, Paquet J, Cournoyer A, et al. Quantity of opioids consumed following an emergency department visit for acute pain: a Canadian prospective cohort study. BMJ Open. 2018;8:e022649. doi:10.1136/bmjopen-2018-022649
  8. Smith BC, Vigotsky AD, Apkarian V. Temporal factors associated with opioid prescriptions for patients with pain conditions in an urban emergency department. JAMA Netw Open. 2020;3(3):e200802. doi:10.1001/jamanetworkopen.2020.0802
  9. Therapeutics Initiative. Tramadol: where do we go from here? Therapeutics Letter. 2021 (May-Jun);131:1-2. https://www.ti.ubc.ca/2021/07/14/131-tramadol/
  10. Ladha KS, Neuman MD, Broms G, et al. Opioid prescribing after surgery in the United States, Canada, and Sweden. JAMA Netw Open. 2019;2(9):e1910734. doi:10.1001/jamanetworkopen.2019.10734
  11. Jeffery MM, Hooten WM, Hess EP, et al. Opioid prescribing for opioid-naïve patients in emergency departments and other settings: characteristics of prescriptions and association with long-term use. Ann Emerg Med. 2018;71(3):326-336.e19. doi:10.1016/j.annemergmed.2017.08.042
  12. Goplen CM, Randall JR, Kang SH, et al. The influence of allowable refill gaps on detecting long-term opioid therapy: an analysis of population-based administrative dispensing data among patients with knee arthritis awaiting total knee arthroplasty. J Manag Care Spec Pharm. 2019;25(10):1064-1072. doi:10.18553/jmcp.2019.25.10.1064
  13. Krebbs EE, Gravely A, Nugent S, et al. Effect of opioid vs nonopioid medications on pain-related function in patients with chronic back pain or hip or knee osteoarthritis pain: the SPACE randomized clinical trial. JAMA. 2018;319(9):872-882. doi:10.1001/jama.2018.0899
  14. Chou R, Turner JA, Devine EB, et al. The effectiveness and risks of long-term opioid therapy for chronic pain: a systematic review for a National Institutes of Health Pathways to Prevention Workshop. Ann Intern Med. 2015;162(4):276-86. doi:10.7326/M14-2559
Tags:
No Comments

Post A Comment