The Role of Perceived Urgency and Barriers in Using Emergency Medical Care for Sepsis: A German-Wide, Representative Survey Using Case Vignettes

Abstract

Objectives:

Delays in emergency medical care (EMC) can increase sepsis mortality. Sepsis symptoms, however, are less known in the public compared with signs of stroke and myocardial infarction. Using case vignettes, we investigated the decision to seek EMC in sepsis-related compared with other emergencies.

DESIGN:

Cross-sectional survey.

SETTINGS AND SUBJECTS:

A representative nationwide survey was conducted in December 2023 among 1013 persons in Germany.

Interventions:

None.

Measurements and Main Results:

A standardized questionnaire was administered via face-to-face interviews to assess perceived urgency, the tendency to seek EMC, sepsis knowledge, and barriers to EMC utilization. Multiple regression analyses were conducted to identify predictors for the perceived urgency and the tendency to seek EMC. Across ten case vignettes, the perceived urgency was lower for sepsis-related (mean = 3.93, sd = 0.64) than for nonsepsis-related emergencies (mean=4.15, sd = 0.67). In sepsis-related vignettes, the immediate EMC option (“Call emergency medical services or go to the emergency department [ED] immediately”) was chosen by 18.3–43.2% of respondents, compared with 20.1–71.9% in nonsepsis-related vignettes. Even when situations were rated as very urgent, EMC use was less likely for sepsis-related emergencies (35%) than for nonsepsis-related emergencies (17%). In regression analyses, sepsis knowledge was the only predictor of perceived urgency, and perceived urgency was the strongest predictor of EMC use. The likelihood of EMC use in sepsis-related emergencies was 14.9 percentage points lower (p < 0.01) than in nonsepsis-related emergencies; “long waiting times in the ED” was the only significant barrier for EMC use.

Conclusions:

Compared with nonsepsis emergencies, sepsis-related emergencies are perceived as less urgent and less often lead to EMC use, even if the level of perceived urgency is the same. As the perceived urgency is associated with sepsis knowledge, this underlines the need to increase efforts to improve sepsis knowledge and the awareness that sepsis requires timely emergency care in the general public.

KEY POINTS.

Question: How important are the perceived urgency and barriers for seeking emergency medical care (EMC) for sepsis?

Findings: In a nationwide survey with 1013 persons in Germany, perceived urgency was found to be the strongest predictor of seeking EMC. The likelihood of seeking EMC was significantly lower for sepsis-related emergencies (–14.9%) than for nonsepsis-related emergencies, even when controlling for perceived urgency.

Meaning: While barriers play a secondary role in the decision to use EMC, the decisive factor is the perceived urgency of the emergency, which, in the case of sepsis, also depends on knowledge of the disease.

Sepsis is a life-threatening medical emergency caused by a dysregulated host response to an infection (1). Every year, around 50 million patients are affected, of which 11 million die (2). Sepsis often presents with unspecific early symptoms, such as dyspnea or altered mentation (3). This can make the early diagnosis of sepsis challenging for both patients and healthcare providers, potentially leading to delays in medical treatment (4). In an U.S.-wide study including 21 emergency medical services (EMS) agencies in Pennsylvania between 2010 and 2012, the median time from first medical preclinical contact to antibiotic administration in the emergency department (ED) was 4.2 hours (interquartile range, 2.7–8.0 hr) (5). With each hour of delay in antibiotic administration, the adjusted odds ratio for death increased by 3% per hour (p < 0.01) (5). In a German multicenter study, the sepsis mortality risk increased by 2% per hour delay of antimicrobial therapy and 1% per hour delay of source control (6). However, delays and barriers to sepsis recognition do not begin at the healthcare facility but rather with the patient’s decision or the families/caregivers advice to seek emergency medical care (EMC) (7–9). In a previous survey among sepsis risk groups, we found that most patients would not use EMC in sepsis-related emergencies presented in case vignettes (10). Compared with other medical emergencies such as stroke and acute myocardial infarction (AMI), respondents more frequently decided that they would wait or consult their general practitioner (GP) (10). Although this can have important implications for sepsis treatment, we lack knowledge on what affects the decision making of sepsis patients to seek EMC. A systematic review and meta-analysis found that the decision to use emergency and urgent care services was influenced by the following factors in general: access to and confidence in primary care; perceived urgency/need for EMC; and barriers (costs, location) and advice from family, friends, or healthcare professionals (11). For AMI and stroke, the perception that the symptoms were unrelated to an important health problem (12) and the failure to recognize the urgency of symptoms (13) were identified as most important barriers to the timely utilization of EMS in two previous survey studies from Italy and Korea. We hypothesized that these factors may also be important in sepsis, thus building on our earlier survey (10), the present study extends this work by using a population-representative sample, directly assessing urgency for each vignette, and explicitly capturing potential barriers to seeking EMC. These advances allow us to provide more generalizable insights and to test: 1) if the perceived urgency of sepsis-related emergencies differs from other medical emergencies, 2) if sepsis knowledge and patient demographics are associated with the perceived urgency in sepsis-related emergencies, and 3) what facilitates and hinders seeking EMC in sepsis-related and other emergencies.

METHODS

Study Design

We conducted a prospective, cross-sectional, representative survey study, which was carried out by the Institute for Demoscopy (IfD) Allensbach from December 1–14, 2023. The IfD regularly conducts omnibus surveys on various topics. As part of the IfD survey number 12079, a representative sample of the German population 16 years old or older was interviewed. The sample was drawn using quota sampling. Quotas were determined based on official statistical data for the German population 16 years old or older. Remaining disproportionalities were resolved by weights, which were calculated based on combinations of age and gender, level of education and region (former East and West Germany), occupational group and region, and household size, marital status, Nielsen areas (14), town size, and voting intention (political party).

The study was approved by institutional review board of the Jena University Hospital (Reg.-Nr. 2025-3798-Bef, study title: “Representative Survey on Sepsis Knowledge in the German Population”) and was conducted in accordance with the Helsinki Declaration 1975. Standardized interviews were conducted face-to-face by one of trained 318 interviewers. On average, each interviewer recruited 3–5 participants. Informed consent was not required from participants under data protection law, as the data were processed, evaluated, and passed on in anonymized form. Interviewers signed to confirm appropriate interview conduct in accordance with legal regulations. We followed the Consensus-Based Checklist for Reporting of Survey Studies checklist (15) to present the study results.

Questionnaire

The questionnaire was constructed based on previous studies (10) (W. Abels, unpublished observataions, 2025) and consisted of 15 questions on sociodemographic information, 60 questions on sepsis knowledge, ten case vignettes describing medical emergencies and a question on barriers for seeking EMC (Supplement, https://links.lww.com/CCM/H850).

Variables

Concerning sociodemographic data, we recorded age, gender, region, and educational level. Educational level was defined by categorizing the educational degrees of Germany (low: 8–9 yr of school; intermediate: 10 yr of school; and high: university entrance qualification or university degree). The region reflects whether a participant lives in former East or West Germany. We also recorded if participants themselves or individuals within their social environment had been affected by sepsis.

Sepsis Knowledge

We assessed five domains of sepsis knowledge: definition and epidemiology (16 items), risk factors (15 items), symptoms (16 items), general prevention (three items), and sepsis sequelae (ten items). We used binary indicator variables to code item responses as 1 = correct and 0 = incorrect. An overall score was calculated as sum score of the indicator variables (i.e., the number of correctly answered items) as measure of sepsis knowledge. The score was calculated only for participants familiar with the term “sepsis”; those unaware received a missing value.

Urgency Ratings of Medical Situations and Intended Actions

We assessed urgency ratings and intended actions for different medical emergencies using ten case vignettes, each representing a clear indication for EMC according to medical guidelines (16, 17). These case vignettes included typical emergencies, of which five were sepsis-related (e.g., sepsis caused by urinary tract infection, sepsis caused by respiratory infection) and five were nonsepsis-related (e.g., AMI or stroke). Participants were asked to rate the urgency of the medical situations on a five-point Likert scale from 1 = very low to 5 = very high. Furthermore, they had to choose one of three possible courses of action (1 = “Call the EMS or go to the ED immediately,” 2 = “Visit the GP on the same day or the outpatient emergency service of the association of statutory health insurance physicians,” and 3 = “Wait 1 day and then decide”). In Germany, if patients suffer from an emergency, they are supposed to either call EMS (calling 112 and receiving on-site-care or care during transportation to a hospital) or to directly visit an ED. In case of less urgent cases and out-of-office hours of the outpatient doctors, patients are supposed to call the so-called “ärztliche Bereitschaftsdienst” (outpatient emergency service of the association of statutory health insurance physicians) or to visit the “on duty practices.”

For statistical analysis of the intended action, we created a binary variable to reflect the use of EMC, with the response option “Call EMS immediately or go to the ED” coded as “1” and the other two options coded as “0.”

Barriers for Using the EMC.

Unrelated to specific barriers, we assessed possible subjective barriers for EMC use. Respondents were asked to decide on twelve possible barriers (e.g., concerns about possible costs, long waiting time in the ED, responsibility for children or pets) whether they apply to them (coded as yes = 1 or no = 0).

Statistical Analyses

To compare whether the urgency rating and intended action differed between the sepsis-related emergencies and the nonsepsis emergencies, we calculated the mean and sds for the urgency rating and the relative frequencies of the possible intended actions for each case vignette.

We analyzed in a multiple linear regression how the urgency rating of sepsis-related emergencies is related to sociodemographic characteristics, sepsis knowledge, and previous experience with sepsis. In addition, we used a multiple linear regression to examine how the tendency to use EMC is associated to the urgency rating, sociodemographic characteristics, possible barriers, and the type of emergency (sepsis-related vs. nonsepsis-related). For both regression analyses, we calculated the mean for the tendency to use EMC and the urgency rating separately across the five sepsis-related and nonsepsis-related case vignettes for each person. For a person who chooses the option “Call EMS immediately or go to the ED” in three of five nonsepsis-related medical situations, this results in a value of 0.6 for the tendency to use EMC. We reported R² differences as measure for the unique explained variance in the regression models. Factorial weighting was used to adjust for remaining deviations of the sample structure of the realized quota sample from official population statistics. Accordingly, statistical methods for complex survey data with sampling weights were used to ensure correct statistical inference (18, 19). All statistical analyses were performed with the R package “survey” v.4.4-2 (20) in the software R v.4.2.3. (R Foundation for Statistical Computing, Vienna, Austria). We report unadjusted p values of statistical tests with a significance level of α = 0.05. We used “listwise deletion” to handle missing data. As none of the variables had a proportion of missing data of more than 5 %, no significant bias is to be expected (21).

RESULTS

In total, 1013 persons participated in the survey. The mean age and sd of the participants was 51.2 ± 6.5 years, and 48.6% of the participants were male (Table 1). Almost all participants (95.1%) had previously heard the term “sepsis.” The mean overall sepsis knowledge score across all domains was mean = 29.9 of 60 (sd = 6.5). Thus, on average, people achieved just half of the maximum possible points in the overall sepsis knowledge.

TABLE 1.

Descriptive Statistics for Demographics (n = 1013)

Variables	Characteristic Values
Region (West Germany), n (%)	831 (81.9)
Age (yr), mean (sd)	51.2 (18.7)
Gender (male), n (%)	493 (48.6)
Educational level, n (%)
Low	279 (27.5)
Intermediate	311 (30.8)
High	422 (41.7)
Sepsis knowledge, mean (sd)	29.9 (6.5)

Educational level = categorized highest educational qualification, n = absolute frequency.

Missing proportion for all reported variables below 5%.

The descriptive comparison of the urgency rating and the intended action in sepsis-related and nonsepsis-related emergencies revealed that nonsepsis-related emergencies were rated as more urgent and the response option “Call the EMS or go to the ED immediately” was selected more frequently (Table 2). The mean urgency rating over all sepsis-related emergencies was significantly (t = 12.5; degrees of freedom = 1011; p < 0.01) lower (mean = 3.93; sd = 0.64) than that of nonsepsis-related emergencies (mean = 4.15; sd = 0.67). In sepsis-related emergencies, the response category “Call the EMS or go to the ED immediately” was selected by a minority of participants (between 18.3% and 43.2% of respondents). In contrast, in nonsepsis-related emergencies, this category was the preferred option in most vignettes (between 20.1% and 71.9%). This pattern reverses when considering the category “Visit the GP on the same day or the outpatient emergency service of the association of statutory health insurance physicians.” For all sepsis-related emergencies, the majority of respondents selected this category (ranging from 50.0% to 67.5%). In contrast, for nonsepsis-related emergencies, a significantly smaller proportion (between 23.2% and 51.2%) chose this response option.

TABLE 2.

Descriptive Statistics for Intended Action and Urgency Rating of the Case Vignettes (n = 1013)

Type of Emergency (Sepsis-Related)	Vignette	Chief Complaint	Intended Action Response Categories (%)			Urgency Rating
			1	2	3	Mean (sd)
No	1	Slurred speech	59.3	29.8	10.9	4.28 (1.00)
No	2	Hemiparesis	71.9	23.2	4.9	4.56 (0.77)
No	3	Transient neurologic deficit	20.1	51.2	28.7	3.61 (1.10)
No	4	Progressive thoracic tightness	56.3	36.1	7.6	4.28 (0.91)
No	5	Palpation and dizziness	39.4	48.9	11.7	4.03 (0.94)
Yes	1	Flu and shortness of breath	18.3	61.0	20.7	3.65 (0.94)
Yes	2	Kidney pain and chills	36.5	58.4	5.1	4.14 (0.80)
Yes	3	Infected cut and severe feeling of illness	23.0	65.1	11.9	3.81 (0.93)
Yes	4	Cough, fever, and confusion	43.2	50.0	6.8	4.11 (0.91)
Yes	5	Cat bite, severe pain, fever, and chills	26.5	67.5	6.0	3.94 (0.86)

Response category 1 = “Call the emergency medical services or go to the emergency department immediately,” 2 = “Visit the general practitioner on the same day,” and 3 = “Wait 1 day and then decide.” Missing proportion for the intended actions and urgency ratings of all case vignettes below 5%.

Furthermore, we found that a noticeable proportion of participants would not use EMC, even if they classified the medical situation as very urgent (Fig. 1). On average, this proportion was significantly higher for sepsis-related emergencies (mean = 35%) than for nonsepsis-related emergencies (mean = 17%). A detailed description of the intended actions by vignette can be found in Figure S1 (https://links.lww.com/CCM/H850).

Figure 1.

Average proportions of seeking emergency medical care (EMC) for sepsis- and nonsepsis-related emergencies depending on the self-assessed urgency (from 1 = very low to 5 = very high).

In the multiple regression for the urgency rating of sepsis-related case vignettes (Table 3), only a comparably small proportion of 8.8% of the variance was explained (R² = 0.088), with sepsis knowledge as the only significant predictor, while age, sex, education, and previous experience with sepsis were no significant explanatory variables in this model.

TABLE 3.

Regression Coefficients of the Predictors for the Urgency Rating of Sepsis-Related Emergencies

Predictors	b	se	p	R2 Differences
Age (in 10 years)	–0.022	0.023	0.329	0.001
Gender, male	0.008	0.046	0.857	0.000
Educational level (reference: low)				0.003
Intermediate	–0.087	0.063	0.166
High	–0.085	0.063	0.179
Sepsis knowledge	0.026	0.004	< 0.001a	0.063
Personally affected by sepsis	0.119	0.140	0.394	0.002
Sepsis among acquaintances	0.075	0.046	0.101	0.003

Educational level = categorized highest educational qualification.

^aSignificant on a significance level α = 0.05.

Reading example: In the multiple regression, the unstandardized regression coefficient b for the predictor “Sepsis knowledge” corresponds to the change in the urgency rating, controlling for all other predictors, if the sepsis knowledge increases by one point. R² differences—corresponds to the unique explained variance by this predictor.

Regarding the multiple regression for the tendency to use EMC (Table 4), the perceived urgency was the strongest predictor controlling for all other predictors. The likelihood of EMC use in sepsis-related emergencies was significantly lower compared with nonsepsis-related emergencies (–14.9%; p < 0.01). Among the possible barriers to utilization of EMC, only “long waiting times in the ED” was a significant predictor. Age, sex, and education had no association with the tendency to use EMC. In total, 37.1% of the variance of tendency to use EMC is explained by the set of predictor variables (R² = 0.371).

TABLE 4.

Regression Coefficients for Predictors of Tendency to Use Emergency Medical Care

Predictors	b	se	p	R2 Differences
Urgency	15.00	1.11	< 0.01a	0.229
Age (in 10 yr)	0.14	0.85	0.88	0.000
Gender: male	0.21	1.51	0.89	0.000
Educational level (reference: low)				0.001
Intermediate	–2.42	2.14	0.26
High	–2.50	2.07	0.23
Type of emergency: sepsis-related	–14.85	1.11	< 0.01a	0.059
Barriers
Fear of infection in the ambulance or emergency department	0.14	0.03	0.97	0.000
Concerns about possible costs	–1.30	–0.44	0.66	0.000
Concerns about tying up capacity for more serious emergencies	–1.55	–0.88	0.38	0.000
Long distance to the nearest emergency department	–0.30	–0.09	0.93	0.000
Long waiting times in the emergency department	–4.14	–2.42	0.02a	0.004
No car or poor public transport connections	–2.77	–0.99	0.32	0.001
Having difficulty to walk and needing help	0.59	0.18	0.86	0.000
Responsibility for small children	0.13	0.04	0.97	0.000
Responsibility for the care of relatives	5.52	1.20	0.23	0.001
Responsibility for a pet	3.70	0.86	0.39	0.001
Do not want an ambulance to be seen outside my door	–2.57	–0.64	0.52	0.000
General fear of hospitals	–3.31	–1.46	0.15	0.001

^aSignificant on a significance level α = 0.05.

Reading example: In the multiple regression, the regression coefficient b for the predictor “Type of emergency” corresponds to the change in the tendency to use emergency medical care, controlling for all other predictors, if a sepsis-related emergency is described in the case vignette. R² differences—corresponds to the unique explained variance by this predictor.

DISCUSSION

In this representative survey, respondents rated the urgency of sepsis-related medical emergencies lower compared with nonsepsis-related emergencies. The rating of the perceived urgency of sepsis-related emergencies was only associated with sepsis knowledge, while sociodemographic factors and previous sepsis experience had no significant impact. Notably, controlling for the perceived urgency, respondents chose EMC less frequently in sepsis-related urgencies.

Our results align with previous studies showing a positive association between perceived urgency and the decision to seek EMS or ED care, regardless of specific diagnoses (11). We expand this research and the evidence of our previous study (10) by showing that the type of emergency is associated with the tendency of EMC use, even if controlling for the perceived urgency. In that regard, sepsis-related emergencies are associated with a lower tendency for EMC use, which may contribute to delays in treatment of this time-sensitive emergency with potential increases in sepsis mortality. We lack data to understand why patients who rated a vignette as highly urgent were less likely to choose EMC for sepsis-related cases than for nonseptic cases in our study. In another case vignette study, the “certainty” of choice influenced the likelihood to seek medical services (22). It may be possible that respondents’ certainty of the need for medical treatment is greater for AMI and stroke than for sepsis. To this end, even health providers highlight that the low specificity of sepsis symptoms is hindering themselves from early sepsis diagnoses and communication in qualitative interviews (23). Also in public, sepsis symptoms seem to be less known than typical AMI or stroke symptoms (24–26). This is supported by a German study in which EMS patients were much more likely “not” to choose an EMS for a vignette with “high fever for several days” (adverting sepsis) compared with a vignette with “tight chest pain” (indicating AMI) (8). Furthermore, it is possible that previous awareness campaigns for stroke and AMI have resulted in a higher level of awareness in the population and possibly perceived susceptibility of patients, although it was shown that sepsis incidence in EMS is similar to AMI and stroke incidence (27). In addition, the preference for consulting a GP rather than using EMC in sepsis-related emergencies may reflect the more gradual and less specific onset of typical sepsis symptoms compared with the sudden and clearly recognizable presentations of stroke or AMI. This relative ambiguity may reinforce the perception that immediate EMC is not required, even when respondents rate the situation as highly urgent. However, further research is needed to understand the comparatively lower tendency to use EMC in highly urgent sepsis-like emergencies.

Our study identified factors that are associated with the perceived urgency of sepsis-related emergencies. While sociodemographic characteristics had no impact, the respondents’ sepsis knowledge was the most important predictor of the perceived urgency. However, with regard to sepsis knowledge, large gaps exist in the general population and at-risk patients (10). According to healthcare providers interviewed in a German mixed methods study, patients’ sepsis education is currently considered inferior compared with education on other diseases (23). This is in line with a study from Korea that found a lower awareness for sepsis compared with stroke and AMI (25). The public’s low sepsis knowledge may also be partially due to the fact that sepsis is—different to AMI and stroke—a syndrome, thus, often only communicated/known as the underlying infection, for example, an urinary infection, without the word “sepsis” being used (23). It may reflect the fact that AMI and stroke have been recognized and researched for much longer and have therefore been the focus of public awareness campaigns for decades, whereas broader recognition of sepsis has only emerged more recently. It may also contribute to this lower awareness that, unlike AMI or stroke, sepsis is not clearly linked to a single medical specialty that actively promotes awareness or serves as a direct point of reference for the public. Furthermore, sepsis does not appear to be consistently documented in discharge letters of affected patients (28). The importance of sepsis knowledge for rating the urgency of sepsis underlines the necessity of comprehensive public sepsis education, which might also unfold its potential with regard to correct identification of symptoms of life-threatening infection and timely EMC seeking. This claim is also a major demand of a call for the development of a European sepsis plan just recently issued by sepsis survivors (29). In Germany, there are already several initiatives such as World Sepsis Day, the nationwide campaign “Deutschland erkennt Sepsis” (Germany recognizes sepsis), and scientific public health projects in model regions (e.g., Strengthening health literacy of patients at high risk for sepsis in order to improve early diagnosis and treatment of sepsis [SEPWISS]) (30). However, these activities have so far not been sufficient to reach the general population on a broad scale. Strengthening sepsis-related emergency competence as part of overall health literacy therefore remains essential and requires stronger efforts. Furthermore, training of medical professionals and students, who themselves often show limited knowledge (31, 32), represents another crucial element to disseminate sepsis awareness more effectively.

Beyond these efforts, better data and visibility of sepsis are also required to raise awareness at both professional and societal levels. This includes consistent documentation of sepsis in discharge summaries, explicit coding of sepsis on death certificates, and the establishment of national surveillance systems to more accurately capture the burden of disease (33). Such measures are in line with the World Health Organization Global Sepsis Resolution (34), including public awareness as a key priority.

Another important finding is that the perceived urgency is the major driver for the tendency to use emergency care, and barriers only play a minor role. Among the barriers, only long waiting times in the ED were associated with a decreased tendency for EMC seeking. This barrier was also highlighted in a previous qualitative study on barriers to seeking care for symptoms of bacterial sepsis (35) and emerged as major barrier in studies on the seeking care in the ED for other medical emergencies (36). In contrast to findings from U.S. studies (36), fear of costs or mobility-related barriers were not significantly associated with the tendency to use EMC in our study. This may be due to differences in the healthcare system with compulsory health insurance for all citizens, low costs for EMS use (usually 10 Euros per transport and 10 Euros per day hospitalization) and comparably short distances to hospitals with EDs in Germany (37).

This study is among the first to use case vignettes to analyze perceived urgency, intended actions, and barriers to EMC use, comparing sepsis with nonsepsis-related emergencies. Its representative sample enables generalizable results for the German population and expands the literature by focusing on reasons for not seeking emergency care, whereas most prior studies examined emergency overuse and overcrowding (38).

However, our study has also limitations. Case vignettes represent a thought experiment, and it is possible that the respondents would act differently if they actually experienced the symptoms described in the case vignettes themselves. Furthermore, as our vignettes used typical AMI and stroke symptoms, it may be possible that we overestimated the respondents’ ability to identify the correct actions for those emergencies, which also can present with less specific symptoms, for example, in patients with diabetes or the elderly. Furthermore, information on participants’ occupational background, including potential employment in healthcare, was not collected and could therefore not be considered in our analyses, although this may influence the ability to assess medical emergencies. A further limitation is that information on the total number of individuals approached and on incomplete interviews was not provided by the survey institute, which restricts the assessment of response rates and potential selection bias. Finally, the survey may not have captured all relevant barriers to seeking emergency care, particularly system-level factors such as access, availability, or perceived capacity of EMS. Although we included items approximating this aspect (e.g., concerns about disadvantaging other patients), broader structural barriers may not be fully reflected.

CONCLUSIONS

While barriers play a secondary role in the decision to use EMC, the decisive factor is the perceived urgency of the emergency, which, in the case of sepsis, also depends on knowledge of the disease. Given the rather low level of sepsis knowledge in the general population, this emphasizes the potential and the need for an increase sepsis knowledge in public, with a specific focus on sepsis symptoms and the awareness that sepsis requires timely emergency care.

Supplementary Material

ccm-54-291-s001.docx ^{(949.6KB, docx)}