Abstract
Introduction
Pain associated with sickle cell disease (SCD) causes severe complications and frequent presentation to the emergency department (ED). Patients with SCD frequently report inadequate pain treatment in the ED, resulting in hospital admission. A retrospective analysis was conducted to assess a quality improvement project to standardize ED care for patients presenting with pain associated with SCD.
Methods
A 3-year prospective quality improvement initiative was performed. Our multidisciplinary team of providers implemented an ED order set in 2019 to improve care and provide adequate analgesia management. Our primary outcome was the overall hospital admission rate for patients after the intervention. Secondary outcome measures included ED disposition, rate of return to the ED within 72 hours, ED pain scores at admission and discharge, ED treatment time, in-patient length of stay, non-opioid medication use, and opioid medication use.
Results
There was an overall 67% reduction in the hospital admission rate after implementation of the order set (P = 0.005) and a significant decrease in the percentage admission rate month over month (P = 0.047). Time to the first non-opioid analgesic decreased by 71 minutes (P > 0.001), and there was no change in time to the first opioid medication. The rate of return to the ED within 72 hours remained unchanged (7.0% vs 7.1%) (P = 0.93), and the ED elopement rate remained unchanged (1.3% vs 1.85%) (P = 0.93). After the implementation, there were significant increases in the prescribing of orally administered acetaminophen (7%), celecoxib (1.2%), and tizanidine (12.5%) and intravenous ketamine (30.5%) and ketorolac (27%). ED pain scores at discharge were unchanged for both hospital-admitted (7.12 vs 7.08) (P = 0.93) and non-admitted (5.51 vs 6.11) (P = 0.27) patients. The resulting potential cost reduction was determined to be $193,440 during the 12-month observation period, with the mean cost per visit decreasing by $792.
Conclusions
Use of a standardized and multimodal ED order set reduced hospital admission rates and the timeliness of analgesia without negatively impacting patients’ pain.
Keywords: Sickle Cell Disease, Nociplastic Pain, Admission, Emergency Department, Protocol
Introduction
Sickle cell disease (SCD) is an inherited blood disorder associated with severe complications and, in the United States, frequent emergency department (ED) visits [1]. Patients with SCD have a single-point mutation causing mutated hemoglobin proteins and rigid, sickle-shaped red blood cells [2]. SCD leads to acute pain episodes, where ischemia and hypoxia give rise to vaso-occlusive pain crises (VOCs) [3]. In addition to acute VOC events, patients with SCD commonly develop chronic and persistent pain that are characterized as neuropathic or nociplastic [4, 5].
It is estimated that 200,000 patients with SCD present to EDs in the United States each year [6]. Approximately 135,000 of these visits are due to pain crises. For many patients, the ED serves as their primary source of pain care, and patients often present to the ED multiple times per year [7, 8]. However, the ED is not the most effective location for treating the complex nature of SCD-related pain because of crowding, time constraints, minimal personnel, and lack of consistency of care [9–11].
Patients with SCD face barriers to care related to their disease, which requires more frequent ED visits for pain medications [12], as well as to social determinants of health and health care. Deficiencies in the quality of patient care for those with SCD are well documented [13, 14]. These include a disconnect that exists between ED clinicians’ perceptions and patients’ perceptions. A cross-sectional study determined that 98% of ED clinicians were “confident” in their knowledge about caring for patients with SCD. Disproportionately, only half of these clinicians’ patients reported that they were “at most, or sometimes” satisfied with their care [12]. Previous quality improvement (QI) efforts have targeted this treatment gap through nursing education about SCD and pain care for SCD, but only a temporary improvement in pain care was found [9]. However, these projects lacked a standardized and easy-to-follow multimodal pain management regimen.
With more than 100,000 patients with SCD citing pain as their primary concern when presenting to ED in the United States, improving consistency and mechanism-based treatment would be an important step forward [6]. Following consistent care plans can reduce the impact of bias and stigma and increase confidence in the clinician by lessening intra- and inter-patient care plan variability. Traditionally, clinicians in the ED have relied on opioid-centered protocols, which have led to ineffective pain control and high admission rates [9, 15, 16]. Managing SCD-related pain requires providers to understand the intricate differences between the acute and chronic pain associated with SCD. Recent data have shifted the paradigm for patients with SCD from a nociceptive (tissue injury) model of pain toward a nociplastic (tissue injury and neuropathic alterations) model [5, 17, 18]. Treating the neuropathic or nociplastic pain with opioids is less effective and can lead to opioid dependence, tolerance, and hyperalgesia [19]. Therefore, exclusive reliance on opioid analgesics is not the ideal approach. Non-opioid analgesics have been shown to effectively treat chronic pain in patients with SCD, as well as to manage acute pain episodes when avoidance of opioid-related side effects is a goal [20–23]. Prescribing non-opioids in the context of individual care plans has support in the treatment of SCD-related pain for the inpatient population [24, 25]. Therefore, implementing a standardized and multimodal ED order set could provide guidance for ED teams on SCD pain care protocols and improve care outcomes for SCD-related pain in the ED. However, the effectiveness of a standardized ED order set for adult patients with SCD has not yet been evaluated.
We developed a multimodal ED pain management order set with the goal of improving pain care for patients with SCD to reduce the need for hospital admission for more intensive pain treatment. Time to the receipt of an opioid analgesic has been used as an indirect measure of system responsiveness to a patient’s pain [26]. Researchers have reported that the time to first opioid administration was reduced from 180 minutes to 86 minutes after an individualized care plan program in a pediatric ED [26]. Compared with the previously used primary pain relief strategy in the ED, the order set implemented in this QI/QA project prioritizes the use of two rapid and short-acting analgesics, intravenous fentanyl and low-dose ketamine, as well as nonsteroidal anti-inflammatory drugs and other non-opioid analgesics [23, 24, 27]. The order set was designed to ensure that patients receive rapid and consistent analgesic care each time they present to the ED, while also supporting inter-patient consistency. In addition to patient benefits, effective pain management lowers health care costs associated with the population of patients with SCD [24, 28].
To our knowledge, multimodal analgesia order sets have not been widely implemented for adult patients with SCD-related pain presenting to the ED. Our interdisciplinary team designed and implemented a novel ED order set, which was analyzed retrospectively to determine its efficacy. The goal of this order set was to improve pain management for patients with SCD presenting to the ED for pain crises by standardizing treatment protocols and focusing on non-opioid analgesics to optimize care delivery, resulting in decreases in associated health care costs.
Methods
Measures, Outcomes, and Analysis
The primary outcome measure for the QI initiative was the rate of hospital admission from the ED. The secondary outcome measures were treatment time in the ED, discharge disposition, pain scores at admission and discharge, in-patient treatment time, in-patient length of stay (LOS), opioid and non-opioid usage, time to first opioid and first non-opioid analgesic, oral morphine milligram equivalence (MME) for opioids used, and 72-hour rate of re-presentation to ED. ED treatment time was determined by taking the difference between the time of the patient being checked into the ED and the time of the patient being discharged from the ED. In-patient treatment time was determined from the difference between hospital admission time and hospital discharge time. In-patient LOS was determined by the number of “midnights” a patient stayed [24]. MME was determined via the conversion standard of the U.S. Centers for Disease Control and Prevention [29]. The rate of return to the ED was determined as the percentage of ED visits within 72 hours of an index ED visit for each patient.
In-patient hospitalization cost was calculated consistent with standard hospital accounting principles by counting admission as day 1 and sequentially each “midnight” accumulated during a hospital stay [24, 25]. All costs were determined from data from our health system for the diagnosis-related group (DRG) relevant to SCD. Total direct cost was determined by considering both the change in hospital admissions after the QI intervention and the change in inpatient LOS. The total cost per day in hospital was determined to be $1,248 for this time interval [24].
Univariate statistical analysis was performed on the outcome variables in the pre-intervention and post-intervention data sets. Continuous data were analyzed with an independent-samples t test and analysis of variance (ANOVA). Categorical data were analyzed with Pearson’s chi-squared test. Discrete data were analyzed with a two-sample t test. Categorical data were recorded as count data and are reported as percentages. Continuous data are presented with means, standard deviations, and 95% confidence intervals. Effect size was calculated with Cohen’s d. A “before–after” approach to the implementation of this QI project was used. This approach has clear potential biases, such as time-dependent confounds, regression to the mean, the Hawthorne effect, and historical events [30]. To limit the major biases of the before–after approach, we examined the change in monthly variables per unit, and an interrupted time-series analysis was performed with autoregressive integrated moving average (ARIMA) modeling [31]. The impact of the implementation of the ED order set in May 2019 was evaluated with a t test on the dummy codes indicating pre- and post-intervention levels. To determine how this intervention has been sustained, a t test on the dummy codes was performed to evaluate changes in admission rate per month, inpatient LOS in days, and ED treatment time in hours for before and after implementation of the order set. Initial autoregressive and moving-average structures were chosen on the basis of examining the autocorrelation and partial autocorrelation of time series. The resulting model was ARIMA (1,0,0) with autoregression over one lag, no differencing, and no moving average. The model’s stability and fit were evaluated with stationary R2 and Ljung-Box statistics. The residuals were examined to ensure correct model choice. A two-tailed P<0.05 was considered statistically significant. All statistical analyses were conducted in IBM SPSS, version 24 (IBM, Armonk, NY, USA). To account for the historical event bias of the coronavirus disease 2019 (COVID-19) pandemic impact on the health system, we have excluded data from December 2019 to the present.
Context
From May 2017 to May 2020, we performed a prospective QI initiative within a large integrated academic learning health system to address the care of VOC or VOC-like pain related to SCD within the ED. The goal of the QI initiative was to address the need to provide consistent, rapid analgesia to adult patients with SCD-related pain. A standardized ED order set was customized to address these care concerns.
Intervention
A multidisciplinary team of advanced practice providers and physicians from Pain Medicine, Hospital Medicine, and Emergency Medicine came together as a working group (WG) to develop a standardized ED order set for patients with SCD-related pain. The hospital quality data demonstrated inconsistent and limited use of non-opioid pain medications during ED visits. The ED-to-hospital admission rate for patients with SCD in the United States ranges from 47% to 54% [32, 33]. WG review of hospital utilization trends showed that our health system had higher numbers for pain-related admissions of adult patients with SCD than the national average. Therefore, at the initial meeting, the WG decided to address the ED-related pain care and levels of admission for people with SCD through a QI project to improve pain care. Similar to work previously published by this team [24], a barrier analysis was conducted through the use of a modified Delphi strategy to identify the most significant barriers and develop an order set to address them [34] (Table 1). Variability in treatment approaches among ED providers, lack of pharmaceutical diversity, and lack of standardized pain order availability were identified as significant barriers that could be addressed in this initiative. The order set used a multistep approach to managing SCD-related pain (Figure 1). Each step contained specific dosing strategies for non-opioid analgesics and full-agonist opioids (i.e., intravenous fentanyl) to treat SCD-related pain. The complete order set developed by the WG is also included (Figure 2). The implementation was initiated in the ED upon approval, dissemination, and training of the ED providers and staff by the ED Medical Director and WG study team. Order sets coded into the electronic health record and direct links were supplied to ED providers. Protocols for the use of individual medications (e.g., intravenous fentanyl, intravenous ketamine) were already in use in the ED for acute and chronic pain not of SCD origin. Training on the use of order sets was provided by the ED study champion.
Table 1.
Modified Delphi barrier and facilitator process
| ||
| Barriers identified in Rounds 0 and 1 | Facilitators | Barriers and facilitators after Rounds 2 and 3 |
| Social determinants of care | Develop standard set of analgesics | Develop standard care with treatment options |
| Availability of analgesics in ED | Make non-opioid analgesics available | An order set with opioid and non-opioid analgesics was developed |
| Racial Bias | Provide education and training | Provide education |
| Provider bias in care for SCD | Involve risk management to reduce opioid-related concerns | ED providers received a one-time introduction to the QI project and education on standardized SCD-related pain care with opioids and non-opioid analgesics |
| Lack of standard pain care options | Develop SCD-related pain order set | |
| Lack of SCD care knowledge among providers | Provide care in dedicated SCD site | |
| Provide transportation to ambulatory sites | ||
| Wrong location of service for SCD care | ||
| Concern about the use of opioids | ||
| Inter- and intra-patient variability in treatment | ||
Figure 1.
Three-step ED algorithm for breakthrough pain related to SCD. PO= oral administration; IV= intravenous administration.
Figure 2.
ED SCD pain order set. PO= oral administration; IV= intravenous administration.
Retrospective Review
Data were retrospectively collected from a report that captured the International Classification of Diseases (ICD) code for patients 18 years and older presenting to the ED with a primary diagnosis code of ICD.57. Inpatient costs were calculated for the DRG 812 (Red Blood Cell Disorders without Major Complication or Comorbidity). The other primary DRG applicable to patients with SCD, DRG 811 (Red Blood Cell Disorders with Major Complication or Comorbidity Data), was not included to reduce overestimation of hospital stay due to non–pain-related conditions increasing the LOS and average daily cost, according to accepted methodology [24].
Data included times of ED presentation and departure (total ED treatment time), hospital admission, medications, pain scores, hospital discharge disposition, and patient demographics. Patients with medical complaints not primarily related to an acute painful crisis, which included acute chest syndrome and priapism, were excluded. Patients who were considered “high utilizing,” defined as having more than four admissions per year, were also excluded, as these patients were already managed through individualized care plans containing personalized order sets accessible by the same ED and health system providers included in this QI effort [24]. Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was first reported in the United States on January 20, 2020. As a result, we chose to curtail the post-intervention analysis to the dates that were unaffected by COVID-19. The retrospective analysis of this QI project met internal review board standards that did not require informed consent because of the retrospective nature of the study. Review of records was approved by the Institutional Review Board.
Results
The demographic information for the cohort is presented in Table 2. One hundred forty-one patients visited the ED with a primary diagnosis of pain related to SCD a total of 539 times over the study period. Much of the cohort was female, self-identified as Black and/or African American, and self-identified as not Hispanic or Latino. Payer methods used were predominantly Medicare (32%) and Medicaid (35%). Most patients were diagnosed with homozygous hemoglobin S (hbSS) (65%).
Table 2.
Patient demographics
| Patients Visiting the ED During Study Period | n = 141 |
|---|---|
| Median age, years (range) | 29 (18–71) |
| Sex, n (%) | |
| Men | 55 (39) |
| Women | 86 (61) |
| Race and ethnicity, n (%) | |
| Black or African American | 131 (93) |
| Hispanic or Latino | 10 (7) |
| SCD genotype, n (%) | |
| SS | 92 (65) |
| SC | 33 (23) |
| Sβthal+ | 11 (8) |
| Sβthal° | 0 (0) |
| Insurance coverage, n (%) | |
| Commercial | 26 (18) |
| Medicare | 45 (32) |
| Medicaid | 50 (35) |
| Veterans Health Administration (VA) | 0 (0) |
| None | 20 (14) |
SS = homozygous for hemoglobin S; SC = heterogenous for hemoglobin S and hemoglobin C; Sβthal+ = hemoglobin S beta plus thalassemia; Sβthal°= hemoglobin S beta zero thalassemia.
The absolute hospital admission rate of patients with SCD reduced by 13% (P = 0.0049) after the intervention. The elopement rate and the 72-hour rate of readmission to the ED did not change. No statistically significant change for time to the first opioid was found (95 minutes vs 104 minutes, P = 0.19), but time to the first non-opioid analgesic decreased significantly (207 minutes vs 136 minutes, P = 0.0003), with an effect size of 0.51. The MMEs given in the ED decreased (22 mg vs 18 mg, P = 0.0053) (Table 3), with an effect size of 0.27. No patients died during the retrospective analysis period (data not shown). The ARIMA plot of the admission rate of patients with SCD is shown in Figure 3. After the order set protocol implementation, there was a statistically significant decrease in the percentage hospital admission rate per month (t = –2.11, P = 0.047). Before the intervention, the admission rate was increasing non–statistically significantly per month (t = 1.496 P = 0.15). The implementation of the protocol resulted in a nonsignificant initial decrease in the admission rate per month (t = –0.171, P = 0.866).
Table 3.
Effect of ED protocol on patients presenting with pain related to SCD
| Outcomes | Before Protocol | After Protocol | P Value |
|---|---|---|---|
| Disposition, n (% of ED visits) | |||
| Admitted | 189 (51) | 63 (38) | 0.0049 |
| Eloped | 5 (1.3) | 3 (1.8) | 0.94 |
| Time to first opioid, minutes | |||
| N, missing | 337, 35 | 150, 17 | |
| Median (range) | 71 (16–299) | 89 (16–290) | |
| Mean (SD) | 95 (67) | 104 (66) | 0.19 |
| Time to first non-opioid analgesic, minutes | |||
| N, missing | 98, 274 | 105, 62 | |
| Median (range) | 185(5–587) | 102 (26–512) | |
| Mean (SD) | 207 (166) | 136 (104) | <0.001 |
| Morphine equivalents, mg | |||
| No. | 355 | 129 | |
| Median (range) | 19 (2–136) | 13 (2–115) | |
| Mean (SD) | 22 (16) | 18 (14) | 0.0053 |
| 72-hour return to ED, n (%) | 26 (7.0) | 12 (7.1) | 0.93 |
SD = standard deviation.
Figure 3.
ARIMA plot of admission rate per month. Red line depicts the actual admission rate. Blue line represents predicted admission rate. Black vertical line displays when the intervention took place in May of 2019.
There was statistically significant decrease in ED treatment time (t = –1.443, P = 0.001) over the period of the QI study, but the intervention resulted in a significant initial increase in ED treatment time by 2.124 (t = 0.2966, P = 0.001) (Figure 4). Hospital treatment time decreased from 3.7 days to 3.0 days (P = 0.029) after the intervention, with an effect size of 0.32 (Table 4). There was a non–statistically significant decrease in inpatient LOS (P = 0.39) after the implementation (Figure 5). The total direct cost of treating an average uncomplicated patient with SCD admitted to our health system was determined to be $1,248 per day [24]. LOS decreased from 3.65 hospital days to 3.02 days. This accounted for a potential decrease in inpatient hospital cost of $792 per admission. Decreased LOS and decreased admissions led to a total potential cost reduction that was estimated to be $193,440 (Table 4).
Figure 4.
ARIMA plot of ED LOS per month. Red line depicts the actual ED LOS. Blue line represents predicted ED LOS. Black vertical line displays when the intervention took place in May of 2019.
Table 4.
Effect of the ED order set on ED treatment time and inpatient LOS
| Outcomes | Before Protocol | After Protocol | P Value |
|---|---|---|---|
| ED treatment time, hours | |||
| Median (range) | 6.2 (0.9–16) | 5.9 (1.9–15) | |
| Mean (SD) | 6.7 (3.1) | 6.3 (2.6) | 0.19 |
| Inpatient LOS, days | |||
| Median (range) | 3.4 (0.09–12.4) | 2.6 (0.5–10.6) | |
| Mean (SD) | 3.7 (2.4) | 3.0 (2.0) | 0.029 |
| Cost of inpatient care, $ | |||
| Total $/year (admission × LOS) | 430,560.00 | 237,120.00 | |
| Cost per admission | 4,556.19 | 3,763.81 | 0.063 |
SD = Standard deviation.
Figure 5.
ARIMA plot of inpatient LOS per month. Red line depicts the actual inpatient LOS. Blue line represents predicted inpatient LOS. Black vertical line displays when the intervention took place in May of 2019.
Pain scores at ED admission (first pain) and at ED discharge (final pain) were captured for 75.8% and 65.2% of pre- and post-protocol patients, respectively (Table 5). Pain scores were subsequently subdivided into groups: patients discharged from the ED and patients admitted to the hospital. Although first pain for ED-discharged patients significantly increased after protocol implementation, the patient-reported pain score remained consistent at discharge, with a similar decrease of 3 in pain score (P = 0.27). Similarly, the pain score significantly decreased for admitted patients. Final pain scores were consistent between patients in the pre- and post-implementation groups (P = 0.93).
Table 5.
Pain scores of patients before and after implementation of ED order set
| Outcomes | Before Protocol | After Protocol | P Value |
|---|---|---|---|
| ED pain score (all patients) | |||
| N, missing | 282, 90 | 109, 58 | |
| First pain (SD) | 8.74 (1.5) | 8.61 (1.9) | 0.80 |
| Final pain (SD) | 6.59 (2.8) | 6.60 (2.6) | 0.96 |
| ED pain score (discharged patients) | |||
| Pain on ED presentation (SD) | 8.49 (1.6) | 9.33 (1.1) | 0.0042 |
| Final pain (SD) | 5.51 (2.9) | 6.11 (2.9) | 0.27 |
| ED pain score (admitted patients) | |||
| Pain on ED presentation (SD) | 8.97 (1.3) | 8.27 (2.0) | 0.014 |
| Final pain (SD) | 7.12 (26) | 7.08 (2.2) | 0.93 |
SD = Standard deviation.
Use of intravenous fentanyl increased by 38% and use of intravenous hydromorphone decreased by 24% after protocol implementation. The use of intravenous ketamine increased significantly (P < 0.001) (Table 5). No commonly reported adverse effects (including but not limited to nausea, dizziness, dysphoria, or mental status changes) [23] of ketamine were reported (data not shown). Non-opioid prescribing of protocol medications increased during the post-intervention period, except for orally administered ibuprofen (Table 6). No change was observed in patients’ creatinine levels (data not shown).
Table 6.
Medication use for pain treatment
| Medication | Before Protocol | After Protocol | P Value |
|---|---|---|---|
| Non-opioid analgesic, PO, n (%) | |||
| Acetaminophen PO | 26 (7.0) | 24 (14) | 0.006 |
| Celecoxib PO | 0 (0) | 2 (1.2) | 0.034 |
| Ibuprofen PO | 5 (1.3) | 6 (3.6) | 0.088 |
| Tizanidine PO | 2 (0.5) | 21 (13) | <0.001 |
| Non-opioid analgesic, IV, n (%) | |||
| Ketamine IV | 2 (0.5) | 53 (31) | <0.001 |
| Ketorolac IV | 65 (18) | 75 (45) | <0.001 |
| Opioid, IV, n (%) | |||
| Fentanyl IV | 8 (2.2) | 67 (40) | <0.001 |
| Hydromorphone IV | 321 (86) | 104 (62) | <0.001 |
| Morphine IV | 30 (8.1) | 12 (7.2) | 0.72 |
| Opioid, PO, n (%) | |||
| Hydromorphone PO | 17 (4.6) | 8 (4.8) | 0.91 |
| Methadone PO | 8 (2.2) | 7 (4.2) | 0.18 |
| Morphine PO | 2 (0.5) | 0 (0) | 0.34 |
| Oxycodone PO | 14 (3.8) | 7 (4.2) | 0.81 |
PO = oral administration; IV = intravenous administration.
Discussion
The implementation of a standardized ED order set designed to improve treatment of SCD-related pain led to a decrease in admission rates, a decrease in treatment time for hospital-admitted patients, an increase in the use of non-opioid analgesics, and a decrease in time to delivery of non-opioid analgesics. These outcomes demonstrate efficacy for an ED order set designed to manage acute SCD pain while not affecting patient-reported pain intensity and discharge disposition.
This QI program does not address the many social determinants of health and health care faced by patients with pain related to SCD. Training programs, as well as standardized and individualized interventions to treat pain related to SCD, have been found to reduce admission rates and shift analgesic prescribing regimens [24, 35, 36]. Within the population of pediatric patients with SCD, the hospital admission rate fell by 16% after implementation of a standardized pathway, which is similar to the decrease of 13% in the present project [35]. Interestingly, the overall hospital admission rate for the pediatric intervention and for the present project was approximately 40%. Although MME decreased slightly after the intervention, we do not believe this resulted in a clinically significant change in patient-reported pain. We postulate that the inclusion of potent non-opioid analgesics made up for the decrease in opioid administration. Similarly, prior pediatric ED interventions found a similar lack of change in pain score when non-opioid analgesics were used as adjuvants [36]. Creatinine levels in hospitalized patients were not impacted, and this suggests that the nonsteroidal anti-inflammatory drugs delivered in the ED did not have an acute negative impact on renal function. Consistent with similar ED clinical care pathways for SCD, no changes in the readmission and elopement rates were found [26, 35]. This implies that the care pathway did not negatively impact the patients’ perception of the quality of care received.
A significant decrease in time to first opioid delivery was not observed in the post-intervention analysis, with first opioid delivery occurring approximately 1.5 hours from ED presentation in both the pre-and post-implementation groups. Other researchers have reported a 1.5-hour decrease in time to first opioid, with a post-intervention time of 86 minutes [26]. Interestingly, we reported a time to first opioid of 95 minutes before the intervention, only 9 minutes slower than their post-intervention value [26]. This suggests that although our intervention did not improve the speed at which we provided opioid-based analgesia, the speed did not appear to be significantly slower than that reported by comparable studies. Additional literature on opioid timing for SCD pain care suggests that 90 minutes is standard, and this therefore reinforces that our initiative did not interfere with the speed of opioid medication prescribing [37]. A potential explanation for the slight increase in time to first opioid might be the significant changes in the medications being used in the ED, like the transition from intravenous hydromorphone to intravenous fentanyl per our protocol. Although time to first opioid has face validity and an impact on patient satisfaction, the literature has shown that the reduction in time to first opioid does not affect admission [37]. Further investigation could be needed to improve faster prescribing of analgesia in the ED.
Pairing opioids with an increase in non-opioid medications is a vital aspect of our novel ED order set and contributes to the decrease in admissions and decreased time spent as an inpatient. Prior research on children with SCD-related pain has shown that non-opioids are effective analgesic options in pain treatment and that delivering analgesics to patients with SCD faster is associated with significant improvement in pain management in the ED environment [36]. The present QI project has demonstrated similar findings in our population of adult patients with SCD. More research needs to be done on the direct effects of non-opioids in combination with traditional opioid-based medications to fully understand the role non-opioids play in pain care of the nociplastic mechanism of SCD-related pain.
A total potential reduction of inpatient cost approximated $200,000 during the post-intervention year because of decreases in inpatient LOS and hospital admissions. A comparable intervention focused on improving care of pediatric patients with SCD found a cost savings total of $325,000 [35]. Although the reduction in the present study is slightly less than that observed in the pediatric intervention, the adult ED care pathways show an ability to reduce financial burden on the health care system while not sacrificing patient care outcomes.
Limitations
Although our results demonstrate the efficacy of an ED order set to manage SCD-related pain, our project had several limitations. The COVID-19 pandemic has placed a significant burden on health care access and has fundamentally changed clinical practices. As a result, we chose not to include an additional year for the post-intervention analysis. At the time of this writing, our region continues to experience increases in COVID-19 illnesses in the community. Therefore, we plan to study the effect of COVID-19 on this patient population and QI initiative in the following year. The curtailing of the data collection as a result imposes a limitation on the generalizability of our short-term response data. This will need to be addressed in future reports on the results of the QI initiative. Additionally, we must consider the single-site scope of our project. When we compared our findings with those of several other projects, differences were highlighted. These comparisons were predominantly with the pediatric literature because of the lack of relevant publications in the adult population. Moving forward, a multisite initiative in the adult population would be a viable option to further our findings. Another limitation of our project was a lack of specific patient feedback. However, previous pediatric intervention studies were able to get patient-specific feedback, which found that patients’ perceptions were positive [38]. Only 1 year of post-intervention data analysis and continued follow-up retrospective reviews will be necessary to determine the efficacy and durability of our initiative.
Conclusion
The implementation of a standardized ED order set decreased admission rates and reduced the time to first non-opioid analgesic for patients with SCD visiting the ED with pain related to SCD. Our ED order set also increased the use of non-opioid analgesics while not affecting pain scores at discharge for both admitted and non-admitted patients. The rate of readmission to the ED within 72 hours, and the elopement rate did not change after our intervention. Our intervention provides actionable steps that can be taken to combat the inconsistent and inadequate care patients with SCD report receiving. Along with financial cost savings, improved patient outcomes indicate the efficacy of managing acute or chronic SCD pain with a standardized ED pain management order set.
Contributor Information
Anthony A Wachnik, Department of Anesthesiology and Pain Service Line.
Jena L Welch-Coltrane, Department of Anesthesiology and Pain Service Line.
Meredith C B Adams, Department of Anesthesiology and Pain Service Line.
Howard A Blumstein, Department of Emergency Medicine.
Manoj Pariyadath, Department of Emergency Medicine.
Samuel G Robinson, Department of Anesthesiology and Pain Service Line.
Amit Saha, Department of Anesthesiology and Pain Service Line.
Erik C Summers, Department of Internal Medicine Section of Hospital Medicine.
Robert W Hurley, Department of Anesthesiology and Pain Service Line; Department of Neurobiology and Anatomy, Wake Forest University School of Medicine, Winston-Salem, North Carolina, USA.
Funding sources: This work was funded by departmental resources. RWH is supported by NIH U24 NS115708, NIH R33 DA046085, NIH U01 DA057016, NIH R24 DA055306, and AHRQ R18 HS028584. MCBA is supported by NIH R24 DA055306, NIH K08 EB022631, NIH U24 NS115708, NIH R33 DA046085, and NIH U01 DA057016.
Conflicts of interest: RWH has received a research grant paid to his institution by Avanos for research unrelated to this topic. RWH is the EIC of Pain Medicine and recused himself from all aspects of editorial process regarding this work. No other authors have a conflict of interest.
Anthony A. Wachnik and Jena L. Welch-Coltrane contributed equally to this work.
References
- 1. Shah N, Bhor M, Xie L, Paulose J, Yuce H.. Sickle cell disease complications: Prevalence and resource utilization. PLoS One 2019;14(7):e0214355. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Inusa BPD, Hsu LL, Kohli N, et al. Sickle cell disease—genetics, pathophysiology, clinical presentation and treatment. Int J Neonatal Screen 2019;5(2):20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Martin OY, Thompson SM, Carroll AE, Jacob SA.. Emergency department provider survey regarding acute sickle cell pain management. J Pediatr Hematol Oncol 2020;42(6):375–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Uwaezuoke SN, Ayuk AC, Ndu IK, et al. Vaso-occlusive crisis in sickle cell disease: Current paradigm on pain management. J Pain Res 2018;11:3141–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Fitzcharles M-A, Cohen SP, Clauw DJ, et al. Nociplastic pain: Towards an understanding of prevalent pain conditions. Lancet 2021;397(10289):2098–110. [DOI] [PubMed] [Google Scholar]
- 6. Yusuf HR, Atrash HK, Grosse SD, Parker CS, Grant AM.. Emergency department visits made by patients with sickle cell disease: A descriptive study, 1999–2007. Am J Prev Med 2010;38(Suppl 4):S536–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Wolfson JA, Schrager SM, Coates TD, Kipke MD.. Sickle-cell disease in California: A population-based description of emergency department utilization. Pediatr Blood Cancer 2011;56(3):413–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Paulukonis ST, Feuchtbaum LB, Coates TD, et al. Emergency department utilization by Californians with sickle cell disease, 2005–2014. Pediatr Blood Cancer 2017;64(6):e26390. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Tanabe P, Hafner JW, Martinovich Z, Artz N.. Adult emergency department patients with sickle cell pain crisis: Results from a quality improvement learning collaborative model to improve analgesic management. Acad Emerg Med 2012;19(4):430–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Recio-Saucedo A, Pope C, Dall’Ora C, et al. Safe staffing for nursing in emergency departments: Evidence review. Emerg Med J 2015;32(11):888–94. [DOI] [PubMed] [Google Scholar]
- 11. Polevoi SK, Quinn JV, Kramer NR.. Factors associated with patients who leave without being seen. Acad Emerg Med 2005;12(3):232–6. [DOI] [PubMed] [Google Scholar]
- 12. Linton EA, Goodin DA, Hankins JS, et al. ; Sickle Cell Disease Implementation Consortium. A Survey-based needs assessment of barriers to optimal sickle cell disease care in the emergency department. Ann Emerg Med 2020;76(3s):S64–s72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Oyeku SO, Faro EZ.. Rigorous and practical quality indicators in sickle cell disease care. Hematology Am Soc Hematol Educ Program 2017;2017(1):418–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Smith LA, Oyeku SO, Homer C, Zuckerman B.. Sickle cell disease: A question of equity and quality. Pediatrics 2006;117(5):1763–70. [DOI] [PubMed] [Google Scholar]
- 15. Mager A, Pelot K, Koch K, et al. Opioid management strategy decreases admissions in high-utilizing adults with sickle cell disease. J Opioid Manag 2017;13(3):143–56. [DOI] [PubMed] [Google Scholar]
- 16. Molokie RE, Montminy C, Dionisio C, et al. Opioid doses and acute care utilization outcomes for adults with sickle cell disease: ED versus acute care unit. Am J Emerg Med 2018;36(1):88–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Darbari DS, Sheehan VA, Ballas SK.. The vaso-occlusive pain crisis in sickle cell disease: Definition, pathophysiology, and management. Eur J Haematol 2020;105(3):237–46. [DOI] [PubMed] [Google Scholar]
- 18. Field JJ, Ballas SK, Campbell CM, et al. AAAPT diagnostic criteria for acute sickle cell disease pain. J Pain 2019;20(7):746–59. [DOI] [PubMed] [Google Scholar]
- 19. Carroll CP, Lanzkron S, Haywood C Jr, et al. Chronic opioid therapy and central sensitization in sickle cell disease. Am J Prev Med 2016;51(1 Suppl 1):S69–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Brandow AM, Carroll CP, Creary S, et al. American Society of Hematology 2020 guidelines for sickle cell disease: Management of acute and chronic pain. Blood Adv 2020;4(12):2656–701. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Lubega FA, DeSilva MS, Munube D, et al. Low dose ketamine versus morphine for acute severe vaso occlusive pain in children: A randomized controlled trial. Scand J Pain 2018;18(1):19–27. [DOI] [PubMed] [Google Scholar]
- 22. Tawfic QA, Faris AS, Kausalya R.. The role of a low-dose ketamine-midazolam regimen in the management of severe painful crisis in patients with sickle cell disease. J Pain Symptom Manage 2014;47(2):334–40. [DOI] [PubMed] [Google Scholar]
- 23. Schwenk ES, Viscusi ER, Buvanendran A, et al. Consensus guidelines on the use of intravenous ketamine infusions for acute pain management from the American Society of Regional Anesthesia and Pain Medicine, the American Academy of Pain Medicine, and the American society of Anesthesiologists. Reg Anesth Pain Med 2018;43(5):456–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Welch-Coltrane JL, Wachnik AA, Adams MCB, et al. Implementation of individualized pain care plans decreases length of stay and hospital admission rates for high utilizing adults with sickle cell disease. Pain Med 2021;22(8):1743–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Balsamo L, Shabanova V, Carbonella J, et al. Improving care for sickle cell pain crisis using a multidisciplinary approach. Pediatrics 2019;143(5):e20182218. [DOI] [PubMed] [Google Scholar]
- 26. Della-Moretta S, Delatore L, Purcell M, et al. The effect of use of individualized pain plans in sickle cell patients presenting to the emergency department. Ann Emerg Med 2020;76(3s):S21–S27. [DOI] [PubMed] [Google Scholar]
- 27. Cohen SP, Bhatia A, Buvanendran A, et al. Consensus guidelines on the use of intravenous ketamine infusions for chronic pain from the American Society of Regional Anesthesia and Pain Medicine, the American Academy of Pain Medicine, and the American Society of Anesthesiologists. Reg Anesth Pain Med 2018;43(5):521–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Lee A, Chan SK, Chen PP, et al. The costs and benefits of extending the role of the acute pain service on clinical outcomes after major elective surgery. Anesth Analg 2010;111(4):1042–50. [DOI] [PubMed] [Google Scholar]
- 29. U.S. Centers for Disease Control and Prevention. Calculating Total Daily Dose of Opioids for Safer Dosage; 2020. Available at: https://www.cdc.gov/drugoverdose/pdf/calculating_total_daily_dose-a.pdf (accessed November 12, 2021).
- 30. Ho AMH, Phelan R, Mizubuti GB, et al. Bias in before-after studies: Narrative overview for anesthesiologists. Anesth Analg 2018;126(5):1755–62. [DOI] [PubMed] [Google Scholar]
- 31. Mascha EJ, Sessler DI.. Segmented regression and difference-in-difference methods: Assessing the impact of systemic changes in health care. Anesth Analg 2019;129(2):618–33. [DOI] [PubMed] [Google Scholar]
- 32. Brousseau DC, Owens PL, Mosso AL, Panepinto JA, Steiner CA.. Acute care utilization and rehospitalizations for sickle cell disease. JAMA 2010;303(13):1288–94. [DOI] [PubMed] [Google Scholar]
- 33. Cline DM, Silva S, Freiermuth CE, Thornton V, Tanabe P.. Emergency Department (ED), ED observation, day hospital, and hospital admissions for adults with sickle cell disease. West J Emerg Med 2018;19(2):311–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Fitch KB, Aguilar MD, Burnand B, et al. The RAND/UCLA Appropriateness Method User's Manual. Santa Monica, Calif: RAND; 2001. [Google Scholar]
- 35. Schefft MR, Swaffar C, Newlin J, Noda C, Sisler I.. A novel approach to reducing admissions for children with sickle cell disease in pain crisis through individualization and standardization in the emergency department. Pediatr Blood Cancer 2018;65(10):e27274. [DOI] [PubMed] [Google Scholar]
- 36. Ender KL, Krajewski JA, Babineau J, et al. Use of a clinical pathway to improve the acute management of vaso-occlusive crisis pain in pediatric sickle cell disease. Pediatr Blood Cancer 2014;61(4):693–6. [DOI] [PubMed] [Google Scholar]
- 37. Mathias MD, McCavit TL.. Timing of opioid administration as a quality indicator for pain crises in sickle cell disease. Pediatrics 2015;135(3):475–82. [DOI] [PubMed] [Google Scholar]
- 38. Krishnamurti L, Smith-Packard B, Gupta A, et al. Impact of individualized pain plan on the emergency management of children with sickle cell disease. Pediatr Blood Cancer 2014;61(10):1747–53. [DOI] [PubMed] [Google Scholar]