- Question: What is the incidence of oligoanalgesia in a national physician-based helicopter emergency medical system (p-HEMS) for patients presenting with an initial Glasgow Coma Scale (GCS) score of ≥8 in the out-of-hospital setting, and what factors have a clinically relevant association?
- Findings: We identified statistically significant and clinically relevant factors, such as low or moderate pain on the scene, low-rated severity of the emergency, and distinctive major complaints (eg, gynecological/obstetric), associated with oligoanalgesia.
- Meaning: Although we observed sufficient analgesia in 81.6% of all patients in our study population, 17.1% of all patients did not receive any type of pain therapy, suggesting scope for further improvement in prehospital pain therapy.
OLIGOANALGESIA IN PREHOSPITAL EMERGENCY MEDICINE
Pain is a common but subjective and highly variable symptom of both traumatic and nontraumatic conditions and is an indication for analgesic treatment.1–5 Oligoanalgesia is commonly defined as “underuse of analgesics in the face of valid indication” or “inadequate treatment of pain.” Oberholzer et al6 reported 38% of patients with oligoanalgesia in a physician-staffed helicopter emergency medical service (p-HEMS) setting in 2011. Other working groups have reported comparable results, confirming that pain is a frequently encountered problem in out-of-hospital medicine.2,7 Treating potentially lethal conditions is the first priority in the out-of-hospital setting, but early relief of pain should also be a priority.4,8,9 Out-of-hospital pain therapy is included in several national and international guidelines. Insufficient out-of-hospital pain management is often caused by a complete lack of analgesic use.6,7,10,11 Well-established training formats, such as Prehospital Trauma Life Support, do not address pain relief.8 Several factors predictive of insufficient pain management have been identified; these include high initial pain scores, inexperienced emergency physicians, and the physician’s gender.6,7,12,13
Prehospital emergency medicine in Germany is organized by the federal state. A paramedic-based system is augmented by a physician-based ground Emergency Medical Service (EMS).14 Beginning in 1970, an additional p-HEMS system has been implemented, which covers all of Germany.14 Regional dispatch centers decide which type of vehicle or helicopter should be dispatched.15,16 P-HEMS responds not only to trauma cases but also to all types of medical emergencies. Wherever applicable, they respond in combination with an additional ground EMS ambulance, the so-called “rendezvous system.”14,17 Notably, the p-HEMS physician decides whether the patient is best transported via air or ground, depending on tactical circumstances, as well as the type and severity of the particular emergency. It is also possible to hand the patient over to another EMS provider, for example, the ambulance.18 Typically, but not always, paramedics are employed by an EMS provider (eg, fire department, German Red Cross, or Allgemeiner Deutscher Automobil-Club [ADAC] Air Ambulance). In contrast, physicians engaged in patient care in the prehospital setting are mostly employed by hospitals and work as anesthesiologists, surgeons, or specialists in internal medicine.15,18
Herein, we describe the evolution in the practice of out-of-hospital pain therapy for conscious patients in a p-HEMS setting over a 12-year period. The primary outcome was the incidence of oligoanalgesia and associated factors, including the patients’ and missions’ characteristics, type of analgesic management, and type of emergency. The secondary outcomes were characteristics of subgroups of special interest and trends in the analgesics used over the 12-year study period.
This study was approved by the ethics committee of the University of Ulm, Germany (No. 03/16) and was registered in the German Clinical Trials Register (No. DRKS00015035). It complied with the Helsinki Declaration (October 2013). The need for obtaining written patient consent was waived by the institutional review board, as the study strictly utilized anonymized patient data, the p-HEMS base, and the individual p-HEMS team. The STrengthening the Reporting of OBservational studies in Epidemiology (STROBE) recommendations were applied with an additional focus on study design.19,20
Study Design and Setting
This was a secondary analysis of routine data, including all patients treated by p-HEMS crews of the German ADAC Air Ambulance, which is part of the General German Automobile Club.21
Data were included for patients involved in primary missions (out-of-hospital emergencies) between January 1, 2005 and December 31, 2017, who had on-scene pain with a numeric rating scale (NRS) score ≥4. Data were excluded for cases with a National Advisory Committee for Aeronautics (NACA) score ≥VI, cardiopulmonary resuscitation (CPR) at any time, or a Glasgow Coma Scale (GCS) score on the scene <8 (Figure 1).
The ADAC Air Ambulance is a p-HEMS organization with 35 bases throughout Germany.21 The p-HEMS crew consists of a helicopter pilot, a flight paramedic, and a specially trained HEMS physician. The physicians (mostly board-certified anesthesiologists or surgeons) are required to attend a 2-year emergency medicine training program. Pharmacological pain treatment included strong opioids (fentanyl, sufentanil), moderate opioids (morphine, piritramide), weak opioids (tramadol), ketamine (ketamine or esketamine), and nonopioids (metamizole, paracetamol, hyoscine butylbromide, acetylsalicylic acid). Nonpharmaceutical pain treatment included positioning, splinting, and immobilization. In Germany, physicians have broad therapeutic freedom.22 In addition, although medical national guidelines regarding pain therapy (eg, acute coronary syndrome, severe trauma) are available, the decision to follow these medical guidelines is made at the individual physician’s discretion.23 Furthermore, the ADAC Air Ambulance conducts a yearly quality assurance for all individual p-HEMS bases, but the organization itself provides no standard protocols for pharmacological interventions or for nonpharmacological measures, such as splinting, traction, or immobilization.
The 11-point NRS score ranges from 0 (no pain at all) to 10 (worst pain imaginable).10,24 The NRS was used to evaluate all patients (including children, sedated patients, and patients with dementia) by the treating p-HEMS physician, on arrival and shortly before handover. The time point of “at handover/hospital admission” is defined as the last score evaluated by the p-HEMS physician before they hand the patient over to the admitting physician in an emergency department or to another EMS provider. A priori pain was defined as NRS score ≥4, moderate pain was defined as NRS score 4–7, and severe pain was defined as NRS score ≥8. Sufficient analgesia was defined as NRS score at handover/hospital admission ≤3, evaluated by the p-HEMS physician. Because safe titration of analgesics to an NRS score <4 may take up to 30 minutes, a pain reduction of ≥3 NRS points (ΔNRS = NRSon scene − NRSat handover, ΔNRS ≥ 3) was also considered sufficient because of the short prehospital treatment time.25 All cases without sufficient analgesia as defined in this section were considered to have oligoanalgesia (primary outcome).
The following subgroups were defined a priori:
- bradypnea: respiratory rate (RR) ≤7 min−1,
- tachypnea: RR ≥31 min−1,
- arterial hypotension: systolic blood pressure (SBP) ≤79 mm Hg,
- arterial hypertension: SBP ≥181 mm Hg,
- bradycardia: heart rate (HR) ≤44 min−1,
- tachycardia: HR ≥121 min−1,
- impaired consciousness: GCS ≤12 points,
- hypoxia: oxygen saturation (Spo2) ≤89%, and
- patients’ major complaints, coded by the treating physicians (central nervous system, cardiovascular, pulmonary, and thoracic illness; abdominal illness; psychiatric illness; metabolic disorders; gynecology/obstetrics; trauma [including traumatic brain injury]; and other).
Secondary outcomes were characteristics of subgroups of special interest and trends in the analgesics used over the 12-year study period.
Data Processing, Analysis, and Bias
The ADAC Air Ambulance mission database contains data on every mission. To reduce recall bias, records were completed in parallel with the patients’ treatment using digital Pen and Paper technology.26,27 Patients’ demographics and characteristics, the emergency specification (eg, the NACA score), the coded major complaint, and the analgesics given were extracted and anonymized from this ADAC Air Ambulance database. Participating p-HEMS stations are listed in Supplemental Digital Content, Table 1, Figure 1, http://links.lww.com/AA/C881. No further follow-up examination was performed for study purposes.
Descriptive statistics were calculated, and the frequency and percentages were determined for nominal and ordinal variables. The median with interquartile range (IQR) was used to present discrete data. The mean and standard deviation (SD) were calculated for numerical variables. Although the GCS and NRS scores are ordinal data with discrete values, they were treated as numerical data, because their entire scales were used, and the sample size was large. For ordinal data, mean, SD, median, and IQR were calculated; the 95% confidence intervals (CIs) were calculated for percentages of nominal data. The odds ratio (OR) with 95% CI was calculated where appropriate.
Numerical data were compared across subgroups using t tests, and dichotomous data were compared using χ2 tests. A multivariable binary logistic regression analysis was performed to explore potential associations between various patient-related, emergency-related, and treatment-related factors and oligoanalgesia. As this is an explorative analysis, no adjustments for multiplicity were applied. Listwise deletion was used to handle missing data, and 86,632 cases had complete information.28 Standardized residuals did not reveal outliers, and there was a linear relationship between the logit and the continuous independent variables. Variance inflation factors were calculated to assess multicollinearity among independent variables, and all were <5. Nevertheless, Nagelkerke R2 was calculated. Two-sided P values <.05 were considered statistically significant. Linear regression analyses were performed with year as a continuous variable to verify the increase or decrease in administration of the different treatments over the years. A significant P value for the influential variable “year” means that the increase or decrease of the percentage use over time in patients was statistically significant for the particular treatment. Figures of measured vital signs (eg, RR, HR, Spo2, GCS) were rounded toward the nearest integer, due to their assumed measuring accuracy.
All analyses were calculated using R statistical software Version 3.5.1 (R Foundation for Statistical Computing, Vienna, Austria; http://www.R-project.org/) with Harrell Miscellaneous (Hmisc version 4.1.1) and Functions for Medical Statistics Book with some Demographic Data (fmsb: version 0.6.3; https://cran.r-project.org).
A sample size calculation was not performed for this secondary data analysis. Therefore, we used all data available in the ADAC Air Ambulance database. A significant result of testing the null hypothesis, as well as differences in the 95% CIs of the observed variables, may refer to clinical relevance but cannot provide definitive proof thereof. Whether a result is of clinical relevance can only be determined by clinical judgment.29 Thus, clinical judgment must always be used.
For the period from 2005 to 2017, 108,400 patients were documented in the ADAC Air Ambulance database (master dataset). Due to clinical reasons or missing GCS, NACA, electrocardiogram (ECG), or NRS information, 1670 patients were excluded, resulting in a study dataset of 106,730 patients (Figure 1). The average ratio of missing data of all variables examined was 3.6%. Supplemental Digital Content, Table 2, http://links.lww.com/AA/C881, offers a detailed overview, including all variables and subgroups examined.
Patients’ overall data are summarized in Table 1: 82.9% of patients received some type of analgesic therapy on scene (ΔNRS of 4.4 ± 2.6 [median = 4; IQR = 3]); and 79.1% of all patients received analgesic drugs (ΔNRS of 4.4 ± 2.5 [median = 4; IQR = 3]). In most cases, opioids (67.1%; 95% CI, 66.9–67.4) were used: strong opioids, 37.1% (95% CI, 36.8–37.4); moderate opioids, 30.6% (95% CI, 30.3–30.9); weak opioids, 0.2% (95% CI, 0.2–0.2); ketamine, 15.8% (95% CI, 15.6–16.1); and nonopioid analgesics, 5.6% (95% CI, 5.4–5.7). If medical pain therapy was initiated, the physicians applied a median of 1 drug for analgesia (minimum = 1, maximum = 4). Of all patients, 34.7% received pharmacological and nonpharmacological pain therapy (ΔNRS of 4.7 ± 2.6 [median = 5; IQR = 4]). Patients receiving neither pharmacological pain treatment nor nonpharmacological treatments (17.1%) had a ΔNRS of 2.5 ± 2.3 (median = 2; IQR = 4).
On handover, 81.6% (95% CI, 81.4–81.9) had received sufficient pain therapy. Further, 77.2% (95% CI, 77.0–77.5) of all patients had an NRS score ≤3 at handover. Pain reduction of ≥3 NRS points was achieved in 73.0% (95% CI, 72.8–73.3) until handover.
In 18.4% (95% CI, 18.1–18.6) of all patients, neither of these criteria was met; these individuals were rated as having oligoanalgesia (Figure 1). Further descriptive statistics are shown in Table 2. Patients with oligoanalgesia had an NRS of 5.6 ± 1.6 (median = 5; IQR = 3) at handover in comparison to patients with sufficient pain therapy (1.5 ± 1.3; median = 2; IQR = 2). The rates of moderate pain (84.1% vs 5.4%, P value <.001) and severe pain (15.9% vs 0.0%, P value <.001) at handover also differed.
Multivariable binary logistic regression analysis revealed the association of several factors on oligoanalgesia in the out-of-hospital phase (Table 3). Nagelkerke R2 for the model was 0.11.
Significant associations of male patients (OR = 1.06 [95% CI, 1.02–1.11]), major central nervous system complaints (OR = 1.57 [95% CI, 1.43–1.73]), and gynecology/obstetrics complaints (OR = 1.88 [95% CI, 1.47–2.40]) with trauma as reference group and NACA scores II (OR = 1.93 [95% CI, 1.69–2.19]), III (OR = 1.44 [95% CI, 1.33–1.55]), and IV (OR = 1.49 [95% CI, 1.38–1.60]), with NACA V as reference group, as well as of patients who were not transported by p-HEMS teams, with oligoanalgesia were observed. Patients who received any type of pharmacological pain therapy had an initial NRS score of 6.5 ± 1.7 (median = 6; IQR = 3), and those for whom no pharmacological pain therapy was initiated had an on-scene NRS score of 5.4 ± 1.6 (median = 5; IQR = 3). Higher on-scene NRS scores were related to decreased odds for oligoanalgesia (OR = 0.87 [95% CI, 0.86–0.88] for every NRS point >4).
Administration of strong opioids, moderate opioids, ketamine, or nonopioid analgesics had a significant association with oligoanalgesia in comparison to no administration of that specific analgesic (Table 3). The same applied for nonpharmaceutical measures. Only administration of weak opioids did not show a reduction or an increase in the observed oligoanalgesia.
Characteristics of Subgroups of Special Interest.
As presented in Figure 1 and Table 1, 20.9% (n = 22,313) of all patients did not receive any type of pharmacological pain therapy. They had an initial NRS score of 5.4 ± 1.6 (median = 5; IQR = 2). Moreover, 12.8% (n = 2860) of these patients were rated to have severe pain. Out-of-hospital treatment itself, including provider–patient interaction, reduced the NRS by 2.5 ± 2.4 (median = 2; IQR = 4) points. At handover, 64.5% (n = 14,393) of the patients received sufficient pain therapy per definition. The pain severity was rated, at handover, as moderate in 35.7% (n = 7955) and as severe in 2.5% (n = 565). In this subgroup without pharmaceutical pain therapy, 18.3% (n = 4084) received nonpharmaceutical interventions, such as traction, splinting, or immobilization.
The type of transport used, as a mission characteristic, is at the discretion of the treating p-HEMS physician. A clinically relevant difference in ORs in terms of the occurrence of oligoanalgesia was observed (Table 3). Patients’ data of these 3 subgroups are summarized in Supplemental Digital Content, Table 3, http://links.lww.com/AA/C881. The subgroup of patients handed over to another EMS provider after initiating emergency treatment on scene comprised 21.6% of the study dataset (n = 22,585). In this subgroup, we observed an initial NRS of 6.2 ± 1.8 (median = 6; IQR = 3); 26.2% (n = 5912) of these patients had severe pain on arrival. Moreover, 71.5% (n = 16,138) of all patients handed over to another EMS provider received some type of analgesic therapy on scene (ΔNRS of 4.0 ± 2.5 [median = 4; IQR = 4]); 68.1% (n = 15,383) of this subgroup received analgesic drugs (ΔNRS of 4.1 ± 2.5 [median = 4; IQR = 3]); and 20.9% (n = 4714) of the subgroup received pharmacological and nonpharmacological pain therapy (ΔNRS of 4.4 ± 2.5 [median = 5; IQR = 3]). Patients who received neither pharmacological pain treatment nor nonpharmacological treatments (28.5%; n = 6447) had a ΔNRS of 2.1 ± 2.4 (median = 2; IQR = 4). At handover to another EMS provider, 69.6% (n = 15,718) had an NRS score ≤3. Pain reduction of ≥3 NRS points was achieved in 64.8% (n = 14,639). Thus, until handover to another EMS provider, 74.3% (n = 16,782) of the subgroup received sufficient pain therapy.
Characteristics of patients with severe pain on scene, trauma as the major complaint, or cardiovascular issues as the major complaint are presented in Supplemental Digital Content, Document 1, http://links.lww.com/AA/C881.
Trends in Analgesics Used
Trends in the use of analgesic drugs from 2005 to 2017 are described in Figure 2. The counts of treatments with the different drugs were normalized to the counts of patients during the given year. We observed an increase in the incidence of strong opioid usage from 30.3% to 42.3% (P < .001). Ketamine usage (from 19.8% to 13.1%; P < .001), moderate opioid usage (from 33.0% to 26.1%; P < .001), weak opioid usage (from 0.5% to 0.1%; P < .001), and nonopioid analgesic usage (from 7.9% to 4.0%; P = .001) all decreased during the study period.
This study provided an overview of out-of-hospital pain therapy during p-HEMS missions and included >100,000 patients with an initial GCS score ≥8 and moderate-to-severe pain treated by physicians from 35 German p-HEMS bases from 2005 to 2017. The observed oligoanalgesia rate is among the lowest published rates.7,25,30
In our study, sufficient pain therapy was defined as NRS score ≤3 at handover (last score evaluated by the p-HEMS team) or a pain reduction of ≥3 points during the out-of-hospital treatment. The latter is not considered sufficient by other research groups.6,7,30 However, in the out-of-hospital emergency setting, we disagree with this opinion, because safe titration of pain to NRS score ≤3 may require up to 30 minutes, jeopardizing the patient or extending the prehospital treatment time, which would also be unacceptable.25,31
Compared to an NACA score of V, NACA scores of II–IV are independently associated with oligoanalgesia. Compared to NRS = 10 on scene, pain scores <10 are independently associated with oligoanalgesia, which is in contrast to the observations made by Albrecht et al7 or Oberholzer et al.6 Hypothetically, p-HEMS providers may not be attentive to pain management if the emergency does not seem to be life-threatening or if the physicians’ personal rating of the pain is that it is not severe enough to require treatment, despite existing guidelines.23 Moreover, although these psychological aspects have been addressed in a previous study,12 these hypotheses cannot be accepted or rejected based on our data. However, Neighbor et al32 also reported that trauma patients with obvious causes of pain are prescribed opioids more often in the emergency department than are those without such obvious causes. In addition, Eidenbenz et al3 reported comparable findings for trauma patients in an alpine p-HEMS setting.
In our study, the best results were achieved for trauma patients with obvious causes of pain and for patients with cardiovascular complaints, where analgesia is part of the therapy provided for reducing myocardial oxygen consumption. This may explain the observed negative association of arterial hypertension and tachycardia on arrival with oligoanalgesia (Table 3). For other nontrauma complaints, the indication for pain therapy may not be as apparent, which may lead to inferior evaluation and pain treatment.
Consistent with preexisting results on the administration of strong opioids, moderate opioids, and ketamine, an OR <1 was observed for oligoanalgesia. In this study, nonpharmacological measurements, as well as nonopioid analgesics, had been rated clinically effective for reducing the likelihood of oligoanalgesia, but the administration of weak opioids was not.
Although this study was not designed to investigate the likelihood of administration of analgesics in patients, we found, in accordance with other researchers,3,6,7 that a clinically relevant ratio of patients received no pharmacological pain therapy (20.1%) and that 17.1% of all patients did not receive any type of pain treatment. This is reported as an independent risk factor for oligoanalgesia.6,7,10,11 Interestingly, 64.5% of these patients were rated as having sufficient pain therapy at handover. Some patients may have declined analgesic therapy, but the effects of nonpharmacological measures, a good patient–provider interaction, physicians’ empathy, or other nontechnical skills,12 should be evaluated in further studies. Until these aspects are investigated, this interpretation must be rated as highly speculative, because, apart from nonpharmacological measures (OR = 0.75 [95% CI, 0.71–0.79]), none of the above speculations can be validated with our dataset.
The overall result revealed that only 18.4% patients had oligoanalgesia. Training and education of p-HEMS physicians, in combination with their daily work in a hospital as anesthesiologists, surgeons, or specialists in internal medicine, as well as the freedom to choose the appropriate treatment as per their discretion, may explain this observation.15,18,22
As described in Figure 2, we found several changes in the incidence of analgesic medication use. Thus, practices may have changed; strong opioids may now be used more often. In the international setting, increased opioid use is either considered a burden33–35 or a sign of improvement in pain therapy.36
The ADAC Air Ambulance covers nearly 45% of all p-HEMS missions in Germany (Supplemental Digital Content, Table 1, http://links.lww.com/AA/C881). However, the use of this dataset is associated with limitations, described below.
First, all results were simply observations, which are not proof of causation, and the clinical importance, even of statistically significant results, should be considered very carefully.
Second, we obtained only information anonymized for the patient, treating physician, and responsible p-HEMS base, which precluded more detailed analysis (eg, regarding regional or physician-related trends in opioid use).34
Third, all data were collected for routine documentation purposes. Moreover, as the completeness of the data was high37 (Supplemental Digital Content, Table 2, http://links.lww.com/AA/C881), accuracy and correctness could not be evaluated. Only weak conclusions can be drawn from self-reported data, such as pain scores, especially if patients were anxious or demented.2,12,38,39 Pain and the level of analgesia were assessed using an NRS score. Since patients with an impaired consciousness were included, it must be assumed that some of these patients were unable to report an NRS value. Also, 969 patients of the master dataset were excluded due to missing NRS values; at handover, some of the results were probably estimated by the p-HEMS physician. The same applies to all patients who could not verbalize their pain, such as neonates, infants, intoxicated or sedated patients, or those unable to communicate due to a language barrier.40 The fact that pain relief (NRS at handover) was evaluated by the p-HEMS physician himself and not by an uninvolved observer may have led to further bias.
Fourth, we could not gather all the desired information from the ADAC Air Ambulance database. Vital signs during the course of medical treatment are recorded every 5 minutes, but only the first measurements and those at handover were stored in this database.26 Specific reactions, such as nausea, vomiting, or dizziness, were not recorded in the dataset. Moreover, patients’ race/ethnicity is never documented in Germany.
Furthermore, only patients presenting with an initial GCS score ≥8 and moderate-to-severe on-scene pain were included. Given these inclusion criteria, we did not evaluate the analgesic treatment in general in the out-of-hospital setting. Patients presenting with an initial GCS score <8 or those who underwent CPR at any time were excluded, as were interhospital transfers, which limits the generalizability of our findings to other EMS systems. Transferring the observed results to other p-HEMS systems dealing mainly with trauma patients, without an opportunity to hand the patient over to another EMS provider (“rendezvous system”), or without the ability to transport via ground or air, should be considered with caution.
The observed rate of oligoanalgesia in this secondary data analysis was 18.4%. Nevertheless, if the factors associated with oligoanalgesia are confirmed using other study designs, the implementation of these findings should be enforced in training programs of prehospital care providers. Until then, p-HEMS providers should pay particular attention to delivering adequate (non-)pharmacological pain treatment to all patients, while also focusing on patients with low pain scores or low NACA scores.
We would like to thank Editage (www.editage.com) for English language editing and the Allgemeiner Deutscher Automobil-Club (ADAC) Air Ambulance (part of the General German Automobile Club) for their support.
Name: Matthias Helm, MD, PhD.
Contribution: This author helped acquire the data, develop and design the study, interpret the results regarding the special conditions in out-of-hospital emergency management, and read and approve the final version of the manuscript.
Conflicts of Interest: Outside the submitted work, M. Helm reports grants from the German Federal Ministry of Defense (Research Grant ID SoFo 34K3-17 1515).
Name: Bjoern Hossfeld, MD.
Contribution: This author helped interpret the results regarding the special conditions in out-of-hospital emergency management and read and approve the final version of the manuscript.
Conflicts of Interest: None.
Name: Benedikt Braun, MD.
Contribution: This author helped design the study and read and approve the final version of the manuscript.
Conflicts of Interest: None.
Name: Daniel Werner, MD.
Contribution: This author helped acquire the data and read and approve the final version of the manuscript.
Conflicts of Interest: None.
Name: Lena Peter, MSc.
Contribution: This author helped analyze the data in the revised versions of the manuscript and read and approve the final version of the manuscript.
Conflicts of Interest: None.
Name: Martin Kulla, MD, PhD.
Contribution: This author helped design the study, write the manuscript, analyze and interpret the results, perform the initial statistical analysis, and read and approve the final version of the manuscript.
Conflicts of Interest: Outside the submitted work, M. Kulla reports other support from the German Interdisciplinary Association of Critical Care and Emergency Medicine, grants from the German Federal Ministry of Defense (Research Grant ID SoFo 34K3-17 1515 and Research Grant ID SoFo 37K3-S-20 1616), other support from the Federal Joint Committee (G-BA) (Research Grant ID VSF1_2017-020), and other support from the German Federal Ministry of Education and Research (Research Grant ID 01KX1319A).
This manuscript was handled by: Richard P. Dutton, MD.
1. Berben SA, Schoonhoven L, Meijs TH, van Vugt AB, van Grunsven PM. Prevalence and relief of pain in trauma patients in emergency medical services. Clin J Pain. 2011;27:587–592.
2. Chambers JA, Guly HR. The need for better pre-hospital analgesia. Arch Emerg Med. 1993;10:187–192.
3. Eidenbenz D, Taffé P, Hugli O, Albrecht E, Pasquier M. A two-year retrospective review of the determinants of pre-hospital analgesia administration by alpine helicopter emergency medical physicians to patients with isolated limb injury. Anaesthesia. 2016;71:779–787.
4. Porter K, Morlion B, Rolfe M, Dodt C. Attributes of analgesics for emergency pain relief: results of the consensus on management of pain caused by trauma Delphi initiative. Eur J Emerg Med. 2019 [Epub ahead of print].
5. Ducharme J. Acute pain and pain control: state of the art. Ann Emerg Med. 2000;35:592–603.
6. Oberholzer N, Kaserer A, Albrecht R, et al. Factors influencing quality of pain management in a physician staffed helicopter emergency medical service. Anesth Analg. 2017;125:200–209.
7. Albrecht E, Taffe P, Yersin B, Schoettker P, Decosterd I, Hugli O. Undertreatment of acute pain (oligoanalgesia) and medical practice variation in prehospital analgesia of adult trauma patients: a 10 yr retrospective study. Br J Anaesth. 2013;110:96–106.
8. National Association of Emergency Medical Technician. PHTLS Prehospital Trauma Life Support (Military 7th Edition). 2010.Burlington, MA: Jones & Bartlett Learning.
9. Thomas SH, Shewakramani S. Prehospital trauma analgesia. J Emerg Med. 2008;35:47–57.
10. Vassiliadis J, Hitos K, Hill CT. Factors influencing prehospital and emergency department analgesia administration to patients with femoral neck fractures. Emerg Med (Fremantle). 2002;14:261–266.
11. White LJ, Cooper JD, Chambers RM, Gradisek RE. Prehospital use of analgesia for suspected extremity fractures. Prehosp Emerg Care. 2000;4:205–208.
12. Tait RC, Chibnall JT, Kalauokalani D. Provider judgments of patients in pain: seeking symptom certainty. Pain Med. 2009;10:11–34.
13. Weisse CS, Sorum PC, Dominguez RE. The influence of gender and race on physicians’ pain management decisions. J Pain. 2003;4:505–510.
14. Roessler M, Zuzan O. EMS systems in Germany. Resuscitation. 2006;68:45–49.
15. Gries A, Zink W, Bernhard M, Messelken M, Schlechtriemen T. Realistic assessment of the physician-staffed emergency services in Germany. Anaesthesist. 2006;55:1080–1086.
16. Hegenberg K, Trentzsch H, Gross S, Prückner S. Use of pre-hospital emergency medical services in urban and rural municipalities over a 10 year period: an observational study based on routinely collected dispatch data. Scand J Trauma Resusc Emerg Med. 2019;27:35.
17. Hecker N, Domres BD. The German emergency and disaster medicine and management system-history and present. Chin J Traumatol. 2018;21:64–72.
18. Moecke H. Emergency medicine in Germany. Ann Emerg Med. 1998;31:111–115.
19. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP; STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007;370:1453–1457.
20. Eisenach JC, Kheterpal S, Houle TT. Reporting of observational research in ANESTHESIOLOGY: the importance of the analysis plan. Anesthesiology. 2016;124:998–1000.
21. ADAC Luftrettung gGmbH. ADAC Luftrettung: Zahlen und Fakten 2017. Available at: https://stiftung.adac.de/luftrettung/zahlen-fakten.aspx?ComponentId=294061&SourcePageId=0
. Accessed June 20, 2018.
22. German Medical Assembly. Professional Code for Physicians in Germany - MBO-Ä 1997 - 121st German Medical Assembly 2018 in Erfurt as amended by a Resolution of the Executive Board of the German Medical Association on 14/12/2018. Available at: https://www.bundesaerztekammer.de/weitere-sprachen/english/german-medical-association/
. Accessed March 25, 2019.
23. Association of the Scientific Medical Societies in Germany. Clinical Practice Guidelines of the Association of the Scientific Medical Societies in Germany (Arbeitsgemeinschaft der Wissenschaftlichen Medizinischen Fachgesellschaften e.V., AWMF). Available at: https://www.awmf.org/en/awmf-online-portal-for-scientific-medicine/awmf-news.html
. Accessed March 28, 2019.
24. Jensen MP, Karoly P, Braver S. The measurement of clinical pain intensity: a comparison of six methods. Pain. 1986;27:117–126.
25. Park CL, Roberts DE, Aldington DJ, Moore RA. Prehospital analgesia: systematic review of evidence. J R Army Med Corps. 2010;156:295–300.
26. Helm M, Hauke J, Schlechtriemen T, Renner D, Lampl L. Primary documentation quality for paper-assisted digital mission data documentation: initial results from the air rescue service. Anaesthesist. 2009;58:24–29.
27. Moecke H, Dirks B, Friedrich H-J, et al. DIVI–Emergency medical service protocols, version 4.2. Notfall Rettungsmed. 2004;7:259–261.
28. Schafer JL, Graham JW. Missing data: our view of the state of the art. Psychol Methods. 2002;7:147–177.
29. Schober P, Bossers SM, Schwarte LA. Statistical significance versus clinical importance of observed effect sizes: what do P values and confidence intervals really represent? Anesth Analg. 2018;126:1068–1072.
30. Häske D, Böttiger BW, Bouillon B, et al. Analgesia in patients with trauma in emergency medicine. Dtsch Arztebl Int. 2017;114:785–792.
31. Pierik JG, IJzerman MJ, Gaakeer MI, et al. Pain management in the emergency chain: the use and effectiveness of pain management in patients with acute musculoskeletal pain. Pain Med. 2015;16:970–984.
32. Neighbor ML, Honner S, Kohn MA. Factors affecting emergency department opioid administration to severely injured patients. Acad Emerg Med. 2004;11:1290–1296.
33. Chenaf C, Kaboré JL, Delorme J, et al. Prescription opioid analgesic use in France: trends and impact on morbidity-mortality. Eur J Pain. 2019;23:124–134.
34. Guy GP Jr, Zhang K, Bohm MK, et al. Vital signs: changes in opioid prescribing in the United States, 2006-2015. MMWR Morb Mortal Wkly Rep. 2017;66:697–704.
35. Orpana HM, Lang JJ, Baxi M, et al. Canadian trends in opioid-related mortality and disability from opioid use disorder from 1990 to 2014 through the lens of the global burden of disease study. Health Promot Chronic Dis Prev Can. 2018;38:234–243.
36. Fang W, Liu T, Gu Z, Li Q, Luo C. Consumption trend and prescription pattern of opioid analgesics in China from 2006 to 2015. Eur J Hosp Pharm. 2019;26:140–145.
37. Bell ML, Fiero M, Horton NJ, Hsu CH. Handling missing data in RCTs: a review of the top medical journals. BMC Med Res Methodol. 2014;14:118.
38. Breivik H, Borchgrevink PC, Allen SM, et al. Assessment of pain. Br J Anaesth. 2008;101:17–24.
39. Treede RD. The international association for the study of pain definition of pain: as valid in 2018 as in 1979, but in need of regularly updated footnotes. Pain Rep. 2018;3:e643.
40. Pasero C, McCaffery M. Pasero C, McCaffery M. Section II: assessment. In: Pain Assessment and Pharmacologic Management. 2011:St Louis, MO: Elsevier/Mosby; 13–176.