Uncategorized

Substance use history is associated with lower opioid use for emergency department pain management

a b s t r a c t

Introduction: In the current national opioid crisis, where 10% of the US population has or has had a substance use disorder (SUD), emergency department (ED) clinicians are challenged when treating pain in the ED and when prescribing pain medications to these patients on discharge as there is concern for contributing to the cycle of addiction. The objective of this study was to examine whether acute pain is treated differently in patients with and without current or past SUD by quantifying the amount of Opioid analgesia given in the ED and prescribed on discharge.

Methods: Retrospective cohort study of patients presenting to a 60,000-visit tertiary referral ED with acute frac- ture between January 1, 2016 and June 30, 2019. The primary exposure was indication of SUD (SUD+) versus those without SUD (SUD-). The primary outcome was receipt of opioids in the ED, and the secondary outcome was opioids prescribed at discharge.

Results: 117 matched pairs (n = 234) were included in the sample. Overall, 53.4% and 62.4% of patients received opioids in the ED or a prescription for opioids, respectively. Opioid receipt in the ED was lower among SUD+ pa- tients compared to SUD- patients (48.7% and 58.1%, respectively; aOR: 0.33; 95%CI: 0.14, 0.77). Similarly, receipt of a prescription for opioids was lower among SUD+ patients compared to SUD- patients (56.4% and 68.4%, re- spectively; aOR: 0.50; 95%CI: 0.26, 0.95).

Conclusions: Overall, ED clinicians gave opioids less frequently to SUD+ patients in the ED and on discharge from the ED compared to SUD- patients with acute pain secondary to acute fracture.

(C) 2021

  1. Introduction

In the setting of the national opioid crisis, where 10% of the US pop- ulation has or has had a substance use disorder (SUD) [1], clinicians face a challenge when prescribing pain medications to patients on discharge as there is concern for contributing to the cycle of addiction. Weekly, 17% of patients discharged from the Emergency Department (ED) across the country are discharged with Opioid pain medications [2]. Around 10% of these patients likely have at least one indicator that suggests po- tential risk for inappropriate use [3].

Multiple factors influence the amount of opioid prescribed to pa- tients. Patient race [4], time clinician has been in practice, patient age, presence of trauma, presence of chronic pain [5], clinician assessment of pain, perceived patient trustworthiness, and practice environment

[6] may all influence clinician’s prescribing practices. Larger quantities of initial Opioid prescriptions have been associated with prolonged

* Corresponding author at: 200 Hawkins Drive, Iowa City, IA 52242, United States of America.

E-mail address: [email protected] (J. Radke).

opioid use and increased likelihood for misuse [7,8]. It has been esti- mated that patients use only 35-45 mg Morphine equivalents (MME) of opioid analgesia in the 1-2 weeks following discharge [9,10] while clinicians tend to prescribe 2-3 times that amount [2]. Months later, 9% of these patients will still be using opioids, and 9% of those will be using them for a reason other than to treat pain [11]. State legislation has limited the amount of opioid analgesia that can be prescribed which has effectively decreased the amount prescribed to patients [12]. However, undertreating pain can also be problematic as opioid consumption plays a role in pain trajectory after ED discharge [13].

Patients with history of SUD may have different requirements to ad- equately treat pain than patients without a history of SUD. Though not statistically significant, patients with positive urine drug screens (UDS) for opioids, THC, cocaine and amphetamines consumed larger amounts of opioids than patients with negative UDS, while patients with UDS positive for benzodiazepines and alcohol intoxication were associated with reduced opioid consumption during hospitalization [14]. High opioid requirement in the hospital, history of Opioid misuse, and the distance the patient lives from a hospital [15] can influence cli- nician prescribing practices after an inpatient stay. However, it is

https://doi.org/10.1016/j.ajem.2021.08.005

0735-6757/(C) 2021

unclear how a history of SUD impacts prescribing practices on discharge from the ED as ED clinicians have less time with each patient and will not have follow up with them.

The objective of this study was to examine whether we treat acute pain differently in patients with and without current or past Substance use disorders by quantifying the amount of opioid analgesia received in the ED and the amount prescribed on discharge after acute fracture.

  1. Methods
    1. Study design, sample, and setting

This was a retrospective cohort study of patients seen at a 60,000- visit tertiary referral ED between January 1, 2016 and June 30, 2019. Ad- ministrative data were queried for a fracture diagnosis across any diag- nosis code fields based on the International Classification of Diseases, 10th edition (ICD-10) [Supplemental Content – Appendix A]. Eligible patients for the study were ages 18-65 yo, and exclusion criteria included those indicating chronic opioid medication use, chronic pain, fibromyalgia, prisoner status, or an active Cancer diagnosis. The study was approved by the local Institutional Review Board under waiver of informed con- sent, and this manuscript is reported in accordance with the STrength- ening the Reporting of OBservational studies in Epidemiology (STROBE) statement [16].

    1. Measurement of exposure and covariates

The primary exposure in this study was an indication of SUD as iden- tified from the problem list of the patient chart that would have been available at the time of the ED visit. The dates a SUD diagnostic code [Supplemental Content – Appendix A] was documented and rendered inactive were compared to the ED visit date. If no SUD inactive date was identified, the diagnosis was considered still active on the problem list. Patients who had a SUD diagnosis on their problem list at the time of the ED visit were identified as SUD+ and those without a diagnosis for SUD were considered SUD-. Patients who had a SUD+ exposure were matched with a SUD- patient by gender, age, and Insurance type.

Other covariates assessed from the electronic medical record in- cluded the mechanism of injury (e.g. fall, assault, motor vehicle colli- sion, etc.), pain level as self-reported by the patient, a subjective measure of perceived pain by the clinician, intoxication, psychiatric co- morbidities (anxiety/generalized anxiety disorder, bipolar, depression, schizophrenia, etc.), chronic medical conditions (e.g. cancer, cardiac, gastrointestinal conditions, obesity, neurologic conditions), other con- ditions such as fall risk, and documented allergic reaction to opioids. Prescriber characteristics were evaluated by the level (resident, attend- ing, advance practice clinician), primary prescriber type (primary clini- cian, handoff, none), and consultants involved in the care.

    1. Measurement of outcomes

The primary outcome was any opioid received in the ED. The sec- ondary outcome was whether an opioid was prescribed by the clinician at the time of the visit. Opioids assessed included codeine, fentanyl, hydrocodone, hydromorphone, morphine, oxycodone, and tramadol. Non-opioid medications (e.g. acetaminophen, ketamine, Non-steroidal anti-inflammatory drugs, Muscle relaxants, anxiolytics) administered in the ED were tabulated and evaluated descriptively.

    1. Statistical analysis

We analyzed covariates (e.g. comorbidities, injury characteristics) obtained from medical chart review for the association with SUD and each outcome through conditional logistic regression of matched pairs. The total amount of opioids administered in the ED and total amount prescribed medications were converted to milligram morphine

equivalents (MME) [17]. Differences in the amounts of opioids given and prescribed between SUD+ and SUD- pairs were compared using the Wilcoxon signed rank sum test.

For each outcome assessed, we built bivariate models and multivar- iable models through purposeful selection of covariates. Variables were retained in the final models if they were associated with the exposure, the outcome, or improved overall model fit as determined by the Akaike Information Criterion [18]. This technique is widely utilized as a model selection criteria, which is used to compare different possible candidate models and variables to determine which one is the best fit for the data. In the case where the simplified model containing the exposure variable was the most parsimonious, we reported that as the final model.

  1. Results
    1. Characteristics of sample

There were 234 patients in the final sample, corresponding with 117 SUD+ and 117 SUD- individuals. Approximately 75.2% of all patients were male, and the age distribution was as follows: 18-24 (16.2%), 25-34 (24.4%), 35-44 (13.3%), 45-54 (22.2%), and 55-65 (23.9%). For

insurance, most frequently identified plans included Medicaid (38.5%), Commercial insurance (35.9%), and Medicare (12.8%).

In both SUD+ and SUD-, the most common mechanisms of injury in- cluded falls and assault (Table 1). SUD+ patients were more likely to present with an injury while intoxicated (uOR: 2.17; 95%CI: 1.09, 4.29), a history of depression (uOR: 8.00; 95%CI: 3.16, 20.27), anxiety

(uOR: 3.33; 95%CI: 1.58, 7.02), gastrointestinal chronic conditions

(uOR: 2.08; 95%CI: 1.05, 4.15), respiratory conditions (uOR: 2.14; 1.14,

4.04), and neurologic health conditions (uOR: 3.00; 95%CI: 1.52, 5.94). We observed no significant difference between the two groups in terms of reported pain (p = 0.42) and perceived pain (p = 0.72); how- ever, 50% of reported pain was marked as unknown or not reported and 80% of clinician perceived pain was marked as none.

    1. Outcomes

The most frequently administered medications in the ED included opioids (53.4%), followed by ibuprofen (8.1%), and acetaminophen (6.8%) (Table 2). We observed that 48.7% of SUD+ and 58.1% of SUD- patients were administered opioids in the ED. In the final adjusted model, the odds of receiving opioids among SUD+ was 0.33 (aOR: 0.14, 0.77) times that of SUD- patients (Table 3). Overall, the median amount of opioids administered among those receiving this medication was 2.0 MME (IQR: 0-7.5), and was significantly different between SUD

+ (median: 0, IQR: 0-6) and SUD- pairs (median: 4.0, IQR: 0-8.0), p =

0.052. For the secondary outcome, 56.4% of SUD+ patients and 68.4% of SUD- patients received a prescription for opioids at discharge from the ED. In the final model, the odds of receiving an opioid prescription among SUD+ was 0.50 (95%CI: 0.26, 0.95) times lower than for SUD- patients. There was no significant difference in the median amount of opioids prescribed at discharge between SUD+ (median: 50.0; IQR: 0-100.0) and SUD- (median: 75.0; IQR: 0-100.0) (p = 0.235).

  1. Discussion

The objective of this study was to examine whether we treat acute pain differently in patients with and without current or past substance use disorders by quantifying the amount of opioid analgesia received in the ED and the amount prescribed on discharge after acute fracture. In the setting of a national opioid crisis, it is important for clinicians to be aware of a patient’s current or prior substance use and be thoughtful in what types of medications they are giving them to avoid having a negative impact on their recovery or contributing to their addiction. This is a delicate balance, however, as we need to be sure we are not undertreating pain.

Table 1

Presentation and clinical characteristics of patients by substance use dependence history among patients presenting to the emergency department with a fracture

Characteristic

Total

SUD +

SUD –

uOR

95% CI

N

%

N

%

N

%

Presentation Characteristic

Mechanism of Injury

Fall

106

45.3

57

48.7

49

41.9

Ref

Assault

31

13.2

18

15.4

13

11.1

1.13

0.50, 2.69

Self Inflicted

22

9.4

12

10.3

10

8.5

1.03

0.38, 2.85

Motor Vehicle Collision

15

6.4

7

6.0

8

6.8

0.66

0.22, 2.02

Sport Injury

9

3.8

2

1.7

7

6.0

0.11

0.01, 0.97

Workplace Injury

8

3.4

3

2.6

5

4.3

0.33

0.04, 3.21

Other

43

18.4

18

15.4

25

21.4

0.51

0.23, 1.15

Severity of Pain (Self-Report)

None, mild, or moderate

48

20.5

19

16.2

29

24.8

Ref

Severe

68

29.1

37

31.6

31

26.5

1.67

0.83, 3.35

Unknown

118

50.4

57

48.7

61

52.1

1.58

0.79, 3.18

Severity of Pain (Perceived by Provider)

None, mild, or moderate

218

93.2

109

93.2

109

93.2

Ref

Severe

3

1.3

1

0.9

2

1.7

0.52

0.05, 5.87

Unknown

13

5.6

7

6.0

6

5.1

1.11

0.37, 3.36

Intoxicated while Injured (Ref = No)

44

18.8

29

24.8

15

12.8

2.17

1.09, 4.29

Clinical Characteristic

Documented Psychiatric Comorbidity/History (Ref = No)

Anxiety/Generalized Anxiety Disorder

47

20.1

34

29.1

13

11.1

3.33

1.58, 7.02

Depression

73

31.2

54

46.2

19

16.2

8.00

3.16, 20.27

Other

51

21.8

34

29.1

17

14.5

2.70

1.31, 5.58

Documented Chronic Medical Conditions (Ref = No)

Cardiac

66

28.2

39

33.3

27

23.1

1.60

0.92, 2.80

Gastrointestinal

75

32.1

44

37.6

31

26.5

2.08

1.05, 4.15

Respiratory

60

25.6

38

32.5

22

18.8

2.14

1.14, 4.04

Musculoskeletal

89

38.0

50

42.7

39

33.3

1.61

0.90, 2.90

Neurologic

60

25.6

41

35.0

19

16.2

3.00

1.52, 5.94

Obesity

21

9.0

7

6.0

14

12.0

0.46

0.18, 1.21

Other

99

42.3

60

51.3

39

33.3

2.91

1.47, 5.77

Other Documented Characteristics (Ref = No)

Fall Risk

18

7.7

9

7.7

9

7.7

1.00

0.32, 3.10

Allergy/Reaction to Opioids

13

5.6

9

7.7

4

3.4

2.67

0.71, 10.05

SUD = Substance Use Dependence; uOR = Unadjusted Odds Ratio.

Table 2

Medications (opioids, non-opioids) given in the ED and opioids prescribed at discharge

Table 3

Association between substance use dependence and amount of opioids given and pre-

Medications Patients

receiving medications

Dose of medication1

scribed in the ED

Outcome SUD + SUD – uOR 95% CI aOR 95% CI N % N %

N % Median IQR

Opioids Given in the ED1

Acetaminophen

16

6.8

650

650-650

None

60

51.3

49

41.9

Ref

Ref

Ketamine

4

1.7

25.0

12.8-45.5

Any

57

48.7

68

58.1

0.69

0.41,

0.33

0.14,

NSAID

1.15

0.77

Ibuprofen

19

8.1

600

600-600

Opioids Prescribed in the

Naproxen

0

0.0

***

***

ED2

Ketorolac

7

3.0

15

15-15

None

51

43.6

37

31.6

Ref

Ref

Indomethacin

0

0.0

***

***

Any

66

56.4

80

68.4

0.50

0.26,

0.50

0.26,

Muscle Relaxants/Anxiolytic

0.95

0.95

MME = Milligram morphine equivalent; *** = unable to calculate.

Cyclobenzaprine

0

0.0

***

***

Diazepam

1

0.4

2.0

2.0-2.0

Lorazepam

1

0.4

1.0

1.0-1.0

Opioids (mme) given

125

53.4

7.0

5.0-10.0

Opioids (mme) prescribed

146

62.4

100.0

75.0-150.0

1 Among those receiving these medications.

Our findings indicate that SUD+ patients were less likely to receive opioids in the ED, and when they did the amount they received was less than SUD- patients. Additionally, SUD+ patients were less likely to re- ceive an opioid prescription at discharge, but when they did the amount they received was similar to SUD- patients. One possible explanation for the observed results may be due to differences in presentation by SUD status. For example, SUD+ patients in our sample may have received less opioid analgesia in the ED as they were more likely to present

SUD = Substance Use Dependence; uOR = unadjusted odds ratio; aOR = adjusted odds ratio.

1 Final model of matched pairs (age, sex, insurance) adjusted for SUD exposure and self- reported pain.

2 Final model of matched pairs (age, sex, insurance) adjusted for SUD exposure.

intoxicated. As a result, clinicians may have worried about the interac- tion between alcohol and opioids given sedating properties of both sub- stances. Alternatively, it could be that these patients were undertreated for their pain. Undertreating pain negatively impacts the patients pain trajectory and may increase their likelihood of developing chronic pain [19] and may also influence or be associated with other Psychiatric comorbidities such as PTSD [20]. This is in contrast to the main focus in recent years of preventing over treating pain and over-prescribing opi- oids. At least 36 states have placed some limit on amount of opioids that can be prescribed without checking the state’s prescription drug

monitoring site [21]. This could be the reason that the amount of opioids prescribed at discharge was similar between the two groups, and the amount given may be viewed by clinicians as the minimum amount needed until follow up. While we do not have access to amounts of opi- oids prescribed through the prescription drug monitoring site in our state when used for research purposes, future research in other states may determine if SUD+ patients were more likely to seek additional opioids and which group was more likely to chronically be on opioids.

  1. Limitations

There are several limitations that should be addressed and that may be used to guide future research. First, we relied on SUD documented in the active problem list presented in a chart. This may have presented scenarios where SUD was not documented, resulting in a misclassifica- tion of the exposure. However, it is important to note that we were attempting to capture what an ED provider would have seen at the time of the visit, and whether the presence or absence of SUD on the ac- tive problem list would have subsequently resulted in differences in opioid administration and prescribing patterns. As a result, undocu- mented SUD was of a lesser concern with this particular research ques- tion. Second, the variety of types of fractures may be viewed as a limitation in our study as they may result in varying amounts of discom- fort. We acknowledge this and did attempt to look at differences in per- ceived and reported pain; however, much of these data were left incomplete or not reported. In those that were reported, there was no significant difference in perceived or reported pain between SUD + and – patients. Additionally, all injuries were able to be managed on an outpatient basis and did not require admission for emergent surgery or for IV pain control. Third, we did not evaluate if patients felt that their pain was adequately or inadequately treated. In future prospective stud- ies this could be part of the data collection, or in future retrospective studies a survey could be sent out to patients. Though we generated a matched sample of SUD+ and SUD- patients on pertinent variables, there may still be some unmeasured confounding that could explain a portion of the relationship between SUD and receipt of pain medica- tions. Finally, as this was a single site with a smaller sample size, there may be some limitations on the generalizability of these results to other settings. Future work may involve attempting to replicate these findings in other EDs or health systems.

  1. Conclusion

Overall, this study quantified how we treat acute pain in patients with and without a history of substance use disorder. This is something clinicians should be aware of so that we do not contribute to addiction or negatively impact recovery but also do not undertreat pain in SUD

+ patients. In order to address undertreated pain we must address pa- tient and physician attitudes and education [19], take a multimodal and multidisciplinary approach to pain control, and understand the com- plexities of pain and pain control [22]. There are currently no specific guidelines outlining if acute pain in SUD+ patients should be managed the same or differently than that of SUD- patients, however, we hope that identifying and quantifying our current practices can be the first step in working toward a better understanding acute pain management in SUD+ patients.

Funding

This research did not receive any specific grant from funding agen- cies in the public, commercial, or not-for-profit sectors.

Presentations

Presented at SAEM May 12, 2021.

Declaration of Competing Interest

None.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajem.2021.08.005.

References

  1. 10 Percent of US Adults have Drug use disorder at Some Point in their Lives. NIH News Releases; Nov. 18, 2015. https://www.nih.gov/news-events/news- releases/10-percent-us-adults-have-drug-use-disorder-some-point-their-lives (Accessed May 10, 2021).
  2. Hoppe JA, Nelson LS, Perrone J, et al. opioid prescribing in a cross section of US emer- gency departments. Ann Emerg Med. 2015;66:253-259.E1. https://doi.org/10.1016/ j.annemergmed.2015.03.026.
  3. Logan J, Liu Y, Paulozzi L, Zhang K, Jones C. Opioid prescribing in emergency depart- ments: the prevalence of potentially Inappropriate prescribing and misuse. Med Care. 2013;51(8):646-53. https://www.ncbi.nlm.nih.gov/pubmed/23632597.
  4. Pletcher MJ, et al. Trends in opioid prescribing by race/ethnicity for patients seeking care in US emergency departments. JAMA. 2008;299(1):70-8. https://jamanetwork. com/journals/jama/article-abstract/1149438.
  5. Heins JK, et al. Disparities in analgesia and opioid prescribing practices for patients with Musculoskeletal pain in the emergency department. J Emerg Nurs. June 2006;3(32):219-24. https://www.sciencedirect.com/science/article/abs/pii/S00991

76706000626.

  1. Sinnenberg L, et al. What Factors Affect Physicians’ Decisions to Prescripe Opioids in Emergency Departments? MDM Policy & Practice; January-June 2017. https://doi. org/10.1177/2381468316681006.
  2. Barnett ML, et al. Opioid- prescribing patterns of emergency physicians and risk of long-term use. NEJM. 2017;376:663-73. https://doi.org/10.1056/NEJMsa1610524.
  3. Delgado MK, Huang Y, Meisel Z, et al. National variation in opioid prescribing and risk of prolonged use for opioid-naive patients treated in the emergency Depart- ment for Ankle sprains. Ann Emerg Med. 2018;72:389-400. https://doi.org/10. 1016/j.annemergmed.2018.06.003.
  4. Chai PR, Carreiro S, Innes BJ, et al. Oxycodone ingestion patterns in acute fracture pain with digital pills. Anesth Analg. 2017;125(6):2105-12. https://www.ncbi.nlm. nih.gov/pubmed/29189367.
  5. 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(9):e022649. https://www.ncbi.nlm.nih.gov/pubmed/30224393.
  6. Heitz C, Morgenstern J, Bond C, Milne WK, Heitz C, et al. Hot off the press: SGEM #264-hooked on a feeling? Opioid use and misuse three months after emergency department visit for acute pain. Acad Emerg Med. 2020 Mar;27(3):240-2. https:// doi.org/10.1111/acem.13856.
  7. Potnuru P, Dudaryk R, Gebhard RE, Diez C, Velazquez OC, Candiotti KA, et al. Opioid prescriptions for acute pain after outpatient surgery at a large public university- affiliated hospital: Impact of state legislation in Florida. Surgery. 2019 Sep;166(3): 375-9. https://doi.org/10.1016/j.surg.2019.04.022 (Epub 2019 Jun 10.Surgery.

2019.PMID: 3119670).

  1. Daoust R, Paquet J, Cournoyer A, Piette E, Morris J, Lessard J, et al. Acute pain resolu- tion after an emergency department visit: a 14-day trajectory analysis. Ann Emerg Med. 2019 Aug;74(2):224-32. https://doi.org/10.1016/j.annemergmed.2019.01. 019.
  2. Salottolo K, Peck L, Carrick M, et al. High on drugs: multi-institutional pilot study ex- amining the effects of substance use on acute pain management. Injury. 2019;50(5): 1058-63. https://doi.org/10.1016/j.injury.2019.01.003. https://www.ncbi.nlm.nih. gov/pubmed/30642622.
  3. Anderson JE, Cocanour CS, Galante JM, Anderson JE, et al. Trauma and acute care surgeons report prescribing less opioids over time. Trauma Surg Acute Care Open. 2019 Mar 27;4(1):e000255. https://doi.org/10.1136/tsaco-2018-000255 (PMID: 31058238Free PMC article).
  4. von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Ann Intern Med. 2007 Oct 16;147(8):573-7. https://doi.org/10.7326/0003-4819-147-8-200710160-

00010.

  1. Opioid Oral Morphine Milligram Equivalent (MME) Conversion Factors. https:// www.cms.gov/Medicare/Prescription-Drug-Coverage/PrescriptionDrugCovContra/ Downloads/Oral-MME-CFs-vFeb-2018.pdf; 2021. Accessed May 10, 2021.
  2. Portet S. A primer on model selection using the Akaike Information Criterion. Infect Dis Model. 2020;5:111-28. https://doi.org/10.1016/j.idm.2019.12.010.
  3. Sinatra Raymond. Causes and consequences of inadequate management of acute pain. Pain Med. December 2010;11(12):1859-71.
  4. Kind S, Otis JD. The interaction between chronic pain and PTSD. Curr Pain Headache Rep. 2019;23:91. https://doi.org/10.1007/s11916-019-0828-3.
  5. Opioid Prescription Limits and Policies by State. https://ballotpedia.org/Opioid_ prescription_limits_and_policies_by_state; 2021. (Accessed May 10, 2021).
  6. King NB, Fraser V. Untreated pain, narcotics regulation, and global health ideologies. PLoS Med. 2013;10(4):e1001411. https://doi.org/10.1371/journal.pmed.1001411.