Article, Emergency Medicine

ED triage pain protocol reduces time to receiving analgesics in patients with painful conditions

a b s t r a c t

Background: Studies suggest that collaborative nursing protocols initiated in triage improve emergency depart- ment (ED) throughput and decrease time to treatment.

Objective: The objective of the study is to determine if an ED triage pain protocol improves time to provision of analgesics.

Methods: Retrospective data abstracted via electronic medical record of patients at a safety net facility with 67 000 annual adult visits. Patients older than 18 years who presented to the ED between March 1, 2011, and May 31, 2013, with 1 of 6 conditions were included: back pain, Dental pain, extremity trauma, sore throat, ear pain, or pain from an abscess. A 3-month orientation to an ED nurse-initiated pain protocol began on March 1, 2012. Nurses administered oral analgesics per protocol, beginning with acetaminophen or ibuprofen and progressing to oxycodone. Preimplementation and postimplementation analyses examined differences in time to analgesics. Multivariable analysis modeled time to analgesics as a function of patient factors.

Results: Over a 27-month period, 23 409 patients were included: 13 112 received pain medications and 10 297

did not. A total of 12 240 (52%) were male, 12 578 (54%) were African American, and 7953 (34%) were white, with a mean (SD) age of 39 years (13 years). The pain protocol was used in 1002 patients. There was a significant change in mean time (minutes) to provision of analgesics between preimplementation (238) and postimplementation (168) (P b .0001). Linear regression showed the protocol-delivered medications to younger patients and of lower acuity in a reduced time. Variables not related to time to provision of medication included sex, payer, and race.

Conclusion: Emergency department triage pain protocol decreased time to provision of pain medications and did so without respect to payer category, sex, or race.

(C) 2016

  1. Introduction

Pain, due to a wide variety of underlying etiologies, is one of the most common patient complaints representing up to 80% of emergency department (ED) visits. Recent studies suggest that greater than 50% of these patients reported experiencing pain that was moderate or severe [1,2]. Lack of adequate pain control and ED wait times have been recent- ly reported in Press Ganey as 2 of the most common patient complaints [3]. The Joint Commission has placed an emphasis on pain management,

? Abstract and moderated poster presentation: Society of Academic Emergency Medi- cine Annual Meeting, Dallas, TX, May 16, 2014.

* Corresponding author at: 981150 Nebraska Medical Center, Omaha, NE 68198-1150. Tel.: +1 402 559 9994; fax: +1 402 559 9659.

E-mail address: [email protected] (A.N. Barksdale).

1 Previously at Truman Medical Center/UMKC School of Medicine.

and in a recent report on this topic, stated “at the most fundamental level, improving pain management is simply the right thing to do” [4]. Because of persistent overcrowding of EDs, decreasing patient wait times and length of stay has been a continuously challenging problem for institutions to overcome. On the other hand, adequate and timely pain control is an area of patient care that can be improved. Early studies addressing this issue suggested that the implementation of mandated assessment of pain using standard pain scales had decreased the time to pain relief; however, recent reports have shown that greater than 50% of patients presenting with moderate to severe pain wait on aver- age more than 60 minutes before having their pain assessed [5-8].

In addition to ED crowding, other factors such as language barriers, fe- male sex, insurance status, and age (old or young) have been reported as patient characteristics that delay analgesia [9-11]. Recent systematic re- views assessing ED pain management have failed to identify any strong ev- idence on Effective interventions to support widespread adoption. However, the most encouraging intervention to date is the implementation

http://dx.doi.org/10.1016/j.ajem.2016.08.051

0735-6757/(C) 2016

Figure. Emergency department triage pain protocol.

of nurse-initiated pain protocols, specifically those that incorporate the re- assessment of patient’s pain every 30 to 60 minutes [12,13].

The primary objective of this study was to review the time to provi- sion of analgesics in patients presenting to the ED before and after the

implementation of a nurse-driven triage pain protocol. A secondary ob- jective was to assess if there are differences among those who received pain medication in the postimplementation phase as a function of age, race, sex, language, and presenting painful complaint.

Table 1

Proportion of patients in Diagnostic categories from postimplementation phase who re- ceived vs did not receive analgesics under pain protocol

Table 2

Mean time (SD) in minutes to provision of pain medications by phase

Pain protocol (n = 853) No protocol (n = 9769)

Phase Preimplementation

(n = 5298)

Orientation

(n = 1511)

Postimplementation

(n = 6303)

Time (min) 238.0 (153.5)a 198.9 (132.8)b 168.0 (127.8)c

diagnostic category

Abscess

33 (3.9%)

1431 (14.6%)

Back pain

127 (14.9%)

3637 (37.2%)

Dental pain

116 (13.6%)

1944 (19.9%)

Ear pain

24 (2.8%)

662 (6.8%)

Extremity trauma

36 (4.2%)

1003 (10.3%)

Sore throat

34 (4.0%)

1263 (12.9%)

Other

483 (56.6%)

Sex

Male

474 (55.6%)

5060 (51.8%)

Race

White

313 (36.7%)

3549 (36.3%)

Black

473 (55.5%)

5438 (55.7%)

Other

67 (7.9%)

782 (8.0%)

Language

English

819 (96.0%)

9347 (95.7%)

Spanish

21 (2.5%)

273 (2.8%)

Other

13 (1.5%)

149 (1.5%)

Payer category Commercial

44 (5.2%)

422 (4.3%)

Medicaid

183 (21.5%)

2167 (22.2%)

Medicare

81 (9.5%)

929 (9.5%)

Self-pay

500 (58.6%)

5757 (58.9%)

Other/unknown

45 (5.3%)

494 (5.1%)

Means with different superscript letters are significantly different (P = .0001).

  1. Methods

This was a retrospective cross-sectional observational study evaluat- ing the time to provision of analgesics in the preimplementation and postimplementation phases of a nurse-driven ED triage pain protocol. This study was approved by the local institutional review board.

This study was conducted at an urban safety net, level 1 trauma cen- ter with an emergency medicine residency program and approximately 67 000 annual adult patient visits. The ED is completely staffed by American Board of Emergency Medicine-certified (or prepared) physi- cians and uses an electronic medical record (EMR) for all Order entry and documentation. All patients 18 years or older who presented to the ED between March 1, 2011 and May 31, 2013, with 1 of the follow- ing 6 complaints were included in the study: back pain, dental pain, ex- tremity trauma, sore throat, ear pain, or pain due to an abscess.

A protocol was developed that allowed nursing staff to administer pain medications to patients with 1 of the 6 listed complaints while in triage and/or after being placed in a room, without a direCT order from a physician. The protocol was built into the EMR and formatted in a stepwise fashion beginning with ibuprofen or acetaminophen and progressed to oxycodone/acetaminophen or oxycodone alone, after re- assessment at 1 hour (Figure).

The protocol was initially introduced as an orientation phase begin- ning on March 1, 2012, through May 31, 2012, to allow for training and for the nursing staff to become familiar and comfortable with the proto- col. Official implementation was considered to be June 1, 2012.

After the 1-year postimplementation phase, a retrospective chart re-

view of the EMR was conducted, and data were recorded for the time from triage to receiving pain medication for all patients who had pre- sented with one of the diagnoses over the 27-month period of the study. Additional descriptive data abstracted from the EMR included age, race, sex, language, triage acuity, payer status, and disposition category.

The primary objective was to determine if statistically significant dif- ferences occurred in time to provision of pain medications between preimplementation and postimplementation phases, March 1, 2011, to February 29, 2012, vs June 1, 2012, to May 31, 2013. A secondary analy- sis explored whether there were significant differences among those who received pain medication in the postimplementation phase when

comparing the following variables: age, race, sex, language, payer status, triage acuity, and presenting painful complaint.

Descriptive statistics (frequencies) were used to characterize those individuals who received the pain protocol vs those who did not. To test whether there was a statistically significant difference in the time to provision of pain medications among the 3 phases of the study (preimplementation, orientation, and postimplementation), distribu- tional characteristics in the data were explored to determine if paramet- ric statistics were appropriate for this comparison. Although data were slightly skewed, the distributions met the underlying assumptions for parametric analysis, and data were analyzed using analysis of variance, with least significant difference for pairwise comparisons. In the postimplementation phase, comparison of time to provision of pain medication was compared between those patients who were treated with the new protocol and those not receiving the protocol using an in- dependent t test. Linear regression was used to model time to provision of pain medication as a function of pain protocol while controlling for patient-level characteristics using multiple linear regression. Lastly, dis- position of patient as a function of pain protocol was assessed using Fisher exact test.

  1. Results

Over a 27-month period, 23 409 patients were included in the study, 13 112 received pain medications, and 10 297 did not. Of these subjects 12 240 (52%) were male, 12 578 (54%) were African American, and 7953 (34%) were white, with a mean (SD) age of 39 years (13 years). The pain protocol was used in 1002 patients, in which 149 occurred during the orientation phase. A total of 483 patients received pain medications per protocol for entities other than the 6 originally described. In the postimplementation phase, there was no statistical significant differ- ence in those who received analgesics under the pain protocol vs those outside of the protocol in regard to diagnostic category, sex, race, language, or payer status (Table 1). Over all, there was a significant difference in mean time (minutes) to provision of analgesics among the preimplementation, orientation, and postimplementation phases, (P b

.0001) (Table 2). Pairwise comparisons among these 3 phases demon- strated a statistically significant decreasing gradient in time across the phases. In addition, in the postimplementation phase, pain protocol pa- tients received medications significantly sooner (minutes) than those for whom the protocol was not used (115 vs 176; P = .0001) (Table 3). A predictive model was fitted using linear regression evaluating the outcome of time to provision of pain meds (minutes), of whether the pain protocol was used. In this model, the overall average predicted time to receiving analgesics was 173.4 minutes. When the protocol was used, this average time decreased by 59.5 minutes (P = .0001). For each increasing year of age, the average time increased by a half a minute (P = .0001) and decreased 7.2 minutes for each level of lower triage acuity (P = .01). There were no statistical differences in regard to race, sex, or payer category; however, these variables were retained in the model to adjust for their respective effects (Table 4). In patients receiving analgesics per the protocol, there was a significant difference in disposition; however, the proportion leaving against medical advice (AMA) and leaving without being seen (LWBS) had comparable rates. There were slight differences in the proportion admitted and

discharged, although these were not Clinically meaningful (Table 5).

Table 3

Postimplementation phase–pain protocol vs no protocol mean time (SD) in minutes to provision of pain medications

Protocol used

Mean time (SD)

n

Yes

115.5 (120.4)?

853

No

176.1 (127.0)?

5463

* P = .0001.

  1. Discussion

Our ED is a busy, safety net facility that is regularly dealing with pa- tient overcrowding and boarding issues, with an average wait time of nearly 4 hours and a rate for leaving without being seen that is persis- tently greater than 10%. The purpose of this study was not an attempt to improve patient flow but rather to implement a process to decrease the time to receiving analgesics in patients presenting with pain, and ideally, without bias in regard to any specific patient characteristics. In comparing the preimplementation and postimplementation phases, pa- tients presenting with 1 of the 6 described painful entities received pain medications on average 70 minutes faster, regardless if the protocol was used or not. The mostly likely explanation is that as a result of this pro- cess, nursing staff adopted an increased awareness towards assessing patients’ pain and administering medications in more a more timely fashion. Specifically, in evaluating the postimplementation phase, our data displayed a 34% (60 minutes) reduction in time to receiving anal- gesics when the pain protocol was used. Both of these measurements were statistically significant improvements. In addition, our study sug- gested that this was achieved without any bias in regard to race, sex, or payer category. A potential concern by the study team before initiat- ing this study was that patients receiving pain medications per the tri- age pain protocol might display increased AMA and LWBS rates, but our data did not suggest any significant difference in those numbers.

Pain will likely continue to be the most common presenting com-

plaint to EDs. A study evaluating patients’ expectations for pain relief re- ported that 72% presenting with pain and 74% without, expected some relief regardless of their age or sex, and up to 18% expected complete resolution of their pain [14]. In another study, 23% of patients reported that their pain was never addressed and of those who were, only 14% ac- quired some pain relief [9]. Previously published systematic reviews have consistently reported similar barriers to assessing and managing patients’ pain in a timely fashion. The most common was not only ED crowding and Patient boarding but also language barriers, insurance status, women, and the elderly [6,10-12]. As previously mentioned, our data did not display any difference in regards to race, sex, or payer status, but we did see a direct relationship between increased age and a delay in administration of pain medications. Our likely explanation is that our nursing staff in general was uncomfortable administering anal- gesics to the older population without a physician order, which has been a consistent theme throughout EDs.

A recently published best practice standards for pain management in the ED suggested the goal of assessment and Administration of analgesia within 30 minutes of arrival. Although we did not meet that parameter, we did significantly reduce our average time, which was consistent with several other studies that had incorporated a pain or nurse/triage initi- atED analgesia protocol. Fosnocht and Swanson [15] conducted a similar

Table 4

Linear model: effect of variables on time to provision of pain medication

Time (min) to medications B SE P

Constant 173.44

Pain protocol (yes)

-59.56

4.70

.0001

Age (y)

0.50

0.13

.0001

Triage acuity

-7.22

2.78

.01

Model adjusted for race, sex, and payer category.

Table 5

Patient disposition as a function of pain protocol

Disposition category Pain protocol, n (%) No pain protocol, n (%)

Admitted

51 (6.0%)

328 (3.4%)

LWBS

113 (13.2%)

1321 (13.5%)

Transferred/behavior health ED

9 (1.1%)

64 (0.7%)

Discharged

675 (79.1%)

7976 (81.6%)

AMA

4 (0.5%)

70 (0.7%)

Unknown

1 (0.1%)

10 (0.1%)

Total

853

9769

Fisher exact test, P = .005.

before-and-after observational study that used a triage pain protocol for patients presenting with traumatic extremity or back pain. They report- ed a reduction in time to medication of 36 minutes, with 70% of patients receiving analgesics compared to 45% before implementation of their protocol. In another study looking specifically at renal colic, a formalized pain protocol reduced the time to analgesia from 72 to 37 minutes (49%) [4]. Two other studies that analyzed the implementation of nurse-initiated pain protocols published statistically significant de- creases in time to analgesia, with median reductions of 26 and 70 mi- nutes [16,17].

Recent systematic reviews have suggested that there is a lack of well-controlled, evidence-based studies to provide clear-cut recom- mendations in solving the issue of poor pain management in the ED [10,12,13]. That being said, our study in conjunction with others discussed, suggests that the implementation of triage and nurse- driven pain protocols is successful and can significantly reduce the time to patients receiving analgesia.

  1. Limitations

First, this study was retrospective in nature and conducted at a sin- gle, safety net institution. Although the results showed a strong associ- ation between implementation of a triage pain protocol and reduction in time to receiving medications, we cannot claim true cause and effect; therefore, our findings may not extrapolate to other similar EDs.

Second, we did not control for variations in ED patient volume. A full

department and patient boarding may or may not have influenced pro- tocol use. Our suspicion is that during busier periods with increased wait times, nurses would be more apt to use the protocol, but we did collect data to evaluate this.

Another potential variable that we did not control for was the pa- tients’ pain score. Initial assessment and decision to use the protocol was left solely to the discretion of the nurse who triaged the patient. We did not compare whether patients reporting higher degrees of pain received analgesia more often or more quickly. The protocol did have a component that required reassessment of pain at 1 hour, with the option to administer additional oxycodone if the patient still report- ed a pain score of 5 or greater.

Finally, we did have 483 subjects who received medication per the protocol, whose painful complaint was not 1 of the 6 entities listed. Re- moving this data would likely affect our results, but we felt that their in- clusion actually supported our findings and that implementation of a nurse-driven triage pain protocol can hastened the administration of pain medications.

  1. Conclusion

This study demonstrated that implementation of an ED triage pain protocol can significantly decrease the time to provision of pain medica- tions in certain painful conditions and did so without respect to payer category, sex, race, or effect on patient disposition.

References

  1. Fry M, Bennetts S, Huckson S. An Australian audit of ED pain management patterns. J

Emerg Nurs 2011;37:269-74.

  1. Venkat A, Fromm C, Isaacs E, Ibarra J, Committee SE. An ethical framework for the man- agement of pain in the emergency department. Acad Emerg Med 2013;20:716-23.
  2. Seguin D. A nurse-initiated pain management advanced triage protocol for ED patients with an extremity injury at a level I trauma center. J Emerg Nurs 2004;30:330-5.
  3. Steinberg PL, Nangia AK, Curtis K. A standardized pain management protocol im- proves timeliness of analgesia among emergency department patients with renal colic. Qual Manag Health Care 2011;20:30-6.
  4. Nelson BP, Cohen D, Lander O, Crawford N, Viccellio AW, Singer AJ. Mandated pain scales improve frequency of ED analgesic administration. Am J Emerg Med 2004; 22:582-5.
  5. Pines JM, Hollander JE. Emergency department crowding is associated with poor care for patients with severe pain. Ann Emerg Med 2008;51:1-5.
  6. Shaban RZ, Holzhauser K, Gillespie K, Huckson S, Bennetts S. Characteristics of effec- tive interventions supporting quality pain management in Australian emergency de- partments: an exploratory study. Aust Emerg Nurs J 2012;15:23-30.
  7. Gueant S, Taleb A, Borel-Kuhner J, Cauterman M, Raphael M, Nathan G, et al. Quality of pain management in the emergency department: results of a multicentre pro- spective study. Eur J Anaesthesiol 2011;28:97-105.
  8. Dale J, Bjornsen LP. Assessment of pain in a Norwegian emergency department. Scand J Trauma Resusc Emerg Med 2015;23:86.
  9. Stang AS, Straus SE, Crotts J, Johnson DW, Guttmann A. Quality indicators for high acuity pediatric conditions. Pediatrics 2013;132:752-62.
  10. Mills AM, Shofer FS, Chen EH, Hollander JE, Pines JM. The association between emer- gency department crowding and analgesia administration in acute abdominal pain patients. Acad Emerg Med 2009;16:603-8.
  11. Hatherley C, Jennings N, Cross R. Time to analgesia and pain score documentation best practice standards for the emergency department–a literature review. Aust Emerg Nurs J 2016;19(1):26-36.
  12. Sampson FC, Goodacre SW, O’Cathain A. Interventions to improve the management of pain in emergency departments: systematic review and narrative synthesis. Emerg Med J 2014;31:e9-18.
  13. Fosnocht DE, Heaps ND, Swanson ER. Patient expectations for pain relief in the ED. Am J Emerg Med 2004;22:286-8.
  14. Fosnocht DE, Swanson ER. Use of a triage pain protocol in the ED. Am J Emerg Med 2007;25:791-3.
  15. 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.
  16. Finn JC, Rae A, Gibson N, Swift R, Watters T, Jacobs IG. Reducing time to analgesia in the emergency department using a nurse-initiated pain protocol: a before-and-after study. Contemp Nurse 2012;43:29-37.

Leave a Reply

Your email address will not be published. Required fields are marked *