Abstract
Background
In our region, as in Western countries, the number of emergencies requesting an ambulance is increasing. Whether this may be due to increased disease burden and/or lower threshold for calling is not clear. This study aimed to examine the development over the years in the number and urgency levels of ambulances and the actual acute disease severity, as indicated by initial vital signs measured on scene, together with patient demographics, comorbidity, and mortality trends over time.
Methods
We conducted a registry-based observational cohort study of patients aged ≥ 18 years assigned an ambulance in the North Denmark Region from 2017 to 2023. The primary outcomes were urgency level and acute critical severity, assessed by a modified National Early Warning Score. Linear regression was applied to examine for alignment between trends in urgency level over the years and critically versus non-critically severity in prehospital patients assigned to urgency level A (most urgent, potentially life-threatening) and B (less urgent), respectively.
Results
Among 202,746 prehospital patients, the proportion assessed as urgency level A decreased from 62% to 59%, while urgency level B increased from 38% to 41%. While the proportion of patients in acute critical condition was 16% vs. 8% among level of urgency A and B, and stable during the period, the number of patients in acute critical condition increased by 3.6 (95% confidence interval (CI) 0.7–6.6) and 6.1 (95% CI 4.4–7.8) per quarter. Across the total population, patient age increased from 64 to 66 years, and the proportion with severe comorbidity increased from 11% to 14%.
Conclusion
From 2017 to 2023, the number of patients assigned with an ambulance through 1-1-2 increased, as did patient age and comorbidity. During the years were similar proportions of acute critical severity, but increasingly more among urgency level B. Thus, the acute severity among prehospital patients has not diminished. The lack of alignment between the level of urgencies and the acuity severity found on scene should be monitored carefully, together with mortality and clinical outcomes.
Supplementary Information
The online version contains supplementary material available at 10.1186/s13049-026-01646-w.
Keywords: Emergency medical dispatch, Emergency medical services, Triage, Patient acuity, Decision support systems, Early warning score
Introduction
As in the Western world, the number of patients requesting an ambulance has increased in Denmark [1–3], most likely due to the increasingly ageing population living with more comorbidities and needing urgent help. Difficulties when navigating the acute health care system may play a part [4, 5], and some studies indicate a lower threshold for calling the Emergency Medical Services (EMS) [6], implying a decrease in the severity of the EMS patients’ clinical conditions over time.
When calling the EMS for an ambulance, the call is assessed to identify the main symptom/injury and triaged to determine how fast a response is needed, and if it is needed at all. In Denmark, the highest level of urgency is A (blue lights and sirens), primarily intended for potential life-threatening or disabling conditions [7], while level B is for urgent, but not potentially life-threatening or disabling conditions. The assessment is based on telephone triage and, therefore, limited to the caller’s information and can prove complex.
In a recent Danish study on ambulance patients, the severity of the presenting condition among younger and older patients (66 years and above) was assessed by a modified National Early Warning Score 2 (NEWS2). The older patients displayed a higher median score, indicating that more patients are critically affected by their acute conditions with age, potentially resulting in more EMS calls [8].
Given the increasing demand and growing complexity of health issues among the population, as reflected in EMS call patterns, it is essential to explore how accurately telephone-based triage reflects patients’ actual clinical conditions.
As such, our aim was to examine the development over the years in the number and urgency levels of ambulances and the actual acute disease severity, as indicated by initial vital signs measured on scene, together with patient demographics, comorbidity, and mortality trends over time.
Methods
Study design and patient population
This observational, historic register-based cohort study included all patients aged 18 years or older in the North Denmark Region who received an ambulance dispatch following a national emergency number (1–1–2) call between January 1, 2017, and December 31, 2023.
Setting
The North Denmark Region has 579,787 inhabitants and consists of both urban and rural areas [9]. Calls to the national emergency number, 1-1-2, involving medical emergencies in Denmark have been forwarded by the police to an Emergency Medical Coordination Centre. Here, health care professionals evaluate the situation using the Danish Index for Emergency Care to determine the appropriate urgency level of ambulance response [10]. No significant changes have been made to this setup during the study period. The index is designed as a criteria-based decision support tool aiding health care professionals in handling EMS calls. The health care professionals are primarily nurses or paramedics with additional training specifically for handling EMS calls. The information available to the health care professionals during the call, apart from the caller’s location, is limited to what the caller provides. All information about the patient’s health care status therefore depends on communication between the healthcare professional call-taker, and the caller.
During the 1-1-2 call, the health care professionals assign one of 37 possible main symptoms or injuries according to the index, as well as determine the appropriate urgency level for the ambulance response.
There are five different urgency levels: (A) life-threatening or potentially life-threatening conditions, (B) urgent but not life-threatening conditions, (C) non-urgent conditions that require an ambulance, (D) non-urgent conditions requiring supine patient transport, and (E) conditions requiring medical advice only [7]. The Danish Index for Emergency Care is regularly audited by the Emergency Medical Services, and a new index is planned for implementation in 2026, completely revising the index used in the current study. Previous studies have assessed and found an association between urgency level and mortality [3]. For ambulances dispatched following 1-1-2 calls in the North Denmark Region, the vast majority were assigned urgency level A (49%) and B (39%) in 2023–2024 [1, 11].
Study population
The patients were identified through the ePMR. If a patient received more than one ambulance during this period, they would appear in the dataset multiple times, except in mortality analyses. Patients were excluded if the Danish Index for Emergency Care was missing, if the urgency level was classified as C, D, or E, as this study focuses on acute patients, or if the civil registration number was missing or invalid. The study’s findings are reported in accordance with the STROBE statement [12].
Data sources and variables
Data on the Danish Index for Emergency Care criteria, urgency level, and Civil Registration Number (CPR number) were extracted from the ePMR [13]. In addition, the level of acute disease severity according to the modified NEWS2 (mNEWS2) score was likewise derived from ePMR data. The mNEWS2 is based on initial vital parameters at the ambulance’s arrival (pulse, respiratory rate, Glasgow Coma Scale, blood pressure, oxygen saturation, and oxygen treatment) [14–16] (Additional file 1). In cases with two or fewer missing vital signs, the missing vital signs were considered to be normal, i.e., not contributing to an increase in severity according to the mNEWS2 score. Otherwise, cases with more than two missing vital signs were considered incomplete registrations and omitted from analysis. The mNEWS2 was stratified into four groups: normal vital signs (NEWS2 = 0), non-critical condition (NEWS2 1–6), critical condition (NEWS2 ≥ 7), and incomplete registration.
Comorbidity was assessed using a modified Charlson Comorbidity Index, based on previous diagnoses (5 years); however, we did not exclude current diagnoses associated with the emergency call [17]. The modified comorbidity was divided into three groups based on the number of comorbidities from the Charlson Comorbidity Index: no comorbidity (comorbidity 0), moderate comorbidity (comorbidity 1–2), and severe comorbidity (comorbidity ≥ 3). Data on previous diagnoses were obtained from the in-hospital electronic medical record.
In Denmark, all citizens have a unique CPR number, which includes information on the individual’s age, sex, and vital status, and is used to retrieve data on their vital status. Data on patients’ age, sex, and vital status were obtained from the Danish Civil Registration System.
Mortality was assessed up to 30 days after the final date of the study to ensure full follow-up.
Outcome
The primary outcomes were levels of urgency and acute disease severity. Patient characteristics, including age, comorbidity, and 0–1-day, 2–7-day, and 0–30-day mortality, were described.
Statistics
Descriptive analyses were performed on the patient population with the total number, median, and percentages.
The trend in urgency levels A and B during the study period (divided into quarters of each included year) was evaluated by linear regression, presented with a 95% confidence interval (CI).
For the statistical analyses of disease severity, we divided the mNEWS2 score into two groups: non-critical condition (normal vital signs and non-critical condition, mNEWS2 < 7) and critical condition (critical condition, mNEWS2 ≥ 7). As such, linear regression analyses were performed to assess trends in patient volume across acute critically and non-critically conditions assigned to urgency level A and B, respectively. P-values of < 0.05 were considered statistically significant. Incomplete registration of mNEWS2 scores (cases with more than two missing vital signs) was excluded from analysis. Meaning that any case with up to two missing vital signs was still included in the analysis and assigned to the group it fitted into, based on the vital signs that were known.
For mortality analysis, a 30-day exclusion period was applied in the 0–30-day mortality for patients with more than one emergency call. This meant that if a patient had multiple calls within a 30-day window, only the first call was included in the analysis. Poisson regression analysis adjusted for comorbidity, age and sex was performed to evaluate trends in mortality by severity groups over time.
All analyses were performed using the STATA statistical software version 18.
Results
During the seven-year study period, a total of 245,266 patient contacts resulted in an ambulance dispatch. Excluding those without a valid CPR number and those under 18 years (34,418), those without a Danish Index category (5,615), and other urgency levels than A or B (2,487) resulted in a total of 202,746 included patient contacts. In this context, each contact is referred to as a “patient,” regardless of whether the same individual had multiple contacts. The 202,746 patient contacts represented 118,740 unique individuals (Fig. 1).
Fig. 1.
Flow diagram of patient inclusion. Legend: Flow diagram of included and excluded patient contacts from January 1, 2017, to December 31, 2023
Overall, urgency level A was most frequent, accounting for 59% compared to level B (41%). For urgency level A, the proportion of patients in acute critical condition was twice as high when compared to level B (16% vs. 8%, respectively) (Table 1). The median age and the number of patients with severe comorbidity increased from 2017 to 2023 for both urgency level A and B, with a larger increase in urgency level B. (Additional file 2). The number of deaths within 0–1, 2–7, and 0–30 days was also approximately twice as high among urgency level A compared to level B (Table 1).
Table 1.
Patient characteristics by urgency level
| Urgency level A | Urgency level B | All | |
|---|---|---|---|
| N (%) | 118,816 (59) | 83,930 (41) | 202,746 (100) |
| Females n (%) | 52,952 (45) | 40,471 (48) | 93,423 (46) |
| Age, median (IQR) | 65 (48–78) | 65 (43–78) | 65 (46–78) |
| mNEWS2, n (%) | |||
| Normal | 25,246 (21) | 22,010 (26) | 47,256 (23) |
| Non-critical | 66,643 (56) | 47,937 (57) | 114,580 (57) |
| Critical | 18,973 (16) | 6500 (8) | 25,473 (13) |
| Incomplete | 7954 (7) | 7483 (9) | 15,437 (8) |
| Comorbidity n (%) | |||
| None | 61,200 (52) | 51,624 (62) | 112,824 (56) |
| Moderate | 41,761 (35) | 23,446 (28) | 65,207 (32) |
| Severe | 15,855 (13) | 8860 (11) | 24,715 (12) |
| Deaths, n (%) | |||
| 0–1 days | 5663 (5) | 1775 (2) | 7434 (4) |
| 2–7 days | 2262 (2) | 867 (1) | 3129 (2) |
| 0–30 days | 10,770 (9) | 4192 (5) | 14,962 (7) |
Outcome rates, sex, age, comorbidity and mortality for all patients (N = 202746)
Urgency level and severity of the patient’s condition
The number of patients assigned with urgency level A and B increased significantly over the study period, with 67.1 (95% CI: 43.6–90.6) and 189.7 (95% CI: 153.2–226.2) patients per quarter, respectively. The largest increase was among patients with urgency level B (Fig. 2).
Fig. 2.
Patients by urgency level (A and B, 2017–2023). Legend: Linear relationship between the number of patients within urgency level A and B over years reported as quarters
Over the seven-year study period, the proportion of patients in acute critical condition within each urgency level was stable (Urgency level A: 15–17%, Urgency level B: 7–8%) (Table 2).
Table 2.
Trends in mNEWS2 and mortality across urgency levels A and B over time
| Year | 2017 | 2018 | 2019 | 2020 | 2021 | 2022 | 2023 | Total |
|---|---|---|---|---|---|---|---|---|
| Urgency level A n = 118,828 | ||||||||
| N (%) | 16,245 (62) | 17,170 (61) | 16,090 (59) | 16,355 (61) | 17,096 (58) | 17,853 (55) | 18,007 (55) | 118,816 (59) |
| mNEWS2, n (%) | ||||||||
| Normal | 3347 (21) | 3525 (21) | 3290 (20) | 3549 (22) | 3876 (23) | 3790 (21) | 3869 (21) | 25,246 (21) |
| Non-critical | 9080 (56) | 9506 (55) | 8993 (56) | 9221 (56) | 9502 (56) | 10,086 (56) | 10,255 (57) | 66,643 (56) |
| Critical | 2494 (15) | 2818 (16) | 2684 (17) | 2447 (15) | 2640 (15) | 2901 (16) | 2989 (17) | 18,973 (16) |
| Incomplete | 1324 (8) | 1321 (8) | 1123 (7) | 1138 (7) | 1078 (6) | 1076 (6) | 894 (5) | 7954 (7) |
| Deaths, n (%) | ||||||||
| 0–1 days | 812 (5) | 908 (5) | 766 (5) | 754 (5) | 816 (5) | 848 (5) | 759 (4) | 5663 (5) |
| 2–7 days | 287 (2) | 327 (2) | 278 (2) | 303 (2) | 324 (2) | 377 (2) | 366 (2) | 2262 (2) |
| 0–30 days | 1470 (9) | 1640 (10) | 1409 (9) | 1442 (9) | 1558 (9) | 1677 (9) | 1574 (9) | 10,770 (9) |
| Urgency level Bn = 83,943 | ||||||||
|---|---|---|---|---|---|---|---|---|
| N (%) | 9819 (38) | 11,191 (39) | 11,081 (41) | 10,564 (39) | 12,451 (42) | 14,344 (45) | 14,480 (45) | 83,930 (41) |
| mNEWS2, n (%) | ||||||||
| Normal | 2675 (27) | 2998 (27) | 2895 (26) | 2890 (27) | 3338 (27) | 3550 (25) | 3664 (25) | 22,010 (26) |
| Non-critical | 5562 (57) | 6335 (57) | 6452 (58) | 6032 (57) | 7042 (57) | 8289 (58) | 8225 (57) | 47,937 (57) |
| Critical | 647 (7) | 798 (7) | 835 (8) | 776 (7) | 1003 (8) | 1225 (9) | 1216 (8) | 6500 (8) |
| Incomplete | 935 (10) | 1060 (9) | 899 (8) | 866 (8) | 1068 (9) | 1280 (9) | 1375 (10) | 7483 (9) |
| Deaths, n (%) | ||||||||
| 0–1 days | 163 (2) | 215 (2) | 223 (2) | 237 (2) | 289 (2) | 337 (2) | 311 (2) | 1775 (2) |
| 2–7 days | 77 (1) | 102 (1) | 118 (1) | 103 (1) | 118 (1) | 185 (1) | 164 (1) | 867 (1) |
| 0–30 days | 391 (4) | 506 (5) | 525 (5) | 549 (5) | 642 (5) | 804 (6) | 775 (5) | 4192 (5) |
Rates of mNEWS2 scores and mortality stratified by urgency levels A and B across all study years (N = 202746)
Evaluated by linear regression, the number of patients in acute critical condition within urgency level A increased (significant) by 3.6 (95% CI 0.7–6.6) per quarter, and for patients in acute critical condition within urgency level B (significant) by 6.1 (95% CI 4.4–7.8) per quarter (Fig. 3, X). For patients in non-critical condition among both urgency level A and B, the increase over the years was significant, with respectively 18.6 (95%CI 10.7–26.6) per quarter and 39 (95%CI 28.6–49.5) per quarter (Fig. 3, Y). For an overview of exclusions for the disease severity analysis and the mortality analysis, and for details on missing vital signs, see Additional file 4 and Additional file 5, respectively.
Fig. 3X.
Trend in acute critical condition across urgency levels A and B. Legend: Linear relationship between the increase in patients with acute critical condition (X) (n = 25,473) within urgency level A and urgency level B over the years
Fig. 3Y.
Trend in non-critical condition across urgency levels A and B. Legend: Linear relationship between the increase in patients with non-critically condition (Y) (n = 161,836) patients within urgency level A and urgency level B over the years
Mortality
Over the years, the proportion of deaths within 0–1, 2–7, and 0–30 days remained stable (Table 3).
Table 3.
Trends in mNEWS2 and mortality across the study population over time
| Year | 2017 | 2018 | 2019 | 2020 | 2021 | 2022 | 2023 | Total |
|---|---|---|---|---|---|---|---|---|
| Total | ||||||||
| N (%) | 26,064 (13) | 28,361 (14) | 27,171 (13) | 26,919 (13) | 29,547 (15) | 32,197 (16) | 32,487 (16) | 202,746 (100) |
| mNEWS2, n (%) | ||||||||
| Normal | 6022 (23) | 6523 (23) | 6185 (23) | 6439 (24) | 7214 (24) | 7340 (23) | 7533 (23) | 47,256 (23) |
| Non-critical | 14,642 (56) | 15,841 (56) | 15,445 (57) | 15,253 (57) | 16,544 (56) | 18,375 (57) | 18,480 (57) | 114,580 (57) |
| Critical | 3141 (12) | 3616 (13) | 3519 (13) | 3223 (12) | 3643 (12) | 4126 (13) | 4205 (13) | 25,473 (13) |
| Incomplete | 2259 (9) | 2381 (8) | 2022 (7) | 2004 (7) | 2146 (7) | 2356 (7) | 2269 (7) | 15,437 (8) |
| Deaths, n (%) | ||||||||
| 0–1 days | 975 (4) | 1123 (4) | 989 (4) | 991 (4) | 1105 (4) | 1185 (4) | 1070 (3) | 7438 (4) |
| 2–7 days | 364 (1) | 429 (2) | 396 (1) | 406 (2) | 442 (2) | 562 (2) | 530 (2) | 3129 (2) |
| 0–30 days | 1861 (7) | 2146 (8) | 1934 (7) | 1991 (7) | 2200 (7) | 2481 (8) | 2349 (7) | 14,962 (7) |
Rates of mNEWS2 scores and mortality for the study population across all study years (N = 202746)
Among patients with a critical mNEWS2 score, the 0–30 mortality was 5,758 (23%) deaths, and among patients with a non-critical mNEWS2 score, it was 4,217 (4%) deaths (Additional file 3).
Among patients with a critical mNEWS2 score within urgency level A, the 0-30-day mortality was 4,712 (25%) deaths, and among patients within urgency level B, it was 1,046 (16%) deaths. Among patients with an incomplete registration of mNEWS2, it was 4,509 (29%) (Fig. 4) (Additional file 3). Across all mortality outcomes, only urgency A and B critical show a significant increase in mortality over the study period (see Additional file 6 for the Poisson regression adjusted for age, sex and comorbidity).
Fig. 4.
Kaplan-Meier plot of survival. Legend: Kaplan Meier plot showing the survival days after the patients were assessed in acute critical condition and non-critical condition within urgency level A and B (n = 202,746)
Discussion
The main findings of this study indicate an overall increase in the total number of patients, with the largest increase seen for urgency level B throughout the study period. Urgency level A had the highest proportion of patients in acute critical condition; however, there was a significant increase in the number of patients in acute critical condition within urgency level B over time. The highest increase in dispatched ambulances was seen for patients in non-critical condition within both urgency level A and B. The age and severe comorbidity increased over the study period, mostly within urgency level B. Mortality was highest with patients in acute critical condition in urgency level A and remained stable over the years across all mNEWS2 subcategories.
We found an increase in the number of patients, aligning with trends observed in a similar rural-urban region in Denmark from 2013 to 2022 and in the city of Berlin between 2018 and 2021 [1, 18, 19]. In addition to this general rise in number, the data revealed an increase in patient age and a higher prevalence of severe comorbidity within the study population, reflecting a broader demographic shift. As each EMS call was treated as an independent event, this reflects current clinical practice where healthcare professionals do not have readily available information regarding previous EMS contacts by the same patient. Consequently, repeated calls from the same individual were not identified or accounted for in the analyses. This may have influenced the results, as repeated calls could reflect either patients with high healthcare needs or patients with relatively few health problems. In addition to this, we know that COVID-19 led to fewer EMS contacts [20]. These deviations from normal patterns may affect the observed trends and should be considered when interpreting the results.
We also found that the most patients in acute critical condition were categorized under urgency level A, which aligns with findings from a Danish study, including patients contacting 1-1-2 in 2011, where patient severity was assessed based on case fatality and risk of hospital admission [3]. These findings are in accordance with our study of severity evaluated using a mNEWS2 score. The mNEWS2 has previously been applied in prehospital settings and has been useful in predicting patient outcomes [14, 21]. As such, it contributes to a more nuanced understanding of the severity of the patient’s clinical condition, using objective measurements obtained by ambulance personnel on scene.
Furthermore, we found a trend of more patients in acute critical condition are assigned urgency level B. The reason for this development remains unclear. However, we found that the median age appears to be increasing more within urgency level B compared to urgency level (A) It may be speculated that changes in demographics, such as an increasing proportion of older patients and a higher prevalence of multimorbidity, along with atypical symptom presentation or limited resources with the EMS system, could help explain this trend [22–24]. Our study revealed mortality among those in acute critical condition, with 25% and 16% within 30 days for urgency levels A and (B) This emphasizes the acute critical condition defined by mNEWS2 ≥ 7, and is cause for concern, considering the increase in the number of urgency level B patients in an acute critical condition. Overall, the study’s findings point towards a change in patient demographics, evolving resource demands, and increasing clinical severity, all of which contribute to a growing pressure on the prehospital care system. The findings may contribute to changes in clinical practice as they underline the importance of relevant information when performing patient triage – especially for complex patient groups such as older and comorbid patients. Several tools already exist to support decision making when assessing an ageing population with increasing comorbidity. For example, providing access to comorbidity measures and the Clinical Frailty Scale may support more accurate telephone triage. Furthermore, for when the prehospital personnel arrive on scene, a sex and age adjusted NEWS2, the International Early Warning Score, has been developed and validated to improve prediction of in hospital mortality among patients in the Emergency Department aged ≥ 18 years [25].
All in all, the findings suggest that the increase in EMS demand is not driven by a rise in low-acuity contacts or by a lower threshold for calling. Rather, the development appears to reflect a broader demographic shift, with more patients contacting EMS, who are generally older and have higher levels of multimorbidity. Within urgency level B, we also observed a gradual increase in clinical severity on scene, while the proportion of critically ill patients within each urgency category remained relatively stable. The overall number of critically ill patients therefore increases because the population using EMS have more complex health issues.
Limitations
We used the mNEWS2 score to assess the severity of patients’ conditions, but this approach has certain limitations. Firstly, the score was modified because temperature was omitted as it is not routinely measured in ambulances. Secondly, mNEWS2 does not account for patients’ clinical presentation, such as neurological status beyond the level of consciousness, pain, or other investigations, such as electrocardiography. These limitations may lead to patients in acute critical condition, where urgency level A is indicated, being overlooked in our study because they could have had normal or non-critical vital signs. Although mNEWS2 was designed for medical deterioration rather than trauma, this limitation is unlikely to affect our results. According to the Danish Trauma Registry, trauma patients represent only a minor fraction of all 1-1-2 responses in the North Denmark Region 442, 443, and 453 cases in 2021, 2022, and 2023, respectively [26–28]. The mNEWS2 score has been used in previous studies to estimate the severity of a patient’s condition, supporting its relevance as a tool in this context despite its limitations [14, 29, 30].
Furthermore, we accepted up to two missing vital signs to calculate a mNEWS2 score, resulting in a total of 7,954 (7%) with more than 2 missing NEWS2 scores in urgency level A and 7,483 (9%) in urgency level B. Patients who did not have a full set of vital signs measured in the ambulance had a higher mortality than those categorized as normal, non-critical, or critical. We also recognize that some severely ill patients may be included in the incomplete-data group. Incomplete documentation of basic vital signs is a known issue across various health care settings [31]. In a U.S.-based trauma study, mortality was more than twice as high among patients with missing prehospital vital signs compared to those with complete registration [32]. Furthermore, research has identified an association between missing vital signs and mortality risk, raising the possibility that a significant proportion of the patients excluded from this study may have already died upon ambulance arrival [32, 33]. This may indicate that the patients in this study were either too severely ill or injured to be fully assessed or had died before ambulance arrival. As a result, the missing data could impact the results, potentially leading to an underestimation of the number of patients in acute critical condition. At present, we are unable to systematically identify patients who were found dead, but work is ongoing within the region to develop a methodological solution that will allow us to do so in future works.
To avoid skewing the 0–30 mortality analysis, we applied a 30-day exclusion period. With this approach, the possible overestimation of mortality is reduced by not including the same patients’ deaths multiple times. However, this may have led to an underestimation of mortality in cases where repeated ambulance contacts were close in time before death. This approach aimed to balance these risks by ensuring each patient was represented only once in the mortality analysis, which is also used in other studies [34].
Conclusion
The number of patients has increased from 2017 to 2023, as has patient age and comorbidity level. During the years, there were similar proportions of acute critical severity, as well as normal and non-critical severity, showing that the acute severity among prehospital patients has not diminished. The lack of alignment between the level of urgencies and the acuity severity found on scene should be monitored carefully, together with mortality and clinical outcomes.
Supplementary Information
Below is the link to the electronic supplementary material.
Supplementary Material 1: Additional file 1: Title of data: The modified NEWS2 without temperature. Description: The modified NEWS2 table (without temperature) assigns 0–3 points to six vital signs based on how abnormal they are. More abnormal values such as low SpO₂, need for oxygen, extreme respiratory or pulse rates, abnormal blood pressure, or reduced consciousness receive higher scores, while normal values score 0
Supplementary Material 2: Additional file 2: Title of data: Age and Comorbidity over the years within urgency levels A and B. Description: Data on the study populations’ age and level of comorbidity divided into urgency level A and B over the study period
Supplementary Material 3: Additional file 3: Title of data: 0–1 day, 2–7 days, and 0–30 days mortality within the mNEWS2 scores. Description: Data showing how 0–30-day mortality varies across mNEWS2 severity groups and across urgency levels A and B, including how mortality increases with higher mNEWS2 scores
Supplementary Material 4: Additional file 4: Title of data: Top five missing vital signs for each group, n (%). Description: In the top five for each group, respiratory rate (RR), Glasgow Coma Scale (GCS) and blood pressure (BP) are represented. The combination GCS + RR appears in the top five in three groups. Both oxygen saturation (SpO₂) + pulse (P) and SpO₂+BP are represented in two groups, while RR + BP appears in the top five in only one group. In the top five for the Incomplete mNEWS2 group, three to five vital signs are missing
Supplementary Material 5: Additional file 5: Title of data: Total number of missing vital signs, n (%). Description: Table demonstrating that 82% of the total population had 0 missing vital signs, and only 3% has now vital signs measured
Supplementary Material 6: Additional file 6: Title of data: Trend in mortality. Description: Poisson regression adjusted for age, sex and comorbidity
Acknowledgements
We would like to thank Jakob Nebeling Hedegaard for his contribution to data analysis.
Abbreviations
- EMS
Emergency Medical Services
- mNEWS2
Modified National Early Warning Score 2
- ePMR
Electronic Prehospital Patient Medical Record
- CPR number
Civil Registration Number
- CI
95% Confidence Interval
Author contributions
KS, TAL, EFC, and MBS designed the study. KS, TAL, TAK, and MBS managed the data, including data analyses. All authors contributed to interpreting the results. KS drafted the manuscript, and EFC, MBS, and TAL contributed substantially to its revision. All authors approved the final manuscript.
Funding
No funding was received for this study.
Data availability
Because the study involves sensitive patient information, access to non-public data is restricted. Researchers who wish to obtain the data must first seek approval from the Danish Patient Safety Authority. Upon receiving approval, data can then be requested from the Centre for Prehospital and Emergency Care in Aalborg, Denmark.
Declarations
Ethics approval and consent to participate
The study was registered as part of ongoing research activities in the North Denmark Region (ID F2024- 176). Permission for the handover of relevant variables from the medical records was given by North Denmark Region (ID 1-45-72-4859-24). Registry-based studies that do not involve biological material do not require approval from a Committee on Health Research Ethics.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Møller TP, Jensen JT, Ersbøll AK, Blomberg SNF, Christensen HC. Emergency call utilization over a 10-years period: an observational study in Region Zealand, Denmark, 2013–2022. Scand J Trauma Resusc Emerg Med. 2024;32(1):129. 10.1186/s13049-024-01307-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Lowthian JA, Jolley DJ, Curtis AJ, Currell A, Cameron PA, Stoelwinder JU, et al. The challenges of population ageing: accelerating demand for emergency ambulance services by older patients, 1995–2015. Med J Aust. 2011;194(11):574–8. 10.5694/j.1326-5377.2011.tb03107.x. PubMed PMID: 21644869. [DOI] [PubMed] [Google Scholar]
- 3.Andersen MS, Johnsen SP, Sørensen JN, Jepsen SB, Hansen JB, Christensen EF. Implementing a nationwide criteria-based emergency medical dispatch system: A register-based follow-up study. Scand J Trauma Resusc Emerg Med. 2013;21(1):53. 10.1186/1757-7241-21-53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Booker MJ, Simmonds RL, Purdy S. Patients who call emergency ambulances for primary care problems: a qualitative study of the decision-making process. Emerg Med J EMJ. 2014;31(6):448–52. 10.1136/emermed-2012-202124. PubMed PMID: 23535018. [DOI] [PubMed] [Google Scholar]
- 5.Coster JE, Turner JK, Bradbury D, Cantrell A. Why Do People Choose Emergency and Urgent Care Services? A Rapid Review Utilizing a Systematic Literature Search and Narrative Synthesis. Acad Emerg Med Off J Soc Acad Emerg Med. 2017;24(9):1137–49. 10.1111/acem.13220. PubMed PMID: 28493626; PubMed Central PMCID: PMC5599959. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Morley C, Unwin M, Peterson GM, Stankovich J, Kinsman L. Emergency department crowding: A systematic review of causes, consequences and solutions. PLoS ONE. 2018;13(8):e0203316. 10.1371/journal.pone.0203316. PubMed PMID: 30161242; PubMed Central PMCID: PMC6117060. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Dansk Indeks for Akuthjælp [Danish Index for Emergency Care]. Danske Regioner; 2017.
- 8.Søvsø MB, Haurum RB, Ebbesen TH, Rasmussen AØ, Ward LM, Mogensen ML, et al. Emergency Call versus General Practitioner Requested Ambulances - Patient Mortality, Disease Severity and Pattern. Clin Epidemiol. 2024;16:513–23. 10.2147. /CLEP.S469430 PubMed PMID: 39101155; PubMed Central PMCID: PMC11297546. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Regionerne - kort fortalt [The Regions - An Overview] [Internet]. Danske Regioner; 2012 [cited 2025 Mar 31]. Available from: https://www.regioner.dk/media/3066/regionerne-kort-fortalt-2011.pdf.
- 10.Lindskou TA, Mikkelsen S, Christensen EF, Hansen PA, Jørgensen G, Hendriksen OM, et al. The Danish prehospital emergency healthcare system and research possibilities. Scand J Trauma Resusc Emerg Med. 2019;27(1):100. 10.1186/s13049-019-0676-5. PubMed PMID: 31684982; PubMed Central PMCID: PMC6829955. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Regionernes Kliniske Kvalitetsudviklingsprogram. Præhospitaldatabasen, Årsrapport 2023 [Prehospital Database. Annual Report 2023] [Internet]. København: Regioneres Kliniske Kvalitetsudviklingsprogram; 2024 Dec. p. 61. Available from: https://www.sundk.dk/media/xnkhxlnq/praeshospitalsdatabasen-2023.pdf.
- 12.Von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007;370(9596):1453–7. 10.1016/S0140-6736(07)61602-X. [DOI] [PubMed] [Google Scholar]
- 13.Rosenkrantz O, Benson C, Lindskou T, Fuglsang C, Pedersen L, Mikkelsen S, et al. Existing Data Sources in Clinical Epidemiology: The Danish Prehospital Medical Record System. Clin Epidemiol. 2025;17:735–45. 10.2147/CLEP.S524197. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sørensen MC, Søvsø MB, Christensen EF, Lindskou TA. Critically deviating vital signs among patients with non-specific diagnoses-A register-based historic cohort study. PLoS ONE. 2023;18(11):e0293762. 10.1371/journal.pone.0293762. PubMed PMID: 37910584; PubMed Central PMCID: PMC10619789. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Luís L, Nunes C. Short national early warning score - developing a modified early warning score. Aust Crit Care. 2018;31(6):376–81. 10.1016/j.aucc.2017.11.004. PubMed PMID: 29242109. [DOI] [PubMed]
- 16.Hansen FG, Lind PC, Stankovic N, Holmberg MJ, Andersen LW, Granfeldt A. Vital signs preceding in-hospital cardiac arrest: a matched case-control study. Resuscitation. 2025;215:110778. 10.1016/j.resuscitation.2025.110778. [DOI] [PubMed] [Google Scholar]
- 17.Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373–83. 10.1016/0021-9681(87)90171-8. PubMed PMID: 3558716. [DOI] [PubMed] [Google Scholar]
- 18.Herr D, Bhatia S, Breuer F, Poloczek S, Pommerenke C, Dahmen J. Increasing emergency number utilisation is not driven by low-acuity calls: an observational study of 1.5 million emergency calls (2018–2021) from Berlin. BMC Med. 2023;21(1):184. 10.1186/s12916-023-02879-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Christensen EF, Bendtsen MD, Larsen TM, Jensen FB, Lindskou TA, Holdgaard HO, et al. Trends in diagnostic patterns and mortality in emergency ambulance service patients in 2007–2014: a population-based cohort study from the North Denmark Region. BMJ Open. 2017;7(8):e014508. 10.1136/bmjopen-2016-014508. PubMed PMID: 28827233; PubMed Central PMCID: PMC5724206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Lindskou TA, Bogh SB, Kløjgaard TA, Fløjstrup M, Folke F, Væggemose U, et al. Patients’ use of Danish emergency medical services before and during the COVID-19 pandemic: a register-based study. Scand J Trauma Resusc Emerg Med. 2024;32(1):92. 10.1186/s13049-024-01267-1. PubMed PMID: 39300552; PubMed Central PMCID: PMC11414125. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Martín-Rodríguez F, Sanz-García A, Ortega GJ, Delgado Benito JF, Aparicio Obregon S, Martínez Fernández FT, et al. Tracking the National Early Warning Score 2 from Prehospital Care to the Emergency Department: A Prospective, Ambulance-Based, Observational Study. Prehosp Emerg Care. 2023;27(1):75–83. 10.1080/10903127.2021.2011995. PubMed PMID: 34846982. [DOI] [PubMed] [Google Scholar]
- 22.Hortmann M, Singler K, Geier F, Christ M. Recognition of infections in elderly emergency patients. Z Gerontol Geriatr. 2015;48(7):601–7. 10.1007/s00391-015-0903-2. PubMed PMID: 25986073. [DOI] [PubMed] [Google Scholar]
- 23.Walsh EE, Cox C, Falsey AR. Clinical features of influenza A virus infection in older hospitalized persons. J Am Geriatr Soc. 2002;50(9):1498–503. 10.1046/j.1532-5415.2002. 50404.x PubMed PMID: 12383146. [DOI] [PubMed] [Google Scholar]
- 24.Lars Sortland LM, Haraldseide. Ingrid Sebjørnsen. Eldre i den akuttmedisinske kjeden [Elderly in the Emergency Medical Chain]. Bergen: Nasjonalt kompetansesenter for legevaktmedisin, NORCE Norwegian Research Centre; 2021. pp. 1–2021. [Google Scholar]
- 25.Candel BGJ, Nissen SK, Nickel CH, Raven W, Thijssen W, Gaakeer MI, et al. Development and External Validation of the International Early Warning Score for Improved Age- and Sex-Adjusted In-Hospital Mortality Prediction in the Emergency Department. Crit Care Med. 2023;51(7):881–91. 10.1097/CCM.0000000000005842. PubMed PMID: 36951452; PubMed Central PMCID: PMC10262984. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Regionerne Kliniske Kvalitetsudviklings Program. Danish Trauma Registry 2021. 2021.
- 27.Regionerne Kliniske Kvalitetsudviklings Program. Danish Trauma Registry 2022. 2022.
- 28.Regionerne Kliniske Kvalitetsudviklings Program. Danish Trauma Registry 2023. 2023.
- 29.Myrstad M, Ihle-Hansen H, Tveita AA, Andersen EL, Nygård S, Tveit A, et al. National Early Warning Score 2 (NEWS2) on admission predicts severe disease and in-hospital mortality from Covid-19 - a prospective cohort study. Scand J Trauma Resusc Emerg Med. 2020;28(1):66. 10.1186/s13049-020-00764-3. PubMed PMID: 32660623; PubMed Central PMCID: PMC7356106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Holland M, Kellett J. A systematic review of the discrimination and absolute mortality predicted by the National Early Warning Scores according to different cut-off values and prediction windows. Eur J Intern Med. 2022;98:15–26. 10.1016/j.ejim.2021.12.024. [DOI] [PubMed] [Google Scholar]
- 31.Kjær J, Milling L, Wittrock D, Nielsen LB, Mikkelsen S. The data quality and applicability of a Danish prehospital electronic health record: A mixed-methods study. PLoS ONE. 2023;18(10):e0293577. 10.1371/journal.pone.0293577. PubMed PMID: 37883522; PubMed Central PMCID: PMC10602337. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Laudermilch DJ, Schiff MA, Nathens AB, Rosengart MR. Lack of emergency medical services documentation is associated with poor patient outcomes: a validation of audit filters for prehospital trauma care. J Am Coll Surg. 2010;210(2):220–7. 10.1016/j.jamcollsurg.2009.10.008. PubMed PMID: 20113943; PubMed Central PMCID: PMC2853922. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Petersen M, Kjeldtoft FG, Christensen EF, Bøggild H, Lindskou TA. A classification system for identifying patients dead on ambulance arrival: a prehospital medical record review. Scand J Trauma Resusc Emerg Med. 2023;31(1):107. 10.1186/s13049-023-01171-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Ibsen S, Lindskou TA, Nickel CH, Kløjgård T, Christensen EF, Søvsø MB. Which symptoms pose the highest risk in patients calling for an ambulance? A population-based cohort study from Denmark. Scand J Trauma Resusc Emerg Med. 2021;29(1):59. 10.1186/s13049-021-00874-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Material 1: Additional file 1: Title of data: The modified NEWS2 without temperature. Description: The modified NEWS2 table (without temperature) assigns 0–3 points to six vital signs based on how abnormal they are. More abnormal values such as low SpO₂, need for oxygen, extreme respiratory or pulse rates, abnormal blood pressure, or reduced consciousness receive higher scores, while normal values score 0
Supplementary Material 2: Additional file 2: Title of data: Age and Comorbidity over the years within urgency levels A and B. Description: Data on the study populations’ age and level of comorbidity divided into urgency level A and B over the study period
Supplementary Material 3: Additional file 3: Title of data: 0–1 day, 2–7 days, and 0–30 days mortality within the mNEWS2 scores. Description: Data showing how 0–30-day mortality varies across mNEWS2 severity groups and across urgency levels A and B, including how mortality increases with higher mNEWS2 scores
Supplementary Material 4: Additional file 4: Title of data: Top five missing vital signs for each group, n (%). Description: In the top five for each group, respiratory rate (RR), Glasgow Coma Scale (GCS) and blood pressure (BP) are represented. The combination GCS + RR appears in the top five in three groups. Both oxygen saturation (SpO₂) + pulse (P) and SpO₂+BP are represented in two groups, while RR + BP appears in the top five in only one group. In the top five for the Incomplete mNEWS2 group, three to five vital signs are missing
Supplementary Material 5: Additional file 5: Title of data: Total number of missing vital signs, n (%). Description: Table demonstrating that 82% of the total population had 0 missing vital signs, and only 3% has now vital signs measured
Supplementary Material 6: Additional file 6: Title of data: Trend in mortality. Description: Poisson regression adjusted for age, sex and comorbidity
Data Availability Statement
Because the study involves sensitive patient information, access to non-public data is restricted. Researchers who wish to obtain the data must first seek approval from the Danish Patient Safety Authority. Upon receiving approval, data can then be requested from the Centre for Prehospital and Emergency Care in Aalborg, Denmark.