- Email: [email protected]
Racial disparity in analgesic treatment for ED patients with abdominal or back pain
American Journal of Emergency Medicine (2011) 29, 752–756
www.elsevier.com/locate/ajem
Original Contribution
Racial disparity in analgesic treatment for ED patients with abdominal or back pain Angela M. Mills MD a,⁎, Frances S. Shofer PhD a , Ann K. Boulis PhD b , Daniel N. Holena MD c , Stephanie B. Abbuhl MD a a
Department of Emergency Medicine, University of Pennsylvania, Philadelphia, PA 19104, USA Department of Sociology, University of Pennsylvania, Philadelphia, PA 19104, USA c Division of Traumatology and Surgical Critical Care, Department of Surgery, University of Pennsylvania, Philadelphia, PA 19104, USA b
Received 6 January 2010; revised 5 February 2010; accepted 24 February 2010
Abstract Objective: Research on how race affects access to analgesia in the emergency department (ED) has yielded conflicting results. We assessed whether patient race affects analgesia administration for patients presenting with back or abdominal pain. Methods: This is a retrospective cohort study of adults who presented to 2 urban EDs with back or abdominal pain for a 4-year period. To assess differences in analgesia administration and time to analgesia between races, Fisher exact and Wilcoxon rank sum test were used, respectively. Relative risk regression was used to adjust for potential confounders. Results: Of 20 125 patients included (mean age, 42 years; 64% female; 75% black; mean pain score, 7.5), 6218 (31%) had back pain and 13 907 (69%) abdominal pain. Overall, 12 109 patients (60%) received any analgesia and 8475 (42%) received opiates. Comparing nonwhite (77 %) to white patients (23%), nonwhites were more likely to report severe pain (pain score, 9-10) (42% vs 36%; P b .0001) yet less likely to receive any analgesia (59% vs 66%; P b .0001) and less likely to receive an opiate (39% vs 51%; P b .0001). After controlling for age, sex, presenting complaint, triage class, admission, and severe pain, white patients were still 10% more likely to receive opiates (relative risk, 1.10; 95% confidence interval, 1.06-1.13). Of patients who received analgesia, nonwhites waited longer for opiate analgesia (median time, 98 vs 90 minutes; P = .004). Conclusions: After controlling for potential confounders, nonwhite patients who presented to the ED for abdominal or back pain were less likely than whites to receive analgesia and waited longer for their opiate medication. © 2011 Elsevier Inc. All rights reserved.
1. Introduction
⁎ Corresponding author. Tel.: +1 215 662 6698. E-mail address: [email protected] (A.M. Mills). 0735-6757/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.ajem.2010.02.023
Acute pain is one of the most common reasons for presenting to an emergency department (ED). The administration of analgesia for acute pain has been studied in the emergency setting, specifically the topic of oligoanalgesia, or the undertreatment of pain. Oligoanalgesia has been well
Race and analgesia treatment documented in the emergency medicine literature and attributed to a variety of factors. These include patient factors such as age [1-3], sex [4,5], and race [6-10]. With respect to race and ethnicity, there have been conflicting results. Two studies of ED patients with long bone fractures showed white patients were more likely to receive analgesics [9,10]. A recent study of a large group of patients from the National Hospital Ambulatory Medical Care Survey (NHAMCS) showed white patients were more likely than nonwhites to receive an opioid prescription for a pain-related visit [11]. On the contrary, 2 studies of analgesia in long bone fractures in both adults [12] and children [13] showed no difference in analgesia administration. Another recent study analyzed analgesic treatment of headache and long bone fractures also from the NHAMCS and demonstrated an improvement in racial inequalities over time and no difference in the treatment of long bone fractures with respect to race in the later study period of 2000 to 2003 [14]. Acute abdominal pain and acute back pain have been shown to be 2 of the most common specific reasons for seeking emergency care [15]. To our knowledge, there have been no large studies examining the effect of race on the administration of analgesia for abdominal or back pain in an emergency setting. We sought to study the effect of race on a large cohort of more than 20 000 patients presenting to the ED with abdominal or back pain. We hypothesized that nonwhite patients would be less likely to receive analgesia for the treatment of these complaints.
2. Methods 2.1. Study design and setting We performed a retrospective cohort study of adult ED patients who presented with a chief complaint of abdominal pain or back pain. The study was performed at 2 urban EDs with 57 000 and 35 000 annual ED visits, respectively. The Institutional Committee on Research Involving Human Subjects at the University of Pennsylvania approved the study.
2.2. Selection of participants Patients were identified using a computerized medical record search of EMTRAC (University of Pennsylvania, Philadelphia, Pa), a computerized charting and order entry system. All adult ED patients, 18 years and older, who presented between July 2003 and February 2007 to triage with a chief complaint of abdominal pain or back pain were included. We excluded patients who left without being seen, were pregnant, were missing race or ethnicity on the medical record, or who had no documented pain score or a triage pain score of “0.” Patients who identified themselves as
753 “white Hispanic” were included in the white group, and those who identified themselves as “black Hispanic” were included in the nonwhite group.
2.3. Methods of measurement A database was created from EMTRAC, a fully electronic emergency medical record, including patient demographics (age, sex, race/ethnicity), triage class (4-point scale [1-4] from emergent [1] to nonemergent [4]), triage pain score, pain medication type and time of administration, and final ED disposition. Triage pain scores were documented using a 10-point scale (1-10), from “no pain” (1) to “worst pain of my life” (10). Reassessments of pain were not available for this analysis. Time to analgesia was defined as time from patient placement in the treatment area (ED room or hallway) to time of analgesia administration as documented in the electronic record. Analgesics were defined as any pain medication administered in the ED including “to go” medications, which almost always consisted of 4 tablets of acetaminophen with oxycodone handed to patients about the time of discharge. Opiate analgesia was defined as any oral, intramuscular, or intravenous medication administered in the ED as documented in the electronic record (eg, acetaminophen with oxycodone, acetaminophen with codeine, morphine, hydromorphone). Nonopiate analgesia included any oral, intramuscular, or intravenous nonnarcotic medication (eg, acetaminophen, ibuprofen, ketorolac). The main outcome measures were analgesic administration and time to analgesic treatment.
2.4. Data analysis The Fisher exact test was used to test for differences between race and analgesia administration. To assess differences between races with regard to time to analgesic treatment between races, Wilcoxon rank sum test was used. To calculate the likelihood of receiving analgesia, an opiate, or receiving analgesia within an hour of room placement, a generalized linear model with a log link, Gaussian error, and robust estimates of the standard errors of the model coefficients was used [16]. These models were controlled for age, sex, severe pain, triage classification, admission status, and chief complaint of back or abdominal pain. Severe pain was defined as a pain score of 9 or 10. This list of confounders was determined a priori, and there were no stepwise techniques used to select variables. Descriptive data are presented as means with SD for continuous data, medians with interquartile ranges for time variables, and frequencies and percentages for categorical data. For the general linear model, data are presented as relative risks with 95% confidence intervals. The generalized linear models were performed using Stata statistical software (Version 10; Stata Corporation, College Station, Tex). All other analyses were performed using SAS statistical software (Version 9.1; SAS
754
A.M. Mills et al.
Institute, Cary, NC). A probability of less than .05 was considered statistically significant.
Table 2
Characteristic
Whites, n = 4681 (23%)
Nonwhites, n = 15 444 (77%)
P
3. Results
Age, y (mean ± SD) Sex Female Male Presenting complaint Abdominal pain Back pain Triage class 1 (emergent) 2 3 4 (nonemergent) Pain score (mean ± SD) 0-8 9-10 Received any analgesia Opiate Nonopiate Disposition Admission Discharge
43 ± 17
42 ± 17
.0008 b.0001
2612 (56) 2069 (44)
10 296 (67) 5148 (33)
3339 (71) 1342 (29)
10 568 (68) 4876 (32)
130 (3) 1902 (41) 2136 (46) 508 (11) 7.2 ± 2.4 3002 (64) 1679 (36) 3067 (66) 2382 (51) 685 (15)
124 (1) 2953 (19) 9639 (62) 2724 (18) 7.5 ± 2.3 9006 (58) 6438 (42) 9042 (59) 6093 (39) 2949 (19)
1582 (20) 3099 (80)
2714 (37) 12 730 (63)
We identified 20 125 patients with abdominal pain or back pain during the study period. Table 1 outlines general patient characteristics. Overall, the group was young (mean age, 42 years), more often female (64%) and black (75%), with most patients presenting with abdominal pain (69%). The mean triage pain score for the group was 7.5. Overall, 12 109 patients (60%) received any analgesia and 8475 (42%) received opiates. Comparing nonwhite (77%) to white patients (23%), nonwhites were more likely to be female (67% vs 56%; P b .0001), less likely to be admitted (63% vs 80%; P b .0001), and more likely to report severe pain (pain score, 9-10; 42% vs 36%; P b .0001). With respect to analgesia, nonwhite patients were less likely to receive any analgesia (59% vs 66%; P b .0001) and less likely to receive an opiate (39% vs 51%; P b .0001). Table 2 describes the racial differences in characteristics. Several potential confounders associated with analgesic treatment were identified: age, sex, chief
Table 1
Overall general patient characteristics
Characteristic
Overall, n = 20 125 (%)
Age, y (mean ± SD) Race/ethnicity Black Asian American Indian Black Hispanic White Hispanic White Sex Female Male Presenting complaint Abdominal pain Back pain Pain score (mean ± SD) 1-8 9-10 Triage class 1 (emergent) 2 3 4 (nonemergent) Received any analgesia Opiate Nonopiate Disposition Admission Discharge
42 ± 16.7 14 980 (74.4) 339 (1.7) 41 (0.2) 79 (0.4) 142 (0.7) 4539 (22.6) 12 908 (64) 7217 (36)
Racial differences in patient characteristics
.0002
b.0001
b.0001 b.0001 b.0001
complaint, triage class, admission, and severe pain (pain score, 9-10). After adjustment for these potential confounders, white patients were still 10% more likely than nonwhite patients to receive opiates (relative risk, 1.10; 95% confidence interval, 1.06-1.13) (Table 3). Of those patients who received analgesia, there was no difference in time to any analgesia among groups (median time, 97 vs 93 minutes; P = .10), but nonwhite patients waited longer than whites when receiving opiate analgesia (median time, 98 vs 90 minutes; P = .004). Table 3 Relative risk regression of variables associated with analgesic treatment Analgesia
Characteristic
Relative risk
13 907 (69) 6218 (31) 7.5 ± 2.4 12 008 (60) 8117 (40)
95% confidence interval
Any
254 (1) 4855 (24) 11 775 (59) 3232 (16) 12 109 (60) 8475 (42) 3634 (18)
Opiate
White Admission Pain score 9-10 Triage class Back pain Female Age N 65 White Admission Pain score 9-10 Triage class Back pain Female Age N 65
1.02 1.39 1.30 0.85 1.35 1.04 0.85 1.10 1.68 1.49 0.83 1.04 1.00 0.87
1.00 1.36 1.28 0.84 1.32 1.02 0.82 1.06 1.62 1.45 0.81 1.00 0.97 0.83
4296 (21) 15 829 (79)
1.04 1.42 1.33 0.87 1.38 1.06 0.88 1.13 1.73 1.54 0.85 1.08 1.02 0.91
Race and analgesia treatment
4. Discussion Multiple studies examining the effects of race and ethnicity on analgesia administration have been published with conflicting results. Most of these studies were conducted in patient populations with musculoskeletal complaints, specifically long bone fractures [9,10,12-14]. This study was conducted to assess whether a racial disparity exists in the treatment of 2 of the most common specific reasons for patients seeking emergency care, abdominal pain and back pain. Our study comprised a large cohort of more than 20 000 ED patients with abdominal and back pain. We demonstrated nonwhite patients were significantly less likely to receive analgesia and specifically opiate analgesia than white patients when presenting to the ED for these complaints despite higher pain scores. Our results are similar to prior studies of ED patients presenting with musculoskeletal pain and long bone fractures [9,10,17] and to a large group of patients from the NHAMCS showing white patients were more likely to receive an opiate prescription for a painrelated visit [11]. Several other studies have found no association between race or ethnicity and analgesia administration in the ED [12-14,18]. These studies were performed in patients with documented long bone fractures. This patient population differs from ours as many patients who present with abdominal pain or back pain do not have a definite etiology or diagnosis for their complaint. Diagnoses with less objective findings, such as abdominal pain or back pain, may require more communication and testing to reach a diagnosis and treatment plan. This may lead to an unconscious racial bias by the provider when deciding to administer analgesia. Potential confounders in analgesia administration may be age, sex, presenting complaint, triage class, admission, and severe pain. In our study, white patients were less likely than the nonwhites to be female. Female sex has been shown in a prior study to be independently associated with oligoanalgesia in a group of ED patients with acute abdominal pain [4]. However, when potential confounders, including sex, were adjusted for in our study, white patients were still more likely to receive opiates than nonwhites. Prior studies concerning the administration of analgesia for acute abdominal pain have also suggested that providers may withhold analgesia out of concern that it may alter the abdominal physical examination and therefore diagnostic accuracy [19,20]. Even when adjusting for presenting complaint, white patients were still more likely than nonwhites to receive opiate analgesia. This discrepancy did not persist when examining receipt of any analgesia or nonopiate analgesia. In patients who received analgesia, our study demonstrated there was no difference in time to any analgesia between groups, although nonwhite patients waited, on average, 8 minutes longer for their opiate medication than white patients, which was statistically significant. Most of our patients (69%) presented with abdominal pain, and this group may require more diagnostic testing in their evaluation than
755 patients who present with other complaints, including long bone fractures. Patients who require more diagnostic evaluation have been shown to wait longer for their analgesia [21]. We demonstrate a significant association between race and oligoanalgesia in 2 urban EDs with nonwhite patients less likely to receive analgesia, and specifically opiates, for the treatment of abdominal pain or back pain. These results were independent of other potential risk factors for oligoanalgesia. Efforts to improve the quality of analgesia administration, including standardized protocols, may help to ameliorate this disparity.
5. Limitations As this was a retrospective study, potential reasons for failure to treat or delays in analgesic treatment could not be studied. Only the initial triage pain score was documented, and reassessments of pain score and patient desire for analgesia were not available. As pain may vary over time, we did not account for whether the evolution of pain influenced the administration of analgesia. We did not assess other potential agents used for analgesia in abdominal pain (eg, aluminum/magnesium hydroxide, H2-receptor antagonists, proton pump inhibitors) or low back pain patients (eg, muscle relaxants). We also did not include discharge prescriptions for analgesics in this study. As the nurse manually enters the time of medication administration into the electronic record, there may have been some discrepancy between this recorded time and the actual time of administration. Finally, our study was conducted in 2 diverse settings, but because both were urban centers, our results may not be generalizable to other practice settings.
6. Conclusion After controlling for potential confounders, nonwhite patients who presented to the ED for abdominal or back pain were less likely than white patients to receive analgesia and waited longer for their opiate medication.
References [1] Friedland LR, Pancioli AM, Duncan KM. Pediatric emergency department analgesic practice. Pediatr Emerg Care 1997;13:103-6. [2] Jones JS, Johnson K, McNinch M. Age as a risk factor for inadequate emergency department analgesia. Am J Emerg Med 1996;14:157-60. [3] Alexander J, Manno M. Underuse of analgesia in very young pediatric patients with isolated painful injuries. Ann Emerg Med 2003;41: 617-22. [4] Chen EH, Shofer FS, Dean AJ, Hollander JE, Baxt WG, et al. Gender disparity in analgesic treatment of emergency department patients with acute abdominal pain. Acad Emerg Med 2008;15:414-8.
756 [5] Raftery KA, Smith-Coggins R, Chen AH. Gender-associated differences in emergency department pain management. Ann Emerg Med 1995;26:414-21. [6] Heins A, Grammas M, Heins JK, Costello MW, Huang K, Mishra S. Determinants of variation in analgesic and opioid prescribing practice in an emergency department. J Opioid Manag 2006;2: 335-40. [7] Hostetler MA, Auinger P, Szilagyi PG. Parenteral analgesic and sedative use among ED patients in the United States: combined results from the National Hospital Ambulatory Medical Care Survey (NHAMCS) 1992-1997. Am J Emerg Med 2002;20:139-43. [8] Miner J, Biros MH, Trainor A, Hubbard D, Beltram M. Patient and physician perceptions as risk factors for oligoanalgesia: a prospective observational study of the relief of pain in the emergency department. [see comment]Acad Emerg Med 2006;13:140-6. [9] Todd KH, Deaton C, D'Adamo AP, Goe L. Ethnicity and analgesic practice [see comment]. Ann Emerg Med 2000;35:11-6. [10] Todd KH, Samaroo N, Hoffman JR. Ethnicity as a risk factor for inadequate emergency department analgesia. JAMA 1993;269:1537-9 [see comment]. [11] Pletcher MJ, Kertesz SG, Kohn MA, Gonzales R. Trends in opioid prescribing by race/ethnicity for patients seeking care in US emergency departments. JAMA 2008;299:70-8. [12] Fuentes EF, Kohn MA, Neighbor ML. Lack of association between patient ethnicity or race and fracture analgesia. Acad Emerg Med 2002;9:910-5.
A.M. Mills et al. [13] Yen K, Kim M, Stremski ES, Gorelick MH. Effect of ethnicity and race on the use of pain medications in children with long bone fractures in the emergency department. Ann Emerg Med 2003;42:41-7. [14] Quazi S, Eberhart M, Jacoby J, Heller M. Are racial disparities in ED analgesia improving? Evidence from a national database. Am J Emerg Med 2008;26:462-4. [15] McCaig LF, Nawar EW. National Hospital Ambulatory Medical Care Survey: 2004 emergency department summary. Adv Data 2006:1-29. [16] Lumley T, Kronmal R, Ma S. Relative risk regression in medical research: models, contrasts, estimators, and algorithms. UW Biostatistics Working Paper Series Working Paper; 2006. p. 293. [17] Heins JK, Heins A, Grammas M, Costello M, Huang K, Mishra S. Disparities in analgesia and opioid prescribing practices for patients with musculoskeletal pain in the emergency department. J Emerg Nurs 2006;32:219-24. [18] Choi DM, Yate P, Coats T, Kalinda P, Paul EA. Ethnicity and prescription of analgesia in an accident and emergency department: cross-sectional study. BMJ 2000;320:980-1. [19] Rupp T, Delaney KA. Inadequate analgesia in emergency medicine. Ann Emerg Med 2004;43:494-503. [20] Wolfe JM, Lein DY, Lenkoski K, Smithline HA. Analgesic administration to patients with an acute abdomen: a survey of emergency medicine physicians. Am J Emerg Med 2000;18:250-3. [21] Kelly AM, Brumby C, Barnes C. Nurse-initiated, titrated intravenous opioid analgesia reduces time to analgesia for selected painful conditions. CJEM 2005;7:149-54.
www.elsevier.com/locate/ajem
Original Contribution
Racial disparity in analgesic treatment for ED patients with abdominal or back pain Angela M. Mills MD a,⁎, Frances S. Shofer PhD a , Ann K. Boulis PhD b , Daniel N. Holena MD c , Stephanie B. Abbuhl MD a a
Department of Emergency Medicine, University of Pennsylvania, Philadelphia, PA 19104, USA Department of Sociology, University of Pennsylvania, Philadelphia, PA 19104, USA c Division of Traumatology and Surgical Critical Care, Department of Surgery, University of Pennsylvania, Philadelphia, PA 19104, USA b
Received 6 January 2010; revised 5 February 2010; accepted 24 February 2010
Abstract Objective: Research on how race affects access to analgesia in the emergency department (ED) has yielded conflicting results. We assessed whether patient race affects analgesia administration for patients presenting with back or abdominal pain. Methods: This is a retrospective cohort study of adults who presented to 2 urban EDs with back or abdominal pain for a 4-year period. To assess differences in analgesia administration and time to analgesia between races, Fisher exact and Wilcoxon rank sum test were used, respectively. Relative risk regression was used to adjust for potential confounders. Results: Of 20 125 patients included (mean age, 42 years; 64% female; 75% black; mean pain score, 7.5), 6218 (31%) had back pain and 13 907 (69%) abdominal pain. Overall, 12 109 patients (60%) received any analgesia and 8475 (42%) received opiates. Comparing nonwhite (77 %) to white patients (23%), nonwhites were more likely to report severe pain (pain score, 9-10) (42% vs 36%; P b .0001) yet less likely to receive any analgesia (59% vs 66%; P b .0001) and less likely to receive an opiate (39% vs 51%; P b .0001). After controlling for age, sex, presenting complaint, triage class, admission, and severe pain, white patients were still 10% more likely to receive opiates (relative risk, 1.10; 95% confidence interval, 1.06-1.13). Of patients who received analgesia, nonwhites waited longer for opiate analgesia (median time, 98 vs 90 minutes; P = .004). Conclusions: After controlling for potential confounders, nonwhite patients who presented to the ED for abdominal or back pain were less likely than whites to receive analgesia and waited longer for their opiate medication. © 2011 Elsevier Inc. All rights reserved.
1. Introduction
⁎ Corresponding author. Tel.: +1 215 662 6698. E-mail address: [email protected] (A.M. Mills). 0735-6757/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.ajem.2010.02.023
Acute pain is one of the most common reasons for presenting to an emergency department (ED). The administration of analgesia for acute pain has been studied in the emergency setting, specifically the topic of oligoanalgesia, or the undertreatment of pain. Oligoanalgesia has been well
Race and analgesia treatment documented in the emergency medicine literature and attributed to a variety of factors. These include patient factors such as age [1-3], sex [4,5], and race [6-10]. With respect to race and ethnicity, there have been conflicting results. Two studies of ED patients with long bone fractures showed white patients were more likely to receive analgesics [9,10]. A recent study of a large group of patients from the National Hospital Ambulatory Medical Care Survey (NHAMCS) showed white patients were more likely than nonwhites to receive an opioid prescription for a pain-related visit [11]. On the contrary, 2 studies of analgesia in long bone fractures in both adults [12] and children [13] showed no difference in analgesia administration. Another recent study analyzed analgesic treatment of headache and long bone fractures also from the NHAMCS and demonstrated an improvement in racial inequalities over time and no difference in the treatment of long bone fractures with respect to race in the later study period of 2000 to 2003 [14]. Acute abdominal pain and acute back pain have been shown to be 2 of the most common specific reasons for seeking emergency care [15]. To our knowledge, there have been no large studies examining the effect of race on the administration of analgesia for abdominal or back pain in an emergency setting. We sought to study the effect of race on a large cohort of more than 20 000 patients presenting to the ED with abdominal or back pain. We hypothesized that nonwhite patients would be less likely to receive analgesia for the treatment of these complaints.
2. Methods 2.1. Study design and setting We performed a retrospective cohort study of adult ED patients who presented with a chief complaint of abdominal pain or back pain. The study was performed at 2 urban EDs with 57 000 and 35 000 annual ED visits, respectively. The Institutional Committee on Research Involving Human Subjects at the University of Pennsylvania approved the study.
2.2. Selection of participants Patients were identified using a computerized medical record search of EMTRAC (University of Pennsylvania, Philadelphia, Pa), a computerized charting and order entry system. All adult ED patients, 18 years and older, who presented between July 2003 and February 2007 to triage with a chief complaint of abdominal pain or back pain were included. We excluded patients who left without being seen, were pregnant, were missing race or ethnicity on the medical record, or who had no documented pain score or a triage pain score of “0.” Patients who identified themselves as
753 “white Hispanic” were included in the white group, and those who identified themselves as “black Hispanic” were included in the nonwhite group.
2.3. Methods of measurement A database was created from EMTRAC, a fully electronic emergency medical record, including patient demographics (age, sex, race/ethnicity), triage class (4-point scale [1-4] from emergent [1] to nonemergent [4]), triage pain score, pain medication type and time of administration, and final ED disposition. Triage pain scores were documented using a 10-point scale (1-10), from “no pain” (1) to “worst pain of my life” (10). Reassessments of pain were not available for this analysis. Time to analgesia was defined as time from patient placement in the treatment area (ED room or hallway) to time of analgesia administration as documented in the electronic record. Analgesics were defined as any pain medication administered in the ED including “to go” medications, which almost always consisted of 4 tablets of acetaminophen with oxycodone handed to patients about the time of discharge. Opiate analgesia was defined as any oral, intramuscular, or intravenous medication administered in the ED as documented in the electronic record (eg, acetaminophen with oxycodone, acetaminophen with codeine, morphine, hydromorphone). Nonopiate analgesia included any oral, intramuscular, or intravenous nonnarcotic medication (eg, acetaminophen, ibuprofen, ketorolac). The main outcome measures were analgesic administration and time to analgesic treatment.
2.4. Data analysis The Fisher exact test was used to test for differences between race and analgesia administration. To assess differences between races with regard to time to analgesic treatment between races, Wilcoxon rank sum test was used. To calculate the likelihood of receiving analgesia, an opiate, or receiving analgesia within an hour of room placement, a generalized linear model with a log link, Gaussian error, and robust estimates of the standard errors of the model coefficients was used [16]. These models were controlled for age, sex, severe pain, triage classification, admission status, and chief complaint of back or abdominal pain. Severe pain was defined as a pain score of 9 or 10. This list of confounders was determined a priori, and there were no stepwise techniques used to select variables. Descriptive data are presented as means with SD for continuous data, medians with interquartile ranges for time variables, and frequencies and percentages for categorical data. For the general linear model, data are presented as relative risks with 95% confidence intervals. The generalized linear models were performed using Stata statistical software (Version 10; Stata Corporation, College Station, Tex). All other analyses were performed using SAS statistical software (Version 9.1; SAS
754
A.M. Mills et al.
Institute, Cary, NC). A probability of less than .05 was considered statistically significant.
Table 2
Characteristic
Whites, n = 4681 (23%)
Nonwhites, n = 15 444 (77%)
P
3. Results
Age, y (mean ± SD) Sex Female Male Presenting complaint Abdominal pain Back pain Triage class 1 (emergent) 2 3 4 (nonemergent) Pain score (mean ± SD) 0-8 9-10 Received any analgesia Opiate Nonopiate Disposition Admission Discharge
43 ± 17
42 ± 17
.0008 b.0001
2612 (56) 2069 (44)
10 296 (67) 5148 (33)
3339 (71) 1342 (29)
10 568 (68) 4876 (32)
130 (3) 1902 (41) 2136 (46) 508 (11) 7.2 ± 2.4 3002 (64) 1679 (36) 3067 (66) 2382 (51) 685 (15)
124 (1) 2953 (19) 9639 (62) 2724 (18) 7.5 ± 2.3 9006 (58) 6438 (42) 9042 (59) 6093 (39) 2949 (19)
1582 (20) 3099 (80)
2714 (37) 12 730 (63)
We identified 20 125 patients with abdominal pain or back pain during the study period. Table 1 outlines general patient characteristics. Overall, the group was young (mean age, 42 years), more often female (64%) and black (75%), with most patients presenting with abdominal pain (69%). The mean triage pain score for the group was 7.5. Overall, 12 109 patients (60%) received any analgesia and 8475 (42%) received opiates. Comparing nonwhite (77%) to white patients (23%), nonwhites were more likely to be female (67% vs 56%; P b .0001), less likely to be admitted (63% vs 80%; P b .0001), and more likely to report severe pain (pain score, 9-10; 42% vs 36%; P b .0001). With respect to analgesia, nonwhite patients were less likely to receive any analgesia (59% vs 66%; P b .0001) and less likely to receive an opiate (39% vs 51%; P b .0001). Table 2 describes the racial differences in characteristics. Several potential confounders associated with analgesic treatment were identified: age, sex, chief
Table 1
Overall general patient characteristics
Characteristic
Overall, n = 20 125 (%)
Age, y (mean ± SD) Race/ethnicity Black Asian American Indian Black Hispanic White Hispanic White Sex Female Male Presenting complaint Abdominal pain Back pain Pain score (mean ± SD) 1-8 9-10 Triage class 1 (emergent) 2 3 4 (nonemergent) Received any analgesia Opiate Nonopiate Disposition Admission Discharge
42 ± 16.7 14 980 (74.4) 339 (1.7) 41 (0.2) 79 (0.4) 142 (0.7) 4539 (22.6) 12 908 (64) 7217 (36)
Racial differences in patient characteristics
.0002
b.0001
b.0001 b.0001 b.0001
complaint, triage class, admission, and severe pain (pain score, 9-10). After adjustment for these potential confounders, white patients were still 10% more likely than nonwhite patients to receive opiates (relative risk, 1.10; 95% confidence interval, 1.06-1.13) (Table 3). Of those patients who received analgesia, there was no difference in time to any analgesia among groups (median time, 97 vs 93 minutes; P = .10), but nonwhite patients waited longer than whites when receiving opiate analgesia (median time, 98 vs 90 minutes; P = .004). Table 3 Relative risk regression of variables associated with analgesic treatment Analgesia
Characteristic
Relative risk
13 907 (69) 6218 (31) 7.5 ± 2.4 12 008 (60) 8117 (40)
95% confidence interval
Any
254 (1) 4855 (24) 11 775 (59) 3232 (16) 12 109 (60) 8475 (42) 3634 (18)
Opiate
White Admission Pain score 9-10 Triage class Back pain Female Age N 65 White Admission Pain score 9-10 Triage class Back pain Female Age N 65
1.02 1.39 1.30 0.85 1.35 1.04 0.85 1.10 1.68 1.49 0.83 1.04 1.00 0.87
1.00 1.36 1.28 0.84 1.32 1.02 0.82 1.06 1.62 1.45 0.81 1.00 0.97 0.83
4296 (21) 15 829 (79)
1.04 1.42 1.33 0.87 1.38 1.06 0.88 1.13 1.73 1.54 0.85 1.08 1.02 0.91
Race and analgesia treatment
4. Discussion Multiple studies examining the effects of race and ethnicity on analgesia administration have been published with conflicting results. Most of these studies were conducted in patient populations with musculoskeletal complaints, specifically long bone fractures [9,10,12-14]. This study was conducted to assess whether a racial disparity exists in the treatment of 2 of the most common specific reasons for patients seeking emergency care, abdominal pain and back pain. Our study comprised a large cohort of more than 20 000 ED patients with abdominal and back pain. We demonstrated nonwhite patients were significantly less likely to receive analgesia and specifically opiate analgesia than white patients when presenting to the ED for these complaints despite higher pain scores. Our results are similar to prior studies of ED patients presenting with musculoskeletal pain and long bone fractures [9,10,17] and to a large group of patients from the NHAMCS showing white patients were more likely to receive an opiate prescription for a painrelated visit [11]. Several other studies have found no association between race or ethnicity and analgesia administration in the ED [12-14,18]. These studies were performed in patients with documented long bone fractures. This patient population differs from ours as many patients who present with abdominal pain or back pain do not have a definite etiology or diagnosis for their complaint. Diagnoses with less objective findings, such as abdominal pain or back pain, may require more communication and testing to reach a diagnosis and treatment plan. This may lead to an unconscious racial bias by the provider when deciding to administer analgesia. Potential confounders in analgesia administration may be age, sex, presenting complaint, triage class, admission, and severe pain. In our study, white patients were less likely than the nonwhites to be female. Female sex has been shown in a prior study to be independently associated with oligoanalgesia in a group of ED patients with acute abdominal pain [4]. However, when potential confounders, including sex, were adjusted for in our study, white patients were still more likely to receive opiates than nonwhites. Prior studies concerning the administration of analgesia for acute abdominal pain have also suggested that providers may withhold analgesia out of concern that it may alter the abdominal physical examination and therefore diagnostic accuracy [19,20]. Even when adjusting for presenting complaint, white patients were still more likely than nonwhites to receive opiate analgesia. This discrepancy did not persist when examining receipt of any analgesia or nonopiate analgesia. In patients who received analgesia, our study demonstrated there was no difference in time to any analgesia between groups, although nonwhite patients waited, on average, 8 minutes longer for their opiate medication than white patients, which was statistically significant. Most of our patients (69%) presented with abdominal pain, and this group may require more diagnostic testing in their evaluation than
755 patients who present with other complaints, including long bone fractures. Patients who require more diagnostic evaluation have been shown to wait longer for their analgesia [21]. We demonstrate a significant association between race and oligoanalgesia in 2 urban EDs with nonwhite patients less likely to receive analgesia, and specifically opiates, for the treatment of abdominal pain or back pain. These results were independent of other potential risk factors for oligoanalgesia. Efforts to improve the quality of analgesia administration, including standardized protocols, may help to ameliorate this disparity.
5. Limitations As this was a retrospective study, potential reasons for failure to treat or delays in analgesic treatment could not be studied. Only the initial triage pain score was documented, and reassessments of pain score and patient desire for analgesia were not available. As pain may vary over time, we did not account for whether the evolution of pain influenced the administration of analgesia. We did not assess other potential agents used for analgesia in abdominal pain (eg, aluminum/magnesium hydroxide, H2-receptor antagonists, proton pump inhibitors) or low back pain patients (eg, muscle relaxants). We also did not include discharge prescriptions for analgesics in this study. As the nurse manually enters the time of medication administration into the electronic record, there may have been some discrepancy between this recorded time and the actual time of administration. Finally, our study was conducted in 2 diverse settings, but because both were urban centers, our results may not be generalizable to other practice settings.
6. Conclusion After controlling for potential confounders, nonwhite patients who presented to the ED for abdominal or back pain were less likely than white patients to receive analgesia and waited longer for their opiate medication.
References [1] Friedland LR, Pancioli AM, Duncan KM. Pediatric emergency department analgesic practice. Pediatr Emerg Care 1997;13:103-6. [2] Jones JS, Johnson K, McNinch M. Age as a risk factor for inadequate emergency department analgesia. Am J Emerg Med 1996;14:157-60. [3] Alexander J, Manno M. Underuse of analgesia in very young pediatric patients with isolated painful injuries. Ann Emerg Med 2003;41: 617-22. [4] Chen EH, Shofer FS, Dean AJ, Hollander JE, Baxt WG, et al. Gender disparity in analgesic treatment of emergency department patients with acute abdominal pain. Acad Emerg Med 2008;15:414-8.
756 [5] Raftery KA, Smith-Coggins R, Chen AH. Gender-associated differences in emergency department pain management. Ann Emerg Med 1995;26:414-21. [6] Heins A, Grammas M, Heins JK, Costello MW, Huang K, Mishra S. Determinants of variation in analgesic and opioid prescribing practice in an emergency department. J Opioid Manag 2006;2: 335-40. [7] Hostetler MA, Auinger P, Szilagyi PG. Parenteral analgesic and sedative use among ED patients in the United States: combined results from the National Hospital Ambulatory Medical Care Survey (NHAMCS) 1992-1997. Am J Emerg Med 2002;20:139-43. [8] Miner J, Biros MH, Trainor A, Hubbard D, Beltram M. Patient and physician perceptions as risk factors for oligoanalgesia: a prospective observational study of the relief of pain in the emergency department. [see comment]Acad Emerg Med 2006;13:140-6. [9] Todd KH, Deaton C, D'Adamo AP, Goe L. Ethnicity and analgesic practice [see comment]. Ann Emerg Med 2000;35:11-6. [10] Todd KH, Samaroo N, Hoffman JR. Ethnicity as a risk factor for inadequate emergency department analgesia. JAMA 1993;269:1537-9 [see comment]. [11] Pletcher MJ, Kertesz SG, Kohn MA, Gonzales R. Trends in opioid prescribing by race/ethnicity for patients seeking care in US emergency departments. JAMA 2008;299:70-8. [12] Fuentes EF, Kohn MA, Neighbor ML. Lack of association between patient ethnicity or race and fracture analgesia. Acad Emerg Med 2002;9:910-5.
A.M. Mills et al. [13] Yen K, Kim M, Stremski ES, Gorelick MH. Effect of ethnicity and race on the use of pain medications in children with long bone fractures in the emergency department. Ann Emerg Med 2003;42:41-7. [14] Quazi S, Eberhart M, Jacoby J, Heller M. Are racial disparities in ED analgesia improving? Evidence from a national database. Am J Emerg Med 2008;26:462-4. [15] McCaig LF, Nawar EW. National Hospital Ambulatory Medical Care Survey: 2004 emergency department summary. Adv Data 2006:1-29. [16] Lumley T, Kronmal R, Ma S. Relative risk regression in medical research: models, contrasts, estimators, and algorithms. UW Biostatistics Working Paper Series Working Paper; 2006. p. 293. [17] Heins JK, Heins A, Grammas M, Costello M, Huang K, Mishra S. Disparities in analgesia and opioid prescribing practices for patients with musculoskeletal pain in the emergency department. J Emerg Nurs 2006;32:219-24. [18] Choi DM, Yate P, Coats T, Kalinda P, Paul EA. Ethnicity and prescription of analgesia in an accident and emergency department: cross-sectional study. BMJ 2000;320:980-1. [19] Rupp T, Delaney KA. Inadequate analgesia in emergency medicine. Ann Emerg Med 2004;43:494-503. [20] Wolfe JM, Lein DY, Lenkoski K, Smithline HA. Analgesic administration to patients with an acute abdomen: a survey of emergency medicine physicians. Am J Emerg Med 2000;18:250-3. [21] Kelly AM, Brumby C, Barnes C. Nurse-initiated, titrated intravenous opioid analgesia reduces time to analgesia for selected painful conditions. CJEM 2005;7:149-54.