The Journal of Emergency Medicine, Vol. 42, No. 1, pp. 15–21, 2012Copyright � 2012 Elsevier Inc.

Printed in the USA. All rights reserved0736-4679/$ - see front matter

doi:10.1016/j.jemermed.2011.05.065

RECEIVED: 24 OACCEPTED: 29 M

OriginalContributions

QUANTIFYING DRUG-SEEKING BEHAVIOR: A CASE CONTROL STUDY

Casey A. Grover, MD,* Reb J. H. Close, MD,† Erik D.Wiele, BA,‡ Kathy Villarreal, RN,† and LeeM. Goldman, MD†

*Stanford/Kaiser EmergencyMedicine Residency, Stanford University Medical Center, Stanford, California, †Division of Emergency Medicine,Community Hospital of the Monterey Peninsula, Monterey, California, and ‡University of California at Berkeley, Berkeley, California

Reprint Address: Casey A. Grover, MD, Stanford University Medical Center, Division of Emergency Medicine, Alway Building, M121, 300Pasteur Dr – MC: 5119, Stanford, CA 94305

, Abstract—Background: Drug-seeking behavior (DSB) iscommon in the Emergency Department (ED), yet the litera-ture describing DSB in the ED consists predominantly ofanecdotal evidence. Study Objectives: To perform a case-control study examining the relative frequency of DSB insuspected drug-seeking patients as compared to all EDpatients. Methods: We performed a retrospective chart re-view of 152 drug-seeking patients and of age- and gender-matched controls, noting which of the following behaviorswere exhibited during a 1-year period: reporting a non-narcotic allergy, requesting addictive medications byname, requesting a medication refill, reporting lost or stolenmedication, three or more ED visits complaining of pain indifferent body parts, reporting 10 out of 10 pain, reporting> 10 out of 10 pain, three or more ED visits within 7 days,reporting being out of medication, requesting medicationsparenterally, and presenting with a chief complaint of head-ache, back pain, or dental pain. Results: The odds ratios foreach studied behavior being used by drug seeking patients ascompared to controls were: non-narcotic allergy: 3.4,medica-tion by name: 26.3, medication refill: 19.2, lost or stolen med-ication: 14.1, three or more pain related visits in differentparts of the body: 29.3, 10 out of 10 pain: 13.9, three visitsin 7 days: 30.8, out of medication: 26.9, headache: 10.9,back pain: 13.6, and dental pain: 6.3. Zero patients in the con-trol group complained of greater than 10-out-of-10 pain or re-quested medication parenterally, resulting in a calculatedodds ratio of infinity for these two behaviors. Conclusions: Re-questing parenteral medication and reporting greater thanten out of ten pain were most predictive of drug-seeking,while reporting a non-narcotic allergy was less predictive ofdrug-seeking than other behaviors. � 2012 Elsevier Inc.

ctober 2010; FINAL SUBMISSION RECEIVED: 3 Febray 2011

15

, Keywords—drug; seeking; behavior; emergency

INTRODUCTION

Pain is a common problem for which patients seek care inthe emergency department (ED), accounting for up to42% of all ED visits (1,2). Despite this, pain control inthe ED can be challenging, with inadequate paincontrol occurring frequently (2,3). There are manyreasons why pain control may be problematic in theED, such as variability in emergency physician (EP)prescribing practices and difficulty in assessinga patient’s level of pain (3–5). Additionally, EPs maybe reluctant to administer analgesia out of concern thata patient complaining of pain may be trying to obtainmedications for non-therapeutic purposes (2–4). Thesepatients, often labeled as ‘‘drug seeking,’’ representa difficult group of patients to manage in the ED. Theyoften present with conditions that may be easily feignedand are difficult to evaluate, such as headache, backpain, and dental pain (6,7). They also are known toengage in deceptive behavior in an attempt to foolclinicians into giving them additional medication. Suchbehaviors include prescription forgery, seeking carefrom multiple providers, reporting allergies to non-opioid analgesics, complaining of lost or stolen medica-tions, requesting refills, exaggerating symptoms, andusing multiple aliases (6–11). These patients are alsonoted to have an extensive knowledge about pain

uary 2011;

16 C. A. Grover et al.

medication, request to receive medications parenterally,request medication by name, and may exhibitthreatening or even violent behavior when denied themedications they want (6,10–12).

Drug-seeking patients are common in the ED,accounting for as many as 20% of all ED visits, and arealso known to consume large amounts of medicalresources (9,12). As a result, many EDs have createdhabitual patient files and case management programs inan attempt to track the use patterns of these patientsand minimize the amount of narcotics and othermedications that they receive (5,7,10,13–15).

Despite the magnitude of the problem and the famili-arity of physicians from all specialties with these patients,there is still much to learn about them. A review of themedical literature reveals that there are many publica-tions on the subject, including several screening tools(Screener and Opioid Assessment for Patients with Pain- Revised [SOAPP-R]; Opioid Risk Tool [ORT]; CurrentOpioid Misuse Measure [COMM]; Diagnosis, Intracta-bility, Risk, and Efficacy score [DIRE]; and AddictionBehaviors Checklist [ABC]) developed by pain manage-ment clinicians to assess for problematic medication usein chronic pain patients (16–23). However, few of thesestudies present any quantitative data on drug-seekingpatients in the ED, and these studies are limited to smallnumbers of patients (6,7,24,25). With this in mind, wechose to perform a case-control study on a large numberof patients referred to a case management program forsuspected narcotic abuse. To the best of our knowledge,this is the first study to date that provides quantitativedata as to the relative frequency of drug-seeking behav-iors in all patients suspected of non-therapeutic ED useas opposed to controls.

The goal of this investigation was to perform a case-control study to determine the relative frequency ofdrug-seeking behaviors in suspected drug-seekingpatients as compared to all ED patients. Given the diffi-culty in studying this group of patients, we are awarethat finding the exact frequency of any given behavioris very unlikely. However, we hope to provide practicingEPs with information as to which drug-seeking behaviorsare more commonly used by drug-seeking patients. Wefeel that, from a clinical standpoint, knowing that one par-ticular behavior is strongly associated with drug-seekingbehavior while another is not as strongly associatedwould be helpful in evaluating a patient suspected ofdrug seeking.

METHODS

This observational retrospective study was performed ata 205-bed community hospital in central California thathas approximately 45,000 visits to the ED each year.

This study was given institutional review board exemp-tion by the hospital committee on research.

The hospital has an existing case management pro-gram that was developed by the ED staff to adequatelymeet the needs and improve the overall care of patientsrecurrently seeking care in the ED for chronic medicalproblems, particularly narcotic addiction. The programis chaired and operated by an ED nurse, who overseesa committee consisting of ED physicians, a chemicaldependency physician, pain management clinicians, be-havioral health physicians and nurses, and social serviceproviders. Patients are enrolled in the case managementprogram if they are identified as having a large numberof visits to the ED in the months before enrollment.Patients can also be enrolled if nursing staff or physiciansrequest a case management evaluation for a particularpatient based on patient use patterns suspicious fordrug-seeking behavior. Additionally, patients can beenrolled if one of the EPs receives a letter from theCalifornia prescription monitoring program (Controlledsubstance Utilization Review and Evaluation System[CURES]) regarding a patient. As a part of the manage-ment of these patients in the case management program,each patient’s ED visits are analyzed, and the chronicproblem or reason for recurrent use is determined.Finally, a plan, such as a chemical dependency evaluationor limitation of narcotic refills for chronic problems, isdeveloped by the case management team to manage thechronic underlying problem for each patient in the outpa-tient setting.

Inclusion criteria for patients in our study group werethe following: any patient enrolled in the case manage-ment program that was given a referral to chemicaldependency and any patient enrolled in the case manage-ment program that had a care plan involving limitation ofnarcotics, benzodiazepines, or muscle relaxants.

Exclusion criteria for patients in our study group werethe following: all patients enrolled in the case manage-ment program whose care plans did not involve eithera chemical dependency evaluation or limitation of nar-cotics, benzodiazepines, or muscle relaxants.

Patients in the control group were randomly chosenfrom the hospital’s medical record system by their medi-cal record number. Each patient is randomly assigneda five- or six-digit medical record number when they firstreceive care at the hospital. For each patient in the casegroup, the medical record number was cut down to fourdigits. This four-digit medical record number was thenentered into the hospital’s database of patients, and thefirst patient that was found that was the same age and gen-der as the case patient was chosen as a control patient.Any patients with zero ED visits in their medical recordwere excluded, and a new control was found with at leastone ED visit in their medical record.

Table 1. List of Studied Behaviors

1. Presenting to the ED and reporting an allergy to a non-narcotic pain medication

2. Presenting to the ED with a chief complaint of headache3. Presenting to the ED with a chief complaint of back pain4. Presenting to the EDwith a chief complaint of dental pain5. Presenting to the ED and asking for narcotics,

benzodiazepines, or muscle relaxants by name6. Presenting to the EDwith a chief complaint of medication

refill for narcotics, benzodiazepines, or musclerelaxants

7. Presenting to the ED and reporting that narcotic,benzodiazepine, or muscle medications had been lostor stolen

8. Presenting to the ED on three or more separateoccasions with pain-related chief complaints indifferent parts of the body

9. Presenting to the ED and complaining of 10 out of 10 pain10. Presenting to the ED and complaining of >10 out of

10 pain11. Presenting to the ED for any reason three or more times

in a 7-day period12. Presenting to the ED and reporting being out of narcotic,

benzodiazepine, or muscle relaxant medications13. Presenting to the ED and requesting to be given narcotic,

benzodiazepine, or muscle relaxant medicationsparenterally

ED = emergency department.

Quantifying Drug Seeking 17

For the study group, all visits to the ED 1 year beforeenrollment in the case management program were re-viewed. For the control group, all visits to the ED duringa 1-year period were reviewed as well. As many of thecontrol patients had few ED visits, the majority of thepatients in our control group did not have visits to theED during the 1-year period of review of their corre-sponding case patient. As such, for each of our casepatients, we reviewed the most recent ED visit in themedical record as well as the year preceding that visit.The hospital’s electronic medical record system was im-plemented in 2004, and we conducted our review in 2010.As such, the largest possible separation in time betweenthe time periods of review for the two groups was 6 years.For the majority of case-control pairs, the separationbetween time periods of review was < 3 years.

Patient medical records were accessed using the hospi-tal’s medical record system, Horizon Patient Folder(McKesson, San Francisco, CA), and all physician andnurse documentation for each visit was carefullyreviewed. For each study patient and each control patient,it was recorded whether or not patients exhibited any ofthe behaviors listed in Table 1 at any point during the1-year time period of study. We did not count the numberof times each behavior was exhibited by a particularpatient; our focus in this analysis was to assess whichbehaviors are used by drug-seeking patients rather thanthe frequency with which they are used by individualdrug-seeking patients. These 13 behaviors were chosenfor assessment as they represent drug-seeking behaviors

reported in the literature (6–12). Whereas certainbehaviors, such as headache and reporting a non-narcotic allergy, are easy to assess in a chart review,behaviors such as exaggeration of symptoms and highfrequency of use are more difficult to measure. We thuschose to make some measurable equivalents for thebehaviors that were more difficult to measure. In tryingto assess exaggeration of symptoms, we chose to lookfor complaining of 10 out of 10 and > 10 out of 10 pain.Also, in trying to assess for frequent use of medicalservices, we chose to look for presenting to the ED formore than three pain-related complaints and presentingto the ED three times within 7 days.

Once collected, the data were analyzed using Excel(Microsoft Corporation, Redmond, WA) and DimensionResearch statistical software (Dimension Research,Drums, PA, 2010). For each of the 13 data parameters col-lected, the percentage of patients in each group exhibitingeach behaviorwas calculated, and a 95%confidence inter-val for each was calculated. Also, an odds ratio was calcu-lated comparing the study group to the control group, aswere 95% confidence intervals for the odds ratios.

RESULTS

Review of the patients in the case management programidentified 152 patients meeting inclusion criteria. Demo-graphically, the average age of this patient group was 43.4years, and the group was composed of 65.8% females.

In the 1 year before enrollment in the case manage-ment program, these patients accounted for 2203 visitsto the ED, which is an average of 14.5 visits per patientper year. In the 1 year of study for the control group, thesepatients accounted for 315 visits to the ED, which is anaverage of 2.1 visits per patient per year.

The number of patients exhibiting each of the studiedbehaviors can be found in Table 2. The calculated odds ra-tio comparing each of the studied behaviors between thestudy group and the control group can be found in Table 3.

DISCUSSION

To the best of our knowledge, these data and analysis rep-resent the largest group of ED patients suspected of drugseeking studied to date. Additionally, to the best of ourknowledge, this is also the first time that drug-seeking pa-tients in the ED have been compared to controls for allchief complaints in an attempt to quantify and comparedifferent behaviors attributed to drug-seeking patients.

One major difficulty in studying patients exhibitingdrug-seeking behavior is that it is nearly impossible todefinitively determine whether or not a patient is trulyseeking care in an attempt to obtain medications fornon-therapeutic reasons. Pseudoaddiction is a condition

Table 2. Number and Percent of Patients Exhibiting Studied Behaviors

Case Groupn = 152 Percent 95% CI

Control Groupn = 152 Percent 95% CI

10 out of 10 pain 114 75.0 68.1–81.9 27 17.8 11.7–23.8Out of medication 91 59.9 52.1–67.7 8 5.3 1.7–8.8Back pain 88 57.9 50.0–65.7 14 9.2 4.6–13.8Request by name 85 55.9 48.0–63.8 7 4.6 1.3–7.9Over 3 pain complaints 83 54.6 46.7–62.5 6 3.9 0.9–7.1Headache 70 46.1 38.1–54.0 11 7.2 3.1–11.4Three visits in 7 days 69 45.4 37.5–53.3 4 2.6 0.1–5.2Chief complaint of refill 60 39.5 31.7–47.2 5 3.3 0.5–6.1Requesting parenteral 46 30.3 23.0–37.6 0 0.0 N/ANon-narcotic allergy 27 17.8 11.7–23.8 9 5.9 2.2–9.7>10 pain 21 13.8 8.3–19.3 0 0.0 N/ADental pain 17 11.2 6.2–16.2 3 2.0 0.0–4.2Lost or stolen medication 13 8.6 4.1–13.0 1 0.7 0.0–2.0

95% Confidence intervals (CI) are for the percentages of patients exhibiting each behavior.

18 C. A. Grover et al.

resulting from inadequate pain management, in whichpatients exhibit drug-seeking behaviors to obtain medica-tion so as to relieve their pain. This condition generallyresolves once the pain is treated. The behaviors exhibitedby patients suffering from pseudoaddiction are difficult todifferentiate from those of true addiction, especially inthe acute care setting (9). In our analysis, we did notmake any attempt to determine motivation for the drug-seeking behavior; rather, we only chose to examine thedifferent behaviors associated with drug-seeking, regard-less of cause.

As this pertains to our study patient population, manyof the patients referred to the case management programfor problems with prescription medications had underly-ing chronic pain issues. As an example, a significantproportion of the patients complaining of headachesand back pain reported histories of migraines and chroniclow back pain, respectively. It would seem that the inclu-sion of these patients in our study group would be a com-plicating factor, in that these patients would be morelikely to be suffering from real pain from their chronicpain condition. However, in the large community sur-rounding our hospital, there are undoubtedly more people

Table 3. Odds Ratios for Studied Behaviors

Odds Ratio 95% CI

Requesting parenteral N N/A>10 pain N N/AThree visits in 7 days 30.8 10.84–87.30Over 3 pain complaints 29.3 12.18–70.33Out of medication 26.9 12.28–58.72Request by name 26.3 11.54–59.86Chief complaint of refill 19.2 7.42–49.52Lost or stolen medication 14.1 1.82–109.3710 out of 10 pain 13.9 7.98–24.19Back pain 13.6 7.17–25.60Headache 10.9 5.48–21.85Dental pain 6.3 1.79–21.81Non-narcotic allergy 3.4 1.55–7.57

CI = confidence interval.

that suffer migraines than the 70 patients in our case man-agement program that repeatedly presented to the EDcomplaining of headache. It stands to reason, then, thatthe patients in our case management program are a differ-ent patient population than other chronic pain patients.They choose to frequent the ED for pain-related com-plaints in an attempt to obtain prescription medicationrather than seek regular care from a primary care physi-cian or specialist. It is perhaps possible that the patientsin our case management program are simply thosepatients with the most severe disease, and their visits tothe ED reflect a desperate effort to control severe pain.However, in reviewing many of the medical records, aninteresting pattern can be seen in a large number of thepatients in the case management program that suggestsa different explanation. Most of the patients begin witha significant disease process, such as migraines, a severetrauma, rheumatoid arthritis, or a work-related injury. Af-ter multiple ED visits and likely multiple visits to otherproviders for pain-related complaints, patients developtolerance for and dependence on the medications theyare taking. Soon, these patients begin presenting to theED in withdrawal and request larger and larger doses ofthe medications upon which they now are dependent.With this in mind, as much as they truly do have underly-ing pain, chemical dependency becomes a major motivat-ing factor for these patients to seek emergency care toobtain medication. This has been described previouslyas a transition in the lives of those with chronic painfrom being pain centered to being both ‘‘opioid centered’’and pain centered (9).

We observed in this study that patients in the drug-seeking group complained of 10 out of 10 pain morethan patients in our control group. Furthermore, our studygroup patients occasionally complained of > 10 out of 10pain, an event that was not observed at all in the controlgroup. This could be explained by the fact that chronicnarcotic use has been shown to increase chronic pain

Quantifying Drug Seeking 19

and can induce long-term changes in the brain that alterpain perception (26). Perhaps the chronic narcotic usehas made these patients more sensitive to pain, and theytruly are suffering from the most severe pain possible.However, this must be considered carefully, as patientswith narcotic abuse and dependency are known to exag-gerate pain complaints in an attempt to obtain desiredmedication (6,8). It is nearly impossible to determinewhich of these two explanations is correct, and thepatients in our study group likely chose to repeatedlyseek ED care to obtain medication for a combination ofboth reasons.

In reviewing the confidence intervals listed for the cal-culated odds ratios in Table 3, there are two main pointsthat should be noted. First, as a result of the fact that zeropatients in the control group requested parenteral medica-tion or reported greater than 10-out-of-10 pain, the oddsratios for these behaviors were calculated as infinity.We interpret this to mean that odds ratios for these behav-iors were significantly higher than all others, and are mostpredictive of drug-seeking behavior. Second, for report-ing a non-narcotic allergy, the odds ratio is only 3.4,and the upper end of the 95% confidence interval forthe odds ratio is below the lower end of the 95% confi-dence interval for the odds ratio of most other behaviors.Thus, we interpret this to mean that reporting a non-nar-cotic allergy is less predictive of drug-seeking behavior ascompared to other behaviors. However, it is important tonote that the odds ratio for a non-narcotic allergy is >1,and is still a behavior that is more commonly used by sus-pected drug-seeking patients than controls.

Limitations

Our study had several limitations. First, as much as this is,to our knowledge, the largest study on drug-seeking pa-tients to date, it remains a small study. Second, we choseto use a retrospective observational study design, whichhas inherent limitations. As our data come from chartreview, we are dependent on nurse and physician chart-ing, which is not always uniform or accurate. In particu-lar, there may have been documentation bias on the partof treating physicians and nurses. In patients suspectedof drug-seeking, treating providers may have been morelikely to document behaviors associated with drug seek-ing, such as being out of medication, as compared topatients not suspected of drug seeking.

Third, the analysis of the data in our study is limited bythe low frequency of our studied behaviors in the controlgroup. This is one of the reasons why our confidenceintervals are so large and why we could not calculateodds ratios for two of our studied behaviors. Fourth, asmuch as all of the patients in our study group were exhib-iting drug-seeking behavior, it would be nearly impossi-

ble to assess whether or not our patients were sufferingfrom addiction or pseudoaddiction, as both groups exhibitdrug-seeking behavior. Fifth, our study population con-sists of patients exhibiting drug-seeking behavior whoare also frequent users of the ED. Our study may thusbe poorly applicable to patients presenting to an ED fora single visit or patients frequenting multiple EDs.Finally, there was a large disparity between the numberof visits between the case group and the control group.Given that two of our studied behaviors (three visitswithin 7 days, three or more pain-related visits) were inpart dependent on the number of visits by a patient, thismay have inappropriately biased our data on these twobehaviors towards being more common in drug-seekingpatients.

Directions for Future Research

Despite the fact that drug-seeking patients are common,research on such patients is difficult. The current litera-ture on these patients consists of small studies; a studywith a large number of patients could prove to be insight-ful. Furthermore, few of the existing studies on drug-seeking patients are prospective; additional studies ofthis design could also provide needed data on this groupof patients. Also, because it is challenging to assesswhether or not a patient is seeking care only to obtainmedication for secondary gain, a study of patients whohave confessed to pure drug seeking would be helpful.

Finally, in this study we chose to look at whether or notpatients exhibited certain behaviors at any point duringa 1-year period. We plan to review all visits in our casegroup to assess the number of times each behavior wasexhibited to gain insight into the frequency of these be-haviors in drug-seeking patients.

CONCLUSIONS

In this study of patients referred to an ED case manage-ment program for drug-seeking behavior, requesting par-enteral medication and complaining of >10 out of 10 painwere most predictive of drug-seeking behavior. Reportinga non-narcotic allergy was the least predictive of drug-seeking behavior, but was still more common in sus-pected drug-seeking patients than in controls. For theremainder of the behaviors studied, the confidence inter-vals were too wide to allow for meaningful interpretationof the data, and further research is needed.

REFERENCES

1. Pletcher MJ, Kertesz SG, Kohn MA, Gonzales R. Trends in opioidprescribing by race/ethnicity for patients seeking care in US emer-gency departments. JAMA 2008;299:70–8.

20 C. A. Grover et al.

2. O’Connor AB, Zwemer FL, Hays DP, Feng C. Outcomes after intra-venous opioids in emergency patients: a prospective cohort analy-sis. Acad Emerg Med 2009;16:477–87.

3. Rupp T, Delaney KA. Inadequate analgesia in emergencymedicine.Ann Emerg Med 2004;43:494–503.

4. Tamayo-Sarver JH, Dawson NV, Cydulka RK, Wigton RS,BakerDW.Variability in emergency physician decisionmaking aboutprescribing opioid analgesics. Ann Emerg Med 2004;43:483–93.

5. Hawkins SC, Smeeks F, Hamel J. Emergency management ofchronic pain and drug-seeking behavior: an alternative perspective.J Emerg Med 2008;34:125–9.

6. McNabb C, Foot C, Ting J, Breeze K, StickleyM. Diagnosing drug-seeking behavior in an adult emergency department. Emerg MedAustralas 2006;18:138–42.

7. Zechnich AD, Hedges JR. Community-wide emergency departmentvisits by patients suspected of drug-seeking behavior. Acad EmergMed 1996;3:312–7.

8. Sim MG, Hulse GK, Khong E. Acute pain and opioid seeking be-havior. Aust Fam Physician 2004;33:1009–12.

9. Hansen GR. The drug-seeking patient in the emergency room.Emerg Med Clin North Am 2005;23:349–65.

10. Carlson MJ, Baker LH. Difficult, dangerous, and drug-seeking: the3D way to better patient care. Am J Public Health 1998;88:1250–2.

11. Longo LP, Parran T Jr, Johnson B, Kinsey W. Addiction: Part II.Identification and management of the drug-seeking patient. AmFam Physician 2000;61:2401–8.

12. McCaffery M, Grimm MA, Pasero C, Ferrell B, Uman GC. On themeaning of ‘‘drug seeking’’ Pain Manag Nurs 2005;6:122–36.

13. Geiderman JM. Keeping lists and naming names: habitual patientfiles for suspected nontherapeutic drug-seeking patients. AnnEmerg Med 2003;41:873–81.

14. Pope D, Fernandes CM, Bouthillette F, Etherington J. Frequentusers of the emergency department: a program to improve careand reduce visits. CMAJ 2000;162:1017–20.

15. Grover CA, Close RJ, Villarreal K, Goldman LM. Emergency de-partment frequent user: Pilot study of intensive case managementto reduce visits and computed tomography. West J Emerg Med2010;11:336–43.

16. Gianutsos LP, Safranek S, Huber T. Clinical inquiries: is therea well-tested tool to detect drug-seeking behaviors in chronic painpatients? J Fam Pract 2008;57:609–10.

17. Chou R, Fanciullo GJ, Fine PG, Miaskowski C, Passik SD,Portenoy RK. Opioids for chronic noncancer pain: prediction andidentification of aberrant drug-related behaviors: a review of theevidence for an American Pain Society and American Academyof Pain Medicine clinical practice guideline. J Pain 2009;10:131–46.

18. Moore TM, Jones T, Browder JH, Daffron S, Passik SD. A compar-ison of common screening methods for predicting aberrant drug-related behavior among patients receiving opioids for chronicpain management. Pain Med 2009;10:1426–33.

19. Wu SM, Compton P, Bolus R, et al. The Addiction BehaviorsChecklist: validation of a new clinician-based measure of inappro-priate opioid use in chronic pain. J Pain SymptomManage 2006;32:342–51.

20. Butler SF, Budman SH, Fernandez KC, et al. Development and val-idation of the Current Opioid Misuse Measure. Pain 2007;130:144–56.

21. Webster LR, Webster RM. Predicting aberrant behaviors in opioid-treated patients: preliminary validation of the Opioid Risk Tool.Pain Med 2005;6:432–42.

22. Butler SF, Fernandez K, Benoit C, Budman SH, Jamison RN. Val-idation of the revised Screener and Opioid Assessment for Patientswith Pain (SOAPPR). J Pain 2008;9:360–72.

23. Belgrade MJ, Schamber CS, Lindgren BR. The DIRE score: pre-dicting outcomes of opioid prescribing for chronic pain. J Pain2006;7:671–81.

24. McNabb C, Foot C, Ting J, Breeze K, Stickley M. Profiling patientssuspected of drug seeking in an adult emergency department. EmergMed Australas 2006;18:131–7.

25. Chan L, Winegard B. Attributes and behaviors associated with opi-oid seeking in the emergency department. J Opioid Manag 2007;3:244–8.

26. Ossipov MH, Lai J, Vanderah TW, Porreca F. Induction of painfacilitation by sustained opioid exposure: relationship to opioidantinociceptive tolerance. Life Sci 2003;73:783–800.

Quantifying Drug Seeking 21

ARTICLE SUMMARY

1. Why is this topic important?With prescription drug abuse at a national all-time high

and drug-seeking patients making up a significant per-centage of all emergency department (ED) visits, dealingwith patients suspected of drug-seeking behavior is some-thing that every emergency physician must face fre-quently. Unfortunately, drug-seeking patients aredifficult to study, as there are no tests that can be usedto confirm or refute whether or not a patient is drug-seeking. As such, the literature on drug-seeking patientsin the ED is limited, and consists of predominantly anec-dotal evidence describing drug-seeking behaviors. Thus,practicing emergency physicians have little evidence-based information to use when trying to assess whetheror not a patient is drug-seeking.2. What does this study attempt to show?

This study attempts to quantify how frequently a partic-ular set of drug-seeking behaviors is used by drug-seekingpatients, as compared to controls. The goal is to demon-strate which behaviors are most likely to be predictiveof drug seeking behaviors.3. What are the key findings?

The behaviors with the highest odds ratios predictingdrug-seeking behavior were complaining of > 10 out of10 pain and requesting narcotic, benzodiazepine, or mus-cle relaxant medication to be given parenterally. Con-versely, the behavior with lowest odds ratio wasreporting an allergy to a non-narcotic pain medication.4. How is patient care impacted?

When caring for patients in the ED, a patient that eithercomplains of > 10 out of 10 pain or that requests narcotic,benzodiazepine, or muscle relaxant medication to begiven parenterally should alert the physician of the poten-tial for drug-seeking. Although reporting a non-narcoticallergy is more common in drug-seeking patients thanall-comers to the ED, it seems to be less predictive ofdrug-seeking behavior than other behaviors and shouldnot be used alone to determine whether or not a patientis drug-seeking.