Time from pain assessment to opioid treatment in the Danish emergency departments—A multicenter cohort study

In this retrospective study of patients treated at Danish emergency departments, the authors studied time to opioid administration in patients with moderate to severe pain. They found slightly longer times to opioid administration in elderly female patients and those treated with paracetamol. The differences were on average nonetheless rather small. Pain is one of the most common reasons for presentation in the emergency department (ED),1, 2 with up to 70% of all ED contacts being related to pain.3-5 This makes pain treatment a core component of emergency medicine, yet it is often a challenge to ensure prompt, efficient, and safe pain management in the ED.6, 7 Prior research indicates that several factors affect pain management in the ED including age, gender, and ethnicity.8 Time to analgesic administration has been previously investigated as an indicator of quality of care with results from previous studies varying significantly from 62 to 116 min.9-11 To our knowledge, no previous studies have assessed time to opioid treatment as the primary outcome in a large multicenter population. In a nationwide survey from 2020, up to 50% of patients in Danish EDs reported insufficient pain management,12 consistent with other European countries.13 Therefore, it is essential to investigate whether time to opioid treatment could be a reason for insufficient pain management, given that prior studies have predominantly concentrated on other analgesic interventions. Since the decision to administer opioids involves a higher level of clinical judgement due to potential risks, this study could provide a better understanding on how severe pain is being addressed and the timeliness of clinical decisions to administer opioids. Inadequate pain management can, besides the obvious discomfort, result in acute complications, prolonged hospital stays, demoralization, sleep deprivation, and slower wound healing.9, 14 International guidelines state that patients should receive appropriate analgesia within 20–30 min of triage.2, 9, 13 The primary aim of the study was to investigate the time from pain assessment to first opioid treatment for pain requiring opioids (NRS ≥4)13 in five Danish EDs among patients presenting with abdominal or flank pain, the most frequent presentation in the ED.15 The secondary aim was to assess associations of known risk factors (age, gender, and other analgesic treatment) that could influence time to opioid treatment initiation. This retrospective multicenter cohort study comprises patients from five Central Denmark Region (CDR) hospitals between February 3, 2019, and December 31, 2019. The five EDs have approximately 300,000 patients annually and serve 1.3 million people.16 The EDs receive patients referred by their general practitioners, out-of-hours physicians, or via prehospital services. Upon arrival at the ED, nurses triage patients using the Danish Emergency Process Triage (DEPT).16 DEPT consists of an assessment of vital parameters and chief complaint. The chief complaint is chosen from a list of fifty specific conditions, and pain assessment using an 11-point numerical rating scale (NRS) is standard of care for abdominal and flank pain.17, 18 The NRS score is a tool to evaluate pain on a scale from 0 to 10, where 0 indicates no pain and 10 indicates worst possible pain. The NRS has been validated and is reliable for assessing pain in the ED with good sensitivity.19 Within Danish EDs in CDR, a nurse-administered medication protocol is followed to commence pain management using a nurse initiated treatment guideline.20 Within the guideline, the nurse can administer the first dose of analgesics without a doctor's prescription. Nurses can administer paracetamol for mild pain (NRS <4), morphine and ibuprofen for moderate pain (NRS score 4–5), and intravenous morphine for severe pain (NRS score >5). Codeine administration is restricted to specially trained nurses. Nurses document medication administration in the electronic patient journal, pending doctor approval during patient evaluation.21 Patients were included if they presented to the ED with abdominal and flank pain as their chief complaint (DEPT card no. 33 “Abdominal and Flank Pain”). Patients with abdominal and flank pain were selected to ensure data quality, as pain assessment using NRS score was a part of the DEPT card for abdominal and flank pain. Only data from the first admission in the study period were included. Patients were stratified into opioid-treated and non-opioid-treated groups. Exclusion criteria included (1) <18 years of age, (2) missing NRS score, and (3) NRS score <4 inconclusive time to opioid treatment as per their medical record. Patients were categorized into age groups based on previous studies, and the definition of elderly patients (>65 years of age) was based on the working age of the population. Data were collected from CDR's Business Intelligence (BI) Portal, which gathers data from various sources, including electronic medical records, allowing for data collection related to patient characteristics, medical history, diagnoses, treatments, and outcomes. Extracted data from BI portal included: In this study, analgesics were defined as paracetamol (PCM), non-steroid anti-inflammatory drugs (NSAID), and opioids. For further classification of analgesics by ATC-codes, see supplementary. Although no sample size calculation was conducted, a study sample of 3351 was considered adequate based on prior published data investigating pain management in the ED.9, 10, 21 Patient characteristics were presented for all patients treated with opioids. Binomial and categorical data were presented as frequencies and percentages, and continuous data as medians and interquartile range (IQR). Distributions were assessed for normality using the Shapiro–Wilk test. Pearson Chi-square test and Wilcoxon rank-sum test were applied to test differences between strata within the study population. In addition, the Kruskal–Wallis test for non-normally distributed groups was used to compare three or more strata. In case of statistically significant differences between strata, Wilcoxon rank-sum test was conducted to assess intra-group comparisons. Differences were reported with a 95% confidence interval (CI). A p-value <.05 was considered statistically significant. All statistical analyses were conducted using Stata version 17.0 (StataCorp, TX, USA). This study was part of a larger project investigating patient flow and treatment in the five EDs in CDR (Internal Research Registry, CDR, ID: 1–16–02-326-23). In addition, the dissemination of data from medical records was approved by CDR (J.No.:1–45–70-40-21). No approval from the Ethical Scientific Committee was required according to Danish law. During the study period, there were 18,657 ED contacts with abdominal and flank pain as the chief complaint. Of these, 4802 were excluded because it was their second presentation to the ED. An additional 6559 patients were excluded due to age < 18, missing NRS score, or NRS score <4 (Figure 1). Subsequently, 7397 patients were divided into opioid-receiving and non-opioid-receiving groups. 3526 patients were treated with opioids; however, 175 of these were excluded due to inconclusive data. A total of 3351 patients were included in the primary analysis. The overall median age for patients treated with opioids was 52.5 years (IQR 37.1–68.6 years), and the majority of patients (56.1%) were female. A total of 2008 (59.9%) patients were in severe pain (NRS score 7–10), 400 patients (11.9%) received NSAIDs before opioid treatment and 643 patients (19.2%) received PCM before opioid treatment (Table 1). The median NRS score at the initial assessment was 7 (IQR 5–8), and the median time from NRS registration to administration of the first opioid was 82 min (IQR 35–192 min). Regarding the time from NRS score to opioid treatment, we found a negative correlation between NRS scores and time to treatment; patients with higher NRS scores had shorter time to opioid treatment (Figure 2). In addition, Figure 2 shows the possibility of greater certainty in opioid treatment with higher NRS scores as the IQR decreases. 47.7% of patients with NRS scores ≥4 received opioid treatment, and the median time to any analgesic treatment was 64 min (IQR 30–140) (Table 2). The comparison of median times to opioid treatment in age-stratified groups indicates the first opioid treatment occurred 79 min (IQR 35–187 min) after the first registered NRS score for the youngest age group (18–25 year-olds), 75 min (IQR 32–180 min) for 26–65 year-olds, and 97 min (IQR 44–225 min) for >65-year-olds (Table 3). The oldest age group (>65 years of age) waited 18 min longer [95% CI: 19–35 min] compared with the youngest age group (18–25 years of age) (p = .0083). There was no significant difference between age groups of 18–25 years and 26–65 years of age. In comparing gender and time to opioid treatment, females waited 15 min longer [95% CI: 5.5–24.5 min] for opioid treatment than males (p = .002). No significant difference in the median time to initiate opioid treatment was observed between patients who received NSAIDs and those who did not (p = .99). However, time to opioid treatment was twice as long for patients receiving PCM compared to those who did not (Table 3). In summary, this study included 3351 patients over an 11-month period who presented to the ED with abdominal and flank pain and an NRS score ≥4. Among included patients, the median time to opioid treatment was 82 (IQR 35–192) minutes. In addition, older patients (>65 years of age), females and patients treated with PCM had longer times to opioid treatment. Treatment with NSAIDs did not significantly impact time to opioid administration. The median time to opioid treatment found in this study is almost three times greater than international recommendations of 20 to 30 min.2, 9, 13 This is substantially higher than what patients deem to be an acceptable wait time for pain medication.8 This study also found a median time to any analgesic treatment of 64 min [IQR: 30–140], consistent with findings from previous studies ranging from 62 to 116 min.9-11 The variation in median time to analgesic treatment from previous studies could be attributed to different factors including overcrowded or understaffed EDs and patients who initially decline analgesic treatment.5, 7, 22 These factors could also affect time to opioid treatment. In the present study, data were collected over a period of 11 months and therefore we would not expect overcrowding or understaffing to have had a significant impact on the results. The median time from NRS score to opioid treatment may also be influenced by ED staff and management practices. One study found pain management was often overlooked in EDs, not being a primary priority or focus in education and training, which could explain delays in opioid treatment.23 Additionally, nurses may underestimate patient pain, relying on their own judgment, leading to further delays in opioid treatment.5, 24 In our study, the median time to opioid treatment was longer than for any analgesic treatment, possibly due to hesitancy in administering stronger pain medications amid the opioid epidemic.25 It should be noted this study did not investigate the effects of analgesic treatment on pain resolution; therefore, we were unable to determine whether one treatment was more effective than the other in the resolution of acute pain. While opioids are indicated for severe acute pain, many acute pain conditions can be managed most effectively with non-opioid therapy.26 The difference between time to any analgesic treatment and time to opioid treatment, likely reflects appropriate clinical decisions to maximize the use of non-opioid options prior to opioid administration to ensure opioid treatment is warranted. This study also adds to the existing literature by examining the influence of age, gender, and other analgesic treatments on the time to opioid treatment. Significant age differences were observed in time to opioid treatment, with patients in the oldest age group (>65 years) experiencing longer durations compared with patients in the youngest age group (18–25 years), consistent with results from other studies where >65 years waited between 19 and 66 min longer than younger patients.27, 28 Possible explanations could be due to older patients refusing opioids to avoid masking pain before undergoing assessment by a physician, cognitive impairment such as dementia or delirium, and generational differences.29-32 In addition, clinicians may hesitate to administer opioids due to contraindications or possible drug interactions with other pharmacological treatments patients may already be receiving for existing medical conditions.33 In this study, females were also more likely to experience longer time to opioid treatment, compared with males. Although mostly consistent with previous studies,15, 34 some studies have demonstrated no differences between gender and time to analgesic treatment in the ED; however, this study examined odds of receiving analgesic treatment and not time to analgesic treatment.35 Our results also demonstrated that 19.2% of patients who received opioid treatment first received PCM and 11.9% first received NSAIDs, prior to opioid treatment. This might indicate that some patients did not have sufficient effect of non-opioid analgesic treatment and did therefore require opioid treatment afterwards. In addition, we identified a negative correlation between a higher pain scores and ´time to opioid treatment, suggesting that more severe pain lead resulted in faster opioid treatment, which is consistent with previous research.7 The IQR becomes narrower with higher NRS scores which could indicate that clinicians were more willing to administer opioids to patients with higher pain intensity scores. As a secondary outcome, this study found that 47.7% of patients with NRS scores ≥4 received opioid treatment, aligning with previous findings.11, 35, 36 A potential explanation for this could be patients' refusal of opioid treatment, which is consistent with other findings from this study. We found that 65.1% of non-opioid-treated patients were in moderate pain compared with 40.1% in the opioid-treated group. This is consistent with prior studies that found 50% of patients in pain declined analgesic treatment due to bearable pain and that the NRS score in the group declining analgesics was lower than the group who accepted analgesic treatment.7, 37 The 47.7% rate of opioid treatment may also be attributed to the preference for non-opioid analgesia due to the opioid epidemic. Physicians and patients may favor non-opioid options to limit opioid use as suggested in recent pain management protocols which favors non-pharmacologic, non-opioid, and multimodal therapies.26 Such interventions can reduce opioid consumption in the ED. Research shows that non-opioid alternatives can maintain patient satisfaction while decreasing opioid administration.38 The study has some limitations. This study focused on patients treated with opioids, excluding those managed only with non-opioid analgesia, and therefore, the findings are specific to opioid-treated patients. Since the primary objective of this study was to examine time to first opioid analgesic administration and protocol adherence, treatment effectiveness was not investigated, and no follow-up was conducted. In addition, our patient sample only included patients who received opioid analgesics; non-opioid therapeutic interventions may have provided sufficient analgesia. Time to opioid treatment was defined as the time from NRS score registration to administration of opioids. The dates and times were collected from electronic medical records, which may be subject to uncertainties including delays in registrations. However, the study has a large sample size across multiple EDs with data collected throughout 11 months. Due to the retrospective nature of the study, we did not control for confounding variables and only assessed unadjusted associations between risk factors and outcomes. Therefore, we cannot rule out confounding factors and explanations for differences in age and gender are speculative. Our study may be used to reflect on possible age and gender differences in opioid analgesic treatment in the ED. However, our study sample was limited to abdominal and flank pain patients, and the results may not be generalizable to patients presenting with other pain conditions in the ED. Nonetheless, this study provides important contributions to the literature which could be used for further investigation including the effect of time to opioid treatment on length of stay or strategies to decrease time to opioid treatment in the ED. This study examined the time from pain assessment to first opioid treatment for opioid requiring pain and found a median time to opioid treatment was 82 min. In addition, older patients (>65 years of age), females and patients treated with PCM had longer times to opioid treatment. Treatment with NSAIDs did not significantly impact time to opioid administration. These findings suggest a potential for optimizing opioid treatment in the ED, prompting further studies to explore alternative methods for reducing the time to opioid administration. It could also be of interest to investigate reasons for not receiving opioid treatment and whether these patients were sufficiently treated with non-opioid therapies. Katrine Harkes Gull and Stine Fjendbo Galili conceived the idea for the study, which all the authors planned. Katrine Harkes Gull acquired the data and conducted the analyses. Sara Viskum Leth provided statistical advice. Katrine Harkes Gull is responsible for the integrity and accuracy of the data analysis. All authors were responsible for the interpretation and reporting of data and results. Katrine Harkes Gull drafted the manuscript, and all authors contributed substantially to its revision. Katrine Harkes Gull holds responsibility for the paper as a whole. The authors would like to express their sincere gratitude to Jacob Redder from Central Denmark Region's Business Intelligence Office for facilitating data collection, without whom this research would not have been possible. This project received no financial funding, and the authors declare no conflicts of interest. No financial support or external funding has been received for the development of this scoping review. The data that support the findings of this study are available upon request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions. Data S1. Table S1. The classification of analgesics utilised in this study, categorised into paracetamol, NSAID, and opioids, each delineated by their respective ATC codes. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.