- Email: [email protected]
Effectiveness of a Rural Emergency Department (ED)-Based Pain Contract on ED Visits Among ED Frequent Users
The Journal of Emergency Medicine, Vol. -, No. -, pp. 1–6, 2018 Ó 2018 Elsevier Inc. All rights reserved. 0736-4679/$ - see front matter
https://doi.org/10.1016/j.jemermed.2018.04.056
Original Contributions
EFFECTIVENESS OF A RURAL EMERGENCY DEPARTMENT (ED)-BASED PAIN CONTRACT ON ED VISITS AMONG ED FREQUENT USERS Abdulaziz Alburaih, MD and Michael D. Witting, MD, MS Department of Emergency Medicine, University of Maryland, Baltimore, Maryland Reprint Address: Abdulaziz Alburaih, MD, Department of Emergency Medicine, University of Maryland, 2302 Annapolis Ridge Ct, Annapolis, MD 21401
, Abstract—Background: Caring for patients with chronic pain in emergency departments (EDs) can be particularly challenging, for both patients and physicians. Objective: This study sought to determine, in a rural setting, the effect of an ED-based pain contract on the rate of ED visits among patients with frequent visits for pain not related to cancer. Methods: This is a multi-ED, retrospective, beforeand-after chart review assessing the effect of a rural EDbased pain contract on the frequency of ED visits. The study setting consisted of four rural EDs representing over 85,000 annual visits. Medical records of patients eligible for a standardized pain contract during a 10-year period (December 2005–December 2015) were reviewed. Only visits involving complaints of pain were included. The number of visits during the year prior to contract initiation was compared with the number of visits during the year after enrollment, using a paired t-test. Results: We enrolled 314 patients, 185 (59%) of whom were female. The study group’s median age was 48 years. The mean number of ED visits was 12.4 visits (95% confidence interval [CI] 11.5–13.3) 1 year prior to the pain contract and 6.5 (95% CI 5.6–7.3) 1 year afterward (p < 0.0001). The mean number of ED visits decreased by 6.0 (95% CI 5.0–7.2). Conclusion: A pain contract protocol was associated with a significant reduction in the number of ED visits to multiple rural EDs. Ó 2018 Elsevier Inc. All rights reserved.
INTRODUCTION Pain is the most common presentation for emergency department (ED) visits; roughly two-thirds of visits are for acute pain or acute exacerbations of chronic pain (1,2). Opioids are widely used to treat acute and chronic pain in the ED, but unfortunately, their use has become an epidemic in the United States, with dire consequences of overdose and addiction. U.S. residents constitute 4.6% of the world’s population but consume approximately 80% of the global supply of opioids (3,4). Multiple data sources have shown that abuse of prescribed drugs is increasing and that once-typical patterns of abuse are changing (5). A recent study revealed that 1 out of 6 patients discharged from the ED received an opioid prescription (6). A study examining the opioidprescribing patterns by medical specialties in the United States between 2007 and 2012 showed emergency medicine among the top five specialties that dispensed opioids (7). In a recent cross-sectional study of patients who reported heroin or nonmedical opioid use, 59% reported they were first exposed to opioids by a legitimate medical prescription and 29% of them said the prescription came from an ED (8). The increased misuse of prescribed opioids, and the resulting addiction and overdose, have made EDs the top destination for people seeking opioid prescriptions. In most states, hospitals and ED administrators have developed programs to curtail abuse of prescribed opioids, with the hope of decreasing the overdose rate.
, Keywords—opioid abuse; pain contract; frequent users; emergency department
RECEIVED: 14 December 2017; FINAL SUBMISSION RECEIVED: 13 April 2018; ACCEPTED: 20 April 2018 1
2
A. Alburaih and M. D. Witting
In this article, we describe a strategy that has been implemented in a multi-hospital, university-affiliated emergency medicine practice: a pain contract policy, intended to prevent patients from over-utilizing the EDs for acute pain or chronic pain not related to cancer. We sought to determine the difference in pain-related ED visits per year associated with pain contract implementation. METHODS Study Design This is a multi-ED, retrospective, before-and-after chart review of all adult patients who were enrolled in a pain contract protocol. It compared the frequency of their pain-related ED visits prior to and after pain contract application. The study was approved by the institutional review boards of all participating hospitals and the affiliated medical school. Setting and Participants Four community-based EDs with more than 85,000 annual ED visits participated in this study. These EDs are located in a four-county contiguous area, covering the western portion of the Eastern Shore of Maryland; the closest major EDs are 32 miles to the southeast and 22 miles to the west of this area. Most of the physicians who staff them are either trained in or board certified in emergency medicine, and the same physicians have the opportunity to work at any of the four EDs. A uniform electronic medical record (EMR) is used in all four EDs, and information from all visits to any of the four EDs is collected in the same database. The group adopted a pain contract policy in December 2005, which aimed to prevent patients from over-utilizing the EDs for noncancer chronic pain and nonacute conditions, and to reduce the number of prescriptions written for controlled substances. The program was started by the director at one hospital and was then adopted by the other area hospitals. Patient enrollment criteria in the pain contract program included the following: Ten or more pain-related ED visits in 1 year, with receipt of a prescription for controlled substance during the majority of those visits. Behavior demonstrating a potential problem with controlled substances, such as altering a prescription, or overdose on a prescribed controlled substance. Multiple prescriptions for controlled substances, identified through a regional database system. This system became available in 2011. Once a patient was deemed eligible for the pain contract, the emergency physician explained it to the patient
and asked the patient to sign a hard copy of the pain contract. A copy of the pain contract was distributed to clinical information management personnel and the ED administration assistant. A notation was entered into the EMR to alert future visits that a pain contract had been initiated. If a patient refused to sign the pain contract, the emergency physician documented this and entered the unsigned pain contract into the patient’s medical record and distributed a copy to the ED administrative assistant. As part of the pain contract protocol, a list of patients selected for the program was maintained. Data Collection Patients were identified using the list maintained under the pain contract protocol. The number of pain-related ED visits 1 year prior to and 1 year after the implementation of the pain contract were tallied in all four ED locations. Only pain-related ED visits were included in the number of ED visits. ED visits that led to hospitalization were excluded. Pain-related visits were identified using the EMR chief complaint for the respective ED visit. A visit was considered pain related if the chief complaint contained the word ‘‘pain’’ or a synonym, such as ‘‘ache,’’ for example. Demographic information (age, sex, ethnicity) was collected. Documentation of patients’ alcohol use, drug use, and tobacco use was extracted from the EMR (MEDITECH client server, Version 5.67 pp 10, Westwood, MA). The EMR included a yes/no field for current smoking and drinking alcohol in the past year. Patients with a current or past history of recreational drug use were categorized as having a history of drug abuse. The EMR contained information about prescriptions for only part of the study period (after 2012), so we did not collect data on ED opiate prescriptions provided. Data were abstracted by an ED administrative assistant who received an hour-long training in the use of the abstraction form, database, and chart elements. The abstractor was not blinded to the objectives of the study. All data were then reviewed by the principal investigator of the study, who made occasional corrections. Data Analysis We hypothesized that pain contract initiation would reduce ED visits. All data analysis was performed using SAS University Edition Studio Version 3.4 (SAS Institute Inc., Cary, NC). A paired t-test was used to compare continuous variables prior to and after the pain contract initiation date. A p value # 0.05 was considered significant. No power calculation was performed, as all eligible patients in the study period were included in the sample.
Rural ED-Based Pain Contract
3
In a post hoc analysis, we performed a subgroup analysis of those with high utilization of ED resources. High utilization was defined as more than 20 visits for painrelated complaints per year. RESULTS There were 314 adult patients in whom the pain contract was initiated between the period of December 2005 and December 2015, identified through an existing list. Of these, 87% (n = 274) were enrolled in the pain contract due to multiple pain-related ED visits (>10 visits in 1 year), 7% were enrolled due to forged or altered prescription, 2.5% (n = 8) after being identified through a regional database system for receiving multiple prescriptions for controlled substances, 2.2% (n = 7) were enrolled because they used an alias, and 1% (n = 3) were enrolled after overdose. Only 17 patients (5.4%) refused to sign the pain contract. ‘‘Time zero’’ was defined as the date the pain contract was signed by the patient or the issuing of the contract by ED care provider. Demographics and Social History We collected data from 314 patients who were enrolled in the pain contract. Table 1 provides descriptive statistics for the study population. Most patients were female, and the median age of the study group was 48 years (48.5 years for female and 46 years for male). Most patients self-identified as White. There was a high frequency of smoking and history of drug abuse in this population. Visits Total pain-related ED visits pre-pain contract was 3898, vs. 2044 post-pain contract, with a 52% reduction in ED visits. The mean number of ED visits 1 year precontract was 12.4 visits (95% confidence interval [CI] 11.5–13.3) and the mean number of ED visits 1 year postcontract was 6.5 (95% CI 5.7–7.3), which is statistically significant (p < 0.0001). The mean number of ED visits decreased Table 1. Descriptive Statistics for the Study Population (N = 314 Patients) Characteristics
n (%)
Female Race/ethnicity White African-American Hispanic Smoking Alcohol use Recreational drug use
185 (59%) 248 (79%) 64 (20%) 2 (1%) 92 (29%) 87 (28%) 82 (26%)
by 6.0 (95% CI 5.0–7.2). Figure 1 shows the distribution of differences in visits prior to and after pain contract initiation. In the post hoc analysis, 41 patients, defined as high ED utilizers, had a mean of 27.9 ED visits for pain (95% CI 24.8–30.9) 1 year pre contract, and a mean of 15.2 visits (95% CI 11.0–19.4) in the year post contract (p < 0.0001). The mean number of ED visits for high utilizers decreased by 12.6 (95% CI 9.5–15.7). Table 2 shows the number of visits for painful complaints, prior to and after contract initiation. Seventeen patients (5.4%) refused to sign the pain contract. These patients had a mean of 14.6 ED visits for pain (95% CI 10.4–18.7) 1 year pre contract and a mean of 6.4 visits for pain (95% CI 3.2–9.5) in the year post contract (p < 0.0019). The mean number of ED visits for these patients decreased by 8.2 (95% CI 3.3–13.1). DISCUSSION The number of ED visits for opioid abuse more than doubled between 2004 and 2008 in the United States. In 2008, drug overdose overtook motor vehicle crash as the leading cause of accidental death in the United States (9,10). The increase in opioid deaths was accompanied by a notable shift from urban to rural areas (11). In this study, implementation of a pain contract in a group of neighboring rural EDs significantly reduced the mean number of ED visits, and this reduction was also statistically significant among high ED-utilizer patients (>20 ED visits per year). Unique features of our study include its relatively large size, the simple intervention of a pain contract application, and its rural setting. In a much smaller patient population (n = 46), Olsen et al. examined the impact of a chronic pain protocol on ED utilization at a suburban Level I trauma center and found a significant reduction in the number of return visits (12). In another small study, Stevenson and Meyer studied the effect of a nonnarcotic pain protocol in 15 patients being treated at a community teaching hospital and reported a significant drop in the number of pain-related visits (13). In a study examining a noncontract intervention, Fox et al. reported on the effect of an ED prescribing guideline on the number of opioid prescriptions written for patients with dental pain; issuance of the guidelines was associated with a 17% reduction in the rate of opioid prescriptions, and the number of ED visits for dental pain decreased from 26 to 21 per 1000 ED visits (95% CI of decrease 2 to 9 visits/1000) (14). Althaus et al. undertook a systemic review of interventions targeting frequent users of EDs (15). This systematic review identified 11 studies: three randomized controlled trials plus two controlled and six
4
A. Alburaih and M. D. Witting 30
25
Total Number of Patients
20
15
10
0
-32 -31 -31 -30 -29 -28 -27 -26 -25 -24 -23 -22 -21 -20 -19 -18 -17 -16 -15 -14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
5
Difference in visits
Figure 1. Distribution of Delta visits (postcontract minus precontract).
noncontrolled before-and-after studies. The intervention most frequently described was case management, a costly option for EDs (15). Specific opioid risk-reduction strategies have been implemented in pain clinics and primary care offices, including written opioid treatment agreements and urine Table 2. Number of Visits, Prior to and After Pain Contract, in Various Subgroups Mean Visits Mean Visits 1 Year Prior 1 Year After to Pain Pain Contract Characteristics Contract (SD) (SD) Gender Female Male Race White AfricanAmerican Hispanic Smoking Yes No Alcohol use Yes No Recreational drug use Yes No
Decrease in Paired Visits (95% CI)
drug testing for patients with chronic pain for whom longterm opioids are prescribed. A systematic review of opioid treatment agreements and urine drug testing showed relatively weak evidence to support the effectiveness of these methods in reducing opioid misuse by patients with chronic pain (16). This systematic review included 11 studies; six were conducted in pain clinics and five in primary care settings. Emergency physicians are on the front line of the opioid epidemic, and ultimately, the ED may be an important site for intervention (17). We believe that implementation of a pain contract or similar protocol is possible with education and motivation of ED staff, and it could result in significant reduction of pain-related ED visits as well as the potential for opioid abuse and overdose.
12.4 (9.5) 12.5 (7.1)
5.9 (7.6) 6.9 (7.3)
6.5 (5.3–7.6) 5.5 (4.5–6.5)
12.1 (8.0) 13.7 (8.7)
6.4 (7.0) 7.2 (8.9)
5.7 (4.9–6.5) 6.5 (4.5–8.5)
Limitations
10 (2.8)
4 (4.2)
6.0 ( 57.5–69.5)
12.3 (7.6) 12.7 (9.4)
6.7 (7.5) 6.1 (7.2)
5.5 (4.7–6.5) 6.5 (5.2–7.9)
12.5 (9.8) 12.4 (7.4) 12.4 (9.3)
6.5 (10.1) 6.5 (6.1) 7.1 (9.6)
6.0 (4.4–7.7) 5.8 (5.0–6.7) 5.3 (3.6–7.1)
12.4 (7.7)
6.3 (6.5)
6.1 (5.3–6.9)
Our study has several limitations. The pain contract was not available at all four hospitals throughout the study period, and patients were not screened routinely for pain contract eligibility; more patients may have been enrolled if there was a systematic screening effort at all 4 hospitals. The abstractor was not blinded to the objective of the study and was subject to review bias. This effect was minimized by having the principal investigator confirm the data abstraction. We recorded only visits to our four EDs, and so we likely missed patients’ other
CI = confidence interval.
Rural ED-Based Pain Contract
potential sources of opioids. However, given the rural setting, it would be difficult for patients to visit EDs outside our study group. It would have been interesting to observe the effect of the pain contract on opioid prescriptions dispensed by emergency physicians, but we could not obtain the required information from a retrospective chart review. We did not calculate the potential cost savings related to the decrease in ED visits after implementation of our pain contract program. The objective of this study was to determine the effect of a pain contract on the number of ED visits. We did not investigate the patient’s perspective; that is, reducing the number of ED visits does not necessarily indicate a reduction in prescription opioid abuse. We did not collect data on patient satisfaction or dissatisfaction with the pain contract program. ED staff members reported that some patients expressed anger over implementation of the pain contract protocol, but the majority seemed to understand the need for follow-up with their primary care physician or in a pain management clinic. The beforeand-after design allows for potential confounding by trends in ED visits and by the phenomenon of regression to the mean; because some patients were selected for the contract due to a high frequency of visits, their next year may have fewer visits not related to the contract. Further studies are required to confirm whether reducing the number of ED visits may reduce prescription opioid abuse. CONCLUSIONS Emergency physicians need to educate frequent ED users about the potential for opioid abuse and addiction. Our study found that implementation of a pain contract protocol, in a group of rural EDs, was associated with a decrease of six pain-related visits per person per year for those enrolled. Additional studies using similar protocols are needed to establish the impact of this strategy on patient well-being, experience, and potential to reduce health care costs. Acknowledgments—The authors wish to acknowledge Mrs. Deborah Stamper for her help with data collections. The manu-
5 script was copyedited by Linda J. Kesselring, MS, ELS, the technical editor/writer in the Department of Emergency Medicine at the University of Maryland School of Medicine.
REFERENCES 1. Cordell WH, Keene KK, Giles BK, et al. The high prevalence of pain in emergency medical care. Am J Emerg Med 2002;20: 165–9. 2. Todd KH, Ducharme J, Choiniere M, PEMI Study Group, et al. Pain in the emergency department: results of the Pain and Emergency Medicine Initiative (PEMI) multicenter study. J Pain 2007;8:460–6. 3. Centers for Disease Control and Prevention (CDC). CDC Grand Rounds: prescription drug overdoses—a US epidemic. MMWR Morb Mortal Wkly Rep 2012;61:10–3. 4. Wang J, Christo PJ. The influence of prescription monitoring programs on chronic pain management. Pain Physician 2009;12:507–15. 5. Maxwell JC. The prescription drug epidemic in the United States: a perfect storm. Drug Alcohol Rev 2011;30:264–70. 6. Hoppe JA, Nelson LS, Perrone J, et al. Opioid prescribing in a cross section us emergency departments. Ann Emerg Med 2015;66:253– 2591. 7. Levy B, Paulozzi L, Mack KA, et al. Trends in opioid analgesic– prescribing rates by specialty, U.S., 2007–2012. Am J Prev Med 2015;49:409–13. 8. Butler MM, Ancona RM, Beauchamp GA, et al. Emergency department prescription opioids as an initial exposure preceding addiction. Ann Emerg Med 2016;68:202–8. 9. Cai R, Crane E, Poneleit K, Paulozzi L. Emergency department visits involving nonmedical use of selected drugs-United States, 2004–2008. MMWR Morb Mortal Wkly Rep 2010;59:705–9. 10. Pitts S, Niska RW, Xu J, Burt C. National Hospital Ambulatory Medical Care Survey: 2006 emergency department summary. Natl Health Stat Report 2008;7:1–39. 11. Okie S. A flood of opioids, a rising tide of deaths. N Engl J Med 2010;363:1981–5. 12. Olsen JC, Ogarek JL, Goldenberg EJ, et al. Impact of a chronic pain protocol on emergency department utilization. Acad Emerg Med 2016;23:424–32. 13. Stevenson JE, Meyer TD. Effectiveness of nonnarcotic protocol for the treatment of acute exacerbations of chronic nonmalignant pain. Am J Emerg Med 2007;25:445–9. 14. Fox RR, Li J, Stevens S, et al. A performance improvement prescribing guideline reduces opioid prescriptions for emergency department dental pain patients. Ann Emerg Med 2013;62:237–40. 15. Althaus F, Paroz S, Hugli O, et al. Effectiveness of interventions targeting frequent users of emergency departments: a systematic review. Ann Emerg Med 2011;58:41–5242. 16. Starrels JL, Becker WC, Alford DP, et al. Systematic review: treatment agreements and urine drug testing to reduce opioid misuse in patients with chronic pain. Ann Intern Med 2010;152:712–20. 17. Hoppe JA, Kim H, Heard K. Association of emergency department opioid initiation with recurrent opioid use. Ann Emerg Med 2015; 65:493–9.
6
A. Alburaih and M. D. Witting
ARTICLE SUMMARY 1. Why is this topic important? The opioid epidemic in the United States is a tragedy, with an increase in overdose deaths involving prescription opioids. The number of emergency department (ED) visits for opioid abuse more than doubled between 2004 and 2008. Emergency physicians are on the front line of the opioid epidemic, and ultimately, the ED may be an important site for intervention. 2. What does this study attempt to show? The purpose of this study was to determine the effect of an ED-based pain contract on the rate of pain-related ED visits among patients with frequent visits for pain not related to cancer. 3. What are the key findings? For those in the pain contract program, pain-related ED visits decreased from 12.4 to 6.5 per year. Pain-related visits decreased by an even greater number among those identified as high utilizers. 4. How is patient care impacted? Implementation of a pain contract or similar protocol is possible with education and motivation of ED staff, and could result in significant reduction of pain-related ED visits.
Rural ED-Based Pain Contract
APPENDIX: PAIN CONTRACT PROGRAM A pain contract is a treatment agreement signed by the patient and clinician that sets out the expectations for a patient using these high-risk medications. The purpose of the administration of the pain contract to patients is to establish a procedure to prevent patients from overutilizing the emergency department (ED) for a chronic noncancer pain, nonacute condition, and to reduce the amount of controlled substances that patients receive, minimizing the potential for abuse and addiction. Patients eligible for a pain contract include: Any patient with 10 or more pain-related ED visits in 1 year and who received a prescription for controlled substance for the majority of the visits. Any patient who has demonstrated a potential problem with controlled substances. Any patient who attempts to alter a prescription. Any patient who has overdosed on a prescribed controlled substance. Any patient with a public sector criminal history regarding abuse, distribution, etc. of controlled substances. Any patient who generates prescriber awareness notification for multiple controlled substance prescription. Once a patient is thought to be eligible for the pain contract and after discussion with the patient, the following actions will take place: The ED physician will contact the patient’s primary physician if the patient has consented and is agreeable to the treatment plan and will notify the primary care physician of the initiation of the pain contract. The ED physician will discuss the contract with the patient, including an individualized plan, contact numbers for local pain specialists, pain clinics, and addiction programs. A copy of the pain contract is distributed to the patient, clinical information management, and ED administration assistant. Information will be entered into the electronic medical record and will alert future visits that a pain contract has been initiated. If the patient refuses to sign the pain contract, the ED physician will document such and enter the unsigned pain contract into the patient’s medical record and a copy distributed to ED administrative assistant for the pain contract.
6.e1
If acute emergency is ruled out and the ED physician has determined that the patient can be safely discharged, then the patient will be discharged from the ED in the usual manner. If a patient is to be admitted, the ED physician will discuss with the hospitalist that a pain contract has been initiated. A Blank Copy of Pain Contract A comprehensive review of your records from [hospital names removed for blinding] has demonstrated numerous visits for pain. These visits are far in excess of the average individual that presents to the Emergency Department. Your numerous visits have resulted in multiple prescriptions for controlled substances in the last year. This places you at high risk for addiction to controlled substances which would endanger your health. The Emergency Department physicians at [names removed for blinding] hospitals are trained in managing acute emergencies. We have no formal training in chronic pain control. Although we hope to continue to serve you for your acute emergency needs, we feel that your pain may be more adequately and consistently maintained by your primary care physician or a chronic pain clinic. Therefore, a goal can be set, focusing on your specific needs. It has been determined that your future emergency department care will be provided accordingly. The emergency department physician will identify any lifethreatening and acute emergencies. If one does not exist, then your pain will be managed with NON CONTROLLED SUBSTANCES ONLY. You will need to contact your primary physician the following day for appropriate follow-up care. This will create consistency in your pain control and reduce the risk of abuse and addiction. Attached you will find a sheet with important numbers for chronic pain clinics and rehab programs. If you feel that you may have an abuse or addiction problem, please contact the rehab numbers for assistance. I acknowledge and understand the above and have had the opportunity to ask and answer any questions. Patient’s Full Name Signature Date Emergency Physician signature Date Time
https://doi.org/10.1016/j.jemermed.2018.04.056
Original Contributions
EFFECTIVENESS OF A RURAL EMERGENCY DEPARTMENT (ED)-BASED PAIN CONTRACT ON ED VISITS AMONG ED FREQUENT USERS Abdulaziz Alburaih, MD and Michael D. Witting, MD, MS Department of Emergency Medicine, University of Maryland, Baltimore, Maryland Reprint Address: Abdulaziz Alburaih, MD, Department of Emergency Medicine, University of Maryland, 2302 Annapolis Ridge Ct, Annapolis, MD 21401
, Abstract—Background: Caring for patients with chronic pain in emergency departments (EDs) can be particularly challenging, for both patients and physicians. Objective: This study sought to determine, in a rural setting, the effect of an ED-based pain contract on the rate of ED visits among patients with frequent visits for pain not related to cancer. Methods: This is a multi-ED, retrospective, beforeand-after chart review assessing the effect of a rural EDbased pain contract on the frequency of ED visits. The study setting consisted of four rural EDs representing over 85,000 annual visits. Medical records of patients eligible for a standardized pain contract during a 10-year period (December 2005–December 2015) were reviewed. Only visits involving complaints of pain were included. The number of visits during the year prior to contract initiation was compared with the number of visits during the year after enrollment, using a paired t-test. Results: We enrolled 314 patients, 185 (59%) of whom were female. The study group’s median age was 48 years. The mean number of ED visits was 12.4 visits (95% confidence interval [CI] 11.5–13.3) 1 year prior to the pain contract and 6.5 (95% CI 5.6–7.3) 1 year afterward (p < 0.0001). The mean number of ED visits decreased by 6.0 (95% CI 5.0–7.2). Conclusion: A pain contract protocol was associated with a significant reduction in the number of ED visits to multiple rural EDs. Ó 2018 Elsevier Inc. All rights reserved.
INTRODUCTION Pain is the most common presentation for emergency department (ED) visits; roughly two-thirds of visits are for acute pain or acute exacerbations of chronic pain (1,2). Opioids are widely used to treat acute and chronic pain in the ED, but unfortunately, their use has become an epidemic in the United States, with dire consequences of overdose and addiction. U.S. residents constitute 4.6% of the world’s population but consume approximately 80% of the global supply of opioids (3,4). Multiple data sources have shown that abuse of prescribed drugs is increasing and that once-typical patterns of abuse are changing (5). A recent study revealed that 1 out of 6 patients discharged from the ED received an opioid prescription (6). A study examining the opioidprescribing patterns by medical specialties in the United States between 2007 and 2012 showed emergency medicine among the top five specialties that dispensed opioids (7). In a recent cross-sectional study of patients who reported heroin or nonmedical opioid use, 59% reported they were first exposed to opioids by a legitimate medical prescription and 29% of them said the prescription came from an ED (8). The increased misuse of prescribed opioids, and the resulting addiction and overdose, have made EDs the top destination for people seeking opioid prescriptions. In most states, hospitals and ED administrators have developed programs to curtail abuse of prescribed opioids, with the hope of decreasing the overdose rate.
, Keywords—opioid abuse; pain contract; frequent users; emergency department
RECEIVED: 14 December 2017; FINAL SUBMISSION RECEIVED: 13 April 2018; ACCEPTED: 20 April 2018 1
2
A. Alburaih and M. D. Witting
In this article, we describe a strategy that has been implemented in a multi-hospital, university-affiliated emergency medicine practice: a pain contract policy, intended to prevent patients from over-utilizing the EDs for acute pain or chronic pain not related to cancer. We sought to determine the difference in pain-related ED visits per year associated with pain contract implementation. METHODS Study Design This is a multi-ED, retrospective, before-and-after chart review of all adult patients who were enrolled in a pain contract protocol. It compared the frequency of their pain-related ED visits prior to and after pain contract application. The study was approved by the institutional review boards of all participating hospitals and the affiliated medical school. Setting and Participants Four community-based EDs with more than 85,000 annual ED visits participated in this study. These EDs are located in a four-county contiguous area, covering the western portion of the Eastern Shore of Maryland; the closest major EDs are 32 miles to the southeast and 22 miles to the west of this area. Most of the physicians who staff them are either trained in or board certified in emergency medicine, and the same physicians have the opportunity to work at any of the four EDs. A uniform electronic medical record (EMR) is used in all four EDs, and information from all visits to any of the four EDs is collected in the same database. The group adopted a pain contract policy in December 2005, which aimed to prevent patients from over-utilizing the EDs for noncancer chronic pain and nonacute conditions, and to reduce the number of prescriptions written for controlled substances. The program was started by the director at one hospital and was then adopted by the other area hospitals. Patient enrollment criteria in the pain contract program included the following: Ten or more pain-related ED visits in 1 year, with receipt of a prescription for controlled substance during the majority of those visits. Behavior demonstrating a potential problem with controlled substances, such as altering a prescription, or overdose on a prescribed controlled substance. Multiple prescriptions for controlled substances, identified through a regional database system. This system became available in 2011. Once a patient was deemed eligible for the pain contract, the emergency physician explained it to the patient
and asked the patient to sign a hard copy of the pain contract. A copy of the pain contract was distributed to clinical information management personnel and the ED administration assistant. A notation was entered into the EMR to alert future visits that a pain contract had been initiated. If a patient refused to sign the pain contract, the emergency physician documented this and entered the unsigned pain contract into the patient’s medical record and distributed a copy to the ED administrative assistant. As part of the pain contract protocol, a list of patients selected for the program was maintained. Data Collection Patients were identified using the list maintained under the pain contract protocol. The number of pain-related ED visits 1 year prior to and 1 year after the implementation of the pain contract were tallied in all four ED locations. Only pain-related ED visits were included in the number of ED visits. ED visits that led to hospitalization were excluded. Pain-related visits were identified using the EMR chief complaint for the respective ED visit. A visit was considered pain related if the chief complaint contained the word ‘‘pain’’ or a synonym, such as ‘‘ache,’’ for example. Demographic information (age, sex, ethnicity) was collected. Documentation of patients’ alcohol use, drug use, and tobacco use was extracted from the EMR (MEDITECH client server, Version 5.67 pp 10, Westwood, MA). The EMR included a yes/no field for current smoking and drinking alcohol in the past year. Patients with a current or past history of recreational drug use were categorized as having a history of drug abuse. The EMR contained information about prescriptions for only part of the study period (after 2012), so we did not collect data on ED opiate prescriptions provided. Data were abstracted by an ED administrative assistant who received an hour-long training in the use of the abstraction form, database, and chart elements. The abstractor was not blinded to the objectives of the study. All data were then reviewed by the principal investigator of the study, who made occasional corrections. Data Analysis We hypothesized that pain contract initiation would reduce ED visits. All data analysis was performed using SAS University Edition Studio Version 3.4 (SAS Institute Inc., Cary, NC). A paired t-test was used to compare continuous variables prior to and after the pain contract initiation date. A p value # 0.05 was considered significant. No power calculation was performed, as all eligible patients in the study period were included in the sample.
Rural ED-Based Pain Contract
3
In a post hoc analysis, we performed a subgroup analysis of those with high utilization of ED resources. High utilization was defined as more than 20 visits for painrelated complaints per year. RESULTS There were 314 adult patients in whom the pain contract was initiated between the period of December 2005 and December 2015, identified through an existing list. Of these, 87% (n = 274) were enrolled in the pain contract due to multiple pain-related ED visits (>10 visits in 1 year), 7% were enrolled due to forged or altered prescription, 2.5% (n = 8) after being identified through a regional database system for receiving multiple prescriptions for controlled substances, 2.2% (n = 7) were enrolled because they used an alias, and 1% (n = 3) were enrolled after overdose. Only 17 patients (5.4%) refused to sign the pain contract. ‘‘Time zero’’ was defined as the date the pain contract was signed by the patient or the issuing of the contract by ED care provider. Demographics and Social History We collected data from 314 patients who were enrolled in the pain contract. Table 1 provides descriptive statistics for the study population. Most patients were female, and the median age of the study group was 48 years (48.5 years for female and 46 years for male). Most patients self-identified as White. There was a high frequency of smoking and history of drug abuse in this population. Visits Total pain-related ED visits pre-pain contract was 3898, vs. 2044 post-pain contract, with a 52% reduction in ED visits. The mean number of ED visits 1 year precontract was 12.4 visits (95% confidence interval [CI] 11.5–13.3) and the mean number of ED visits 1 year postcontract was 6.5 (95% CI 5.7–7.3), which is statistically significant (p < 0.0001). The mean number of ED visits decreased Table 1. Descriptive Statistics for the Study Population (N = 314 Patients) Characteristics
n (%)
Female Race/ethnicity White African-American Hispanic Smoking Alcohol use Recreational drug use
185 (59%) 248 (79%) 64 (20%) 2 (1%) 92 (29%) 87 (28%) 82 (26%)
by 6.0 (95% CI 5.0–7.2). Figure 1 shows the distribution of differences in visits prior to and after pain contract initiation. In the post hoc analysis, 41 patients, defined as high ED utilizers, had a mean of 27.9 ED visits for pain (95% CI 24.8–30.9) 1 year pre contract, and a mean of 15.2 visits (95% CI 11.0–19.4) in the year post contract (p < 0.0001). The mean number of ED visits for high utilizers decreased by 12.6 (95% CI 9.5–15.7). Table 2 shows the number of visits for painful complaints, prior to and after contract initiation. Seventeen patients (5.4%) refused to sign the pain contract. These patients had a mean of 14.6 ED visits for pain (95% CI 10.4–18.7) 1 year pre contract and a mean of 6.4 visits for pain (95% CI 3.2–9.5) in the year post contract (p < 0.0019). The mean number of ED visits for these patients decreased by 8.2 (95% CI 3.3–13.1). DISCUSSION The number of ED visits for opioid abuse more than doubled between 2004 and 2008 in the United States. In 2008, drug overdose overtook motor vehicle crash as the leading cause of accidental death in the United States (9,10). The increase in opioid deaths was accompanied by a notable shift from urban to rural areas (11). In this study, implementation of a pain contract in a group of neighboring rural EDs significantly reduced the mean number of ED visits, and this reduction was also statistically significant among high ED-utilizer patients (>20 ED visits per year). Unique features of our study include its relatively large size, the simple intervention of a pain contract application, and its rural setting. In a much smaller patient population (n = 46), Olsen et al. examined the impact of a chronic pain protocol on ED utilization at a suburban Level I trauma center and found a significant reduction in the number of return visits (12). In another small study, Stevenson and Meyer studied the effect of a nonnarcotic pain protocol in 15 patients being treated at a community teaching hospital and reported a significant drop in the number of pain-related visits (13). In a study examining a noncontract intervention, Fox et al. reported on the effect of an ED prescribing guideline on the number of opioid prescriptions written for patients with dental pain; issuance of the guidelines was associated with a 17% reduction in the rate of opioid prescriptions, and the number of ED visits for dental pain decreased from 26 to 21 per 1000 ED visits (95% CI of decrease 2 to 9 visits/1000) (14). Althaus et al. undertook a systemic review of interventions targeting frequent users of EDs (15). This systematic review identified 11 studies: three randomized controlled trials plus two controlled and six
4
A. Alburaih and M. D. Witting 30
25
Total Number of Patients
20
15
10
0
-32 -31 -31 -30 -29 -28 -27 -26 -25 -24 -23 -22 -21 -20 -19 -18 -17 -16 -15 -14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
5
Difference in visits
Figure 1. Distribution of Delta visits (postcontract minus precontract).
noncontrolled before-and-after studies. The intervention most frequently described was case management, a costly option for EDs (15). Specific opioid risk-reduction strategies have been implemented in pain clinics and primary care offices, including written opioid treatment agreements and urine Table 2. Number of Visits, Prior to and After Pain Contract, in Various Subgroups Mean Visits Mean Visits 1 Year Prior 1 Year After to Pain Pain Contract Characteristics Contract (SD) (SD) Gender Female Male Race White AfricanAmerican Hispanic Smoking Yes No Alcohol use Yes No Recreational drug use Yes No
Decrease in Paired Visits (95% CI)
drug testing for patients with chronic pain for whom longterm opioids are prescribed. A systematic review of opioid treatment agreements and urine drug testing showed relatively weak evidence to support the effectiveness of these methods in reducing opioid misuse by patients with chronic pain (16). This systematic review included 11 studies; six were conducted in pain clinics and five in primary care settings. Emergency physicians are on the front line of the opioid epidemic, and ultimately, the ED may be an important site for intervention (17). We believe that implementation of a pain contract or similar protocol is possible with education and motivation of ED staff, and it could result in significant reduction of pain-related ED visits as well as the potential for opioid abuse and overdose.
12.4 (9.5) 12.5 (7.1)
5.9 (7.6) 6.9 (7.3)
6.5 (5.3–7.6) 5.5 (4.5–6.5)
12.1 (8.0) 13.7 (8.7)
6.4 (7.0) 7.2 (8.9)
5.7 (4.9–6.5) 6.5 (4.5–8.5)
Limitations
10 (2.8)
4 (4.2)
6.0 ( 57.5–69.5)
12.3 (7.6) 12.7 (9.4)
6.7 (7.5) 6.1 (7.2)
5.5 (4.7–6.5) 6.5 (5.2–7.9)
12.5 (9.8) 12.4 (7.4) 12.4 (9.3)
6.5 (10.1) 6.5 (6.1) 7.1 (9.6)
6.0 (4.4–7.7) 5.8 (5.0–6.7) 5.3 (3.6–7.1)
12.4 (7.7)
6.3 (6.5)
6.1 (5.3–6.9)
Our study has several limitations. The pain contract was not available at all four hospitals throughout the study period, and patients were not screened routinely for pain contract eligibility; more patients may have been enrolled if there was a systematic screening effort at all 4 hospitals. The abstractor was not blinded to the objective of the study and was subject to review bias. This effect was minimized by having the principal investigator confirm the data abstraction. We recorded only visits to our four EDs, and so we likely missed patients’ other
CI = confidence interval.
Rural ED-Based Pain Contract
potential sources of opioids. However, given the rural setting, it would be difficult for patients to visit EDs outside our study group. It would have been interesting to observe the effect of the pain contract on opioid prescriptions dispensed by emergency physicians, but we could not obtain the required information from a retrospective chart review. We did not calculate the potential cost savings related to the decrease in ED visits after implementation of our pain contract program. The objective of this study was to determine the effect of a pain contract on the number of ED visits. We did not investigate the patient’s perspective; that is, reducing the number of ED visits does not necessarily indicate a reduction in prescription opioid abuse. We did not collect data on patient satisfaction or dissatisfaction with the pain contract program. ED staff members reported that some patients expressed anger over implementation of the pain contract protocol, but the majority seemed to understand the need for follow-up with their primary care physician or in a pain management clinic. The beforeand-after design allows for potential confounding by trends in ED visits and by the phenomenon of regression to the mean; because some patients were selected for the contract due to a high frequency of visits, their next year may have fewer visits not related to the contract. Further studies are required to confirm whether reducing the number of ED visits may reduce prescription opioid abuse. CONCLUSIONS Emergency physicians need to educate frequent ED users about the potential for opioid abuse and addiction. Our study found that implementation of a pain contract protocol, in a group of rural EDs, was associated with a decrease of six pain-related visits per person per year for those enrolled. Additional studies using similar protocols are needed to establish the impact of this strategy on patient well-being, experience, and potential to reduce health care costs. Acknowledgments—The authors wish to acknowledge Mrs. Deborah Stamper for her help with data collections. The manu-
5 script was copyedited by Linda J. Kesselring, MS, ELS, the technical editor/writer in the Department of Emergency Medicine at the University of Maryland School of Medicine.
REFERENCES 1. Cordell WH, Keene KK, Giles BK, et al. The high prevalence of pain in emergency medical care. Am J Emerg Med 2002;20: 165–9. 2. Todd KH, Ducharme J, Choiniere M, PEMI Study Group, et al. Pain in the emergency department: results of the Pain and Emergency Medicine Initiative (PEMI) multicenter study. J Pain 2007;8:460–6. 3. Centers for Disease Control and Prevention (CDC). CDC Grand Rounds: prescription drug overdoses—a US epidemic. MMWR Morb Mortal Wkly Rep 2012;61:10–3. 4. Wang J, Christo PJ. The influence of prescription monitoring programs on chronic pain management. Pain Physician 2009;12:507–15. 5. Maxwell JC. The prescription drug epidemic in the United States: a perfect storm. Drug Alcohol Rev 2011;30:264–70. 6. Hoppe JA, Nelson LS, Perrone J, et al. Opioid prescribing in a cross section us emergency departments. Ann Emerg Med 2015;66:253– 2591. 7. Levy B, Paulozzi L, Mack KA, et al. Trends in opioid analgesic– prescribing rates by specialty, U.S., 2007–2012. Am J Prev Med 2015;49:409–13. 8. Butler MM, Ancona RM, Beauchamp GA, et al. Emergency department prescription opioids as an initial exposure preceding addiction. Ann Emerg Med 2016;68:202–8. 9. Cai R, Crane E, Poneleit K, Paulozzi L. Emergency department visits involving nonmedical use of selected drugs-United States, 2004–2008. MMWR Morb Mortal Wkly Rep 2010;59:705–9. 10. Pitts S, Niska RW, Xu J, Burt C. National Hospital Ambulatory Medical Care Survey: 2006 emergency department summary. Natl Health Stat Report 2008;7:1–39. 11. Okie S. A flood of opioids, a rising tide of deaths. N Engl J Med 2010;363:1981–5. 12. Olsen JC, Ogarek JL, Goldenberg EJ, et al. Impact of a chronic pain protocol on emergency department utilization. Acad Emerg Med 2016;23:424–32. 13. Stevenson JE, Meyer TD. Effectiveness of nonnarcotic protocol for the treatment of acute exacerbations of chronic nonmalignant pain. Am J Emerg Med 2007;25:445–9. 14. Fox RR, Li J, Stevens S, et al. A performance improvement prescribing guideline reduces opioid prescriptions for emergency department dental pain patients. Ann Emerg Med 2013;62:237–40. 15. Althaus F, Paroz S, Hugli O, et al. Effectiveness of interventions targeting frequent users of emergency departments: a systematic review. Ann Emerg Med 2011;58:41–5242. 16. Starrels JL, Becker WC, Alford DP, et al. Systematic review: treatment agreements and urine drug testing to reduce opioid misuse in patients with chronic pain. Ann Intern Med 2010;152:712–20. 17. Hoppe JA, Kim H, Heard K. Association of emergency department opioid initiation with recurrent opioid use. Ann Emerg Med 2015; 65:493–9.
6
A. Alburaih and M. D. Witting
ARTICLE SUMMARY 1. Why is this topic important? The opioid epidemic in the United States is a tragedy, with an increase in overdose deaths involving prescription opioids. The number of emergency department (ED) visits for opioid abuse more than doubled between 2004 and 2008. Emergency physicians are on the front line of the opioid epidemic, and ultimately, the ED may be an important site for intervention. 2. What does this study attempt to show? The purpose of this study was to determine the effect of an ED-based pain contract on the rate of pain-related ED visits among patients with frequent visits for pain not related to cancer. 3. What are the key findings? For those in the pain contract program, pain-related ED visits decreased from 12.4 to 6.5 per year. Pain-related visits decreased by an even greater number among those identified as high utilizers. 4. How is patient care impacted? Implementation of a pain contract or similar protocol is possible with education and motivation of ED staff, and could result in significant reduction of pain-related ED visits.
Rural ED-Based Pain Contract
APPENDIX: PAIN CONTRACT PROGRAM A pain contract is a treatment agreement signed by the patient and clinician that sets out the expectations for a patient using these high-risk medications. The purpose of the administration of the pain contract to patients is to establish a procedure to prevent patients from overutilizing the emergency department (ED) for a chronic noncancer pain, nonacute condition, and to reduce the amount of controlled substances that patients receive, minimizing the potential for abuse and addiction. Patients eligible for a pain contract include: Any patient with 10 or more pain-related ED visits in 1 year and who received a prescription for controlled substance for the majority of the visits. Any patient who has demonstrated a potential problem with controlled substances. Any patient who attempts to alter a prescription. Any patient who has overdosed on a prescribed controlled substance. Any patient with a public sector criminal history regarding abuse, distribution, etc. of controlled substances. Any patient who generates prescriber awareness notification for multiple controlled substance prescription. Once a patient is thought to be eligible for the pain contract and after discussion with the patient, the following actions will take place: The ED physician will contact the patient’s primary physician if the patient has consented and is agreeable to the treatment plan and will notify the primary care physician of the initiation of the pain contract. The ED physician will discuss the contract with the patient, including an individualized plan, contact numbers for local pain specialists, pain clinics, and addiction programs. A copy of the pain contract is distributed to the patient, clinical information management, and ED administration assistant. Information will be entered into the electronic medical record and will alert future visits that a pain contract has been initiated. If the patient refuses to sign the pain contract, the ED physician will document such and enter the unsigned pain contract into the patient’s medical record and a copy distributed to ED administrative assistant for the pain contract.
6.e1
If acute emergency is ruled out and the ED physician has determined that the patient can be safely discharged, then the patient will be discharged from the ED in the usual manner. If a patient is to be admitted, the ED physician will discuss with the hospitalist that a pain contract has been initiated. A Blank Copy of Pain Contract A comprehensive review of your records from [hospital names removed for blinding] has demonstrated numerous visits for pain. These visits are far in excess of the average individual that presents to the Emergency Department. Your numerous visits have resulted in multiple prescriptions for controlled substances in the last year. This places you at high risk for addiction to controlled substances which would endanger your health. The Emergency Department physicians at [names removed for blinding] hospitals are trained in managing acute emergencies. We have no formal training in chronic pain control. Although we hope to continue to serve you for your acute emergency needs, we feel that your pain may be more adequately and consistently maintained by your primary care physician or a chronic pain clinic. Therefore, a goal can be set, focusing on your specific needs. It has been determined that your future emergency department care will be provided accordingly. The emergency department physician will identify any lifethreatening and acute emergencies. If one does not exist, then your pain will be managed with NON CONTROLLED SUBSTANCES ONLY. You will need to contact your primary physician the following day for appropriate follow-up care. This will create consistency in your pain control and reduce the risk of abuse and addiction. Attached you will find a sheet with important numbers for chronic pain clinics and rehab programs. If you feel that you may have an abuse or addiction problem, please contact the rehab numbers for assistance. I acknowledge and understand the above and have had the opportunity to ask and answer any questions. Patient’s Full Name Signature Date Emergency Physician signature Date Time