Abstract
Objective
We aimed to evaluate the differences in characteristics and illness/injury severity among patients who present to the emergency department (ED) via emergency medical services (EMS) compared to patients who present via other means.
Methods
We analyzed a nationwide sample of ED visits from the 2015–2019 National Hospital Ambulatory Medical Care Survey. We excluded patients <18 or >92 years old, who eloped or left against medical advice, or who arrived via interfacility transport. Mode of presentation was dichotomized to those presenting to the ED via EMS versus any other mode of transportation. Using the appropriate survey sampling weights, we described patient characteristics and compared measures of illness/injury severity between groups using a multivariable logistic regression model.
Results
An unweighted total of 73,397 ED visits, representing a weighted estimate of 528,083,416 ED visits in the United States during 2015–2019, included 18% arriving via EMS and 82% via other means. EMS patients were older, more often male, more often had multiple chronic medical conditions, and less often had private insurance. EMS patients had higher priority triage scores, consumed more resources in the ED, and had longer lengths of stay. Arrival by EMS was associated with higher odds of hospital admission (odds ratio [OR] 2.7, 95% confidence interval [CI] 2.4–2.9) and in‐hospital mortality (OR 11.1, 95% CI 7.3–17.2).
Conclusions
Patients presenting via EMS had significantly different characteristics and outcomes than those presenting via other means. These important differences should be considered when comparing studies of all ED patients versus those who present via EMS.
Keywords: acuity, emergency medical services, prehospital, severity, transport, triage
1. INTRODUCTION
1.1. Background
The 2001 National Emergency Medical Services (EMS) Research Agenda called for the promotion of EMS‐specific research to inform evidence‐based prehospital care guidelines and improve the quality of acute care research. 1 In the 20 years since, total National Institutes of Health funding for EMS research has increased more than 6‐fold and the number of relevant publications has increased th3ree‐fold, both outpacing growth trends in overall medical research during the same period. 2 As the field of EMS research continues to grow, findings from the frontiers of prehospital care often rely upon literature rooted in data from the emergency department for contextualization. Although the majority of EMS patients represent a sizable subset of all ED patients, the degree to which they tend to differ from others remains incompletely understood.
1.2. Importance
Patients arriving by EMS are an important yet distinct subset of overall ED patients, with differences in characteristics, acuity, and outcomes when comparing these 2 overlapping but distinct populations. Failure to appreciate baseline differences in these populations can lead to misinterpretation of new findings and potentially erroneous recommendations for clinical practice, particularly in the prehospital setting. Prior studies have reported patterns of sociodemographic differences between ED patients who present via EMS versus other means, including older age, urban residence, and public health insurance. 3 , 4 EMS patients also appear to have higher acuity than those presenting by other means, with outcomes that include requiring more hospital resources than others and greater acuity in triage scores and diagnosis‐related grouping severity. 3 , 4 , 5 However, it remains unclear whether more reliable outcome measures of clinical severity such as hospital admission, length of stay, and mortality differ between these populations.
1.3. Goals of this investigation
In this study, we aimed to describe the difference between patients arriving by EMS and those arriving by other means in a nationally representative sample of ED visits to evaluate the hypothesis that patients who present to the ED via EMS have greater severity of illness and injury compared to patients who present to the ED using any other means.
2. METHODS
2.1. Study design and data source
We performed a retrospective cohort study using the ED component of the National Hospital Ambulatory Medical Care Survey (NHAMCS) data set, administered by the National Center for Health Statistics (NCHS). 6 The NHAMCS data set uses a national probability sample survey to examine ED visits in the United States and generates weighted estimates of national statistics related to health care use. 7 Data collection for NHAMCS is managed by the US Census Bureau, whose staff follow structured protocols to train field agents, collect survey data, and complete data review and quality assurance to maximize standardization and minimize errors and missingness. 8 This study was deemed exempt by the Institutional Review Board for Human Subjects Research (IRB22‐0855).
2.2. Study population
From all ED visits from 2015–2019 in the NHAMCS data set, we excluded patients <18 years old, patients >92 years old (due to top coding of ages in this range), and patients who did not complete the ED visit (ie, left against medical advice or eloped from the ED). Only the first presentation was included for a given health care encounter marked by interfacility transfer, excluding subsequent presentations at receiving facilities. Due to known differences in EMS use patterns in children, patients under age 18 were excluded to avoid confounding results associated with a pooled sample. 9 Patients who did not complete the ED visit were excluded to avoid patient‐directed effects on outcome measures, such as ED disposition and length of stay. Interfacility transfer encounters were excluded to ensure each health care encounter was captured only once. A combination of demographic, socioeconomic, and clinical data were extracted from the deidentified NHAMCS data set at the patient encounter level. Sampling weights determined by the NCHS were used to generate national estimates. We repeated this procedure for the NHAMCS data set 2020 release as well but chose to analyze 2020 separately due to extensive changes in the US health care system during the COVID‐19 pandemic.
2.3. Measures
Demographic and socioeconomic variables of interest used in this study included patient age, sex, race or ethnicity, type of residence (based on home address provided), medical comorbidities, and health insurance status. Clinical severity markers included resource use and initial acuity. Number of procedures, diagnostic tests, and medications administered in the ED were calculated to evaluate the degree of hospital resource use. Emergency severity index (ESI) triage scores were used to determine initial assessment of illness or injury severity upon arrival to the ED, where “1” signifies immediate need for medical attention and “5” signifies lowest level of priority. ED outcome measures included discharge from the ED, hospital admission (including disposition to the operating room), and death in the ED. Hospital outcome measures included length of stay (in days) and survival to hospital discharge (ie, inpatient mortality).
2.4. Statistical analysis
We calculated weighted descriptive statistics with 95% confidence intervals (CI) for characteristics and outcomes of each patient and ED visit for the years 2015–2019 and 2020, separately, stratified by mode of transportation to the ED. Estimates from 2020 were not pooled with prior years due to the extensive disruption in the overall health care system, including EMS, due to the COVID‐19 pandemic. Using data from 2015–2019, survey weighted multivariable logistic regression was performed using complete case analysis for outcome measures of hospital admission and in‐hospital mortality with respect to mode of transportation to the ED, controlling for key patient characteristics. All statistical analyses were performed using Stata version 15.0 (StataCorp, College Station, TX, USA).
3. RESULTS
During 2015–2019, a total of 73,397 unweighted ED visits met criteria for inclusion in this study, representing a weighted 528,083,416 ED visits in the United States over that period (Figure 1). During 2020, 11,748 ED visits represented a weighted estimate of 105,394,640 ED visits in the United States. Table 1 summarizes the descriptive characteristics of this sample, stratified by mode of transportation to the ED. In this sample, 17.7% used EMS, and the remaining 82.3% presented to the ED via any other means. Patients who used EMS were older (mean age of EMS 56.6 years, 95% CI 55.8–57.5; other means 45.2 years, 95% CI 44.7–45.7) and less often female. EMS patients were less often of Hispanic ethnicity, and no differences were found among non‐Hispanic populations with respect to race or ethnicity (ie, between White, Black, and Asian/other non‐Hispanic populations, as categorized by NCHS).
FIGURE 1.
Selection criteria for ED visits (unweighted) during 2015–2019 included in the primary analysis, Abbreviations: AMA, against medical advice; ED, emergency department.
TABLE 1.
Characteristics of included emergency department visits, National Hospital Ambulatory Medical Care Survey 2015–2019.
| Characteristic | Overall | Arrived by EMS | Arrived by other means |
|---|---|---|---|
| Unweighted n (unweighted %) | 73,397 (100) | 12,963 (17.7) | 60,434 (82.3) |
| Weighted n (weighted %) | 528,083,416 (100) | 92,781,194 (17.6) | 435,302,222 (82.4) |
| Age, weighted mean (95% CI) | 47.2 (46.7–47.7) | 56.6 (55.8–57.5) | 45.2 (44.7–45.7) |
| Sex, weighted % (95% CI) | |||
| Female | 57.4 (56.7–58.1) | 53.5 (52.2–54.7) | 58.3 (57.6–59.0) |
| Male | 42.6 (41.9–43.3) | 46.5 (45.3–47.8) | 41.7 (41.0–43.4) |
| Race or ethnicity, weighted % (95% CI) | |||
| Non‐Hispanic White | 60.7 (57.7–63.6) | 64.4 (61.6–67.2) | 59.9 (56.8–62.9) |
| Non‐Hispanic Black | 23.1 (20.2–26.2) | 22.0 (19.5–24.7) | 23.3 (20.3–26.6) |
| Hispanic | 13.3 (11.7–15.1) | 10.3 (8.9–11.8) | 14.0 (12.2–15.9) |
| Non‐Hispanic other | 3.0 (2.4–3.6) | 3.3 (2.7–4.0) | 2.9 (2.3–3.5) |
| Patient residence, weighted % (95% CI) | |||
| Private residence | 93.1 (92.0–94.0) | 81.9 (80.4–83.3) | 95.5 (94.4–96.3) |
| Nursing home | 1.9 (1.7–2.1) | 8.1 (7.3–9.0) | 0.5 (0.4–0.7) |
| Homeless/homeless shelter | 1.0 (0.9–1.2) | 2.5 (2.1–3.0) | 0.7 (0.6–0.9) |
| Other | 1.5 (1.2–1.8) | 4.0 (3.1–5.0) | 1.0 (0.8–1.2) |
| Missing/unknown | 2.5 (1.8–3.5) | 3.5 (2.6–4.6) | 2.3 (1.6–3.4) |
| Chronic conditions, weighted % (95% CI) | |||
| None | 40.1 (38.9–41.5) | 20.9 (19.6–22.3) | 44.3 (42.9–45.6) |
| 1 | 23.2 (22.7–23.7) | 22.7 (21.6–23.9) | 23.3 (22.7–23.8) |
| ≥2 | 36.7 (35.4–37.9) | 56.4 (54.6–58.2) | 32.5 (31.2–33.8) |
| Insurance status, weighted % (95% CI) | |||
| Private insurance | 25.2 (23.8–26.6) | 15.5 (14.4–16.8) | 27.2 (25.6–28.8) |
| Medicare/Medicaid | 49.7 (47.3–52.1) | 61.9 (58.9–64.8) | 47.1 (44.7–49.5) |
| Other insurance | 3.3 (2.8–3.9) | 3.6 (2.8–4.5) | 3.2 (2.7–3.8) |
| Self‐pay/no charge | 10.0 (8.8–11.3) | 6.7 (5.8–7.8) | 10.7 (9.4–12.1) |
| Missing/unknown | 11.8 (9.2–15.1) | 12.3 (9.3–16.1) | 11.7 (9.2–14.9) |
| Any diagnostic services ordered or provided at visit, weighted % (95% CI) | 67.8 (66.0–69.6) | 74.0 (71.3–76.6) | 66.5 (64.7–68.2) |
| Missing (unweighted) | 1,094 | 164 | 930 |
| Any procedures provided at visit, weighted % (95% CI) | 50.4 (48.3–53.6) | 58.1 (55.6–60.5) | 48.8 (46.6–51.0) |
| Missing (unweighted) | 2,072 | 365 | 1,707 |
| Any medications provided at visit, weighted % (95% CI) | 65.6 (63.7–67.4) | 73.8 (71.6–75.9) | 63.8 (61.9–65.6) |
Note: Percentage estimates are with respect to column (ie, by group for a given measure). Sampling weights used were provided by the National Center for Health Statistics.
Abbreviations: CI; confidence interval; EMS, emergency medical services.
The Bottom Line
This analysis of a national emergency department database demonstrates that patients who present by emergency medical services (EMS) are older, more often male, more often had multiple chronic medical conditions, and less often had private insurance. EMS patients had higher priority triage scores, consumed more resources in the ED, and had longer lengths of stay.
Patients who lived in private residences used EMS less often compared to undomiciled patients or those residing in nursing homes. EMS patients more often had multiple chronic medical conditions documented (EMS 56.4%, 95% CI 54.6–58.2; other means 32.5%, 95% CI 31.2–33.8). EMS patients were less often privately insured, more often used public insurance, and were less often documented as self‐pay status. EMS patients were more likely to consume medical resources in all three measured domains: diagnostic studies, procedures, and medications.
Outcome measures stratified by mode of transportation to the ED are described in Table 2. EMS patients were more often associated with higher priority triage levels (ESI 1–2) and less often associated with lower priority triage levels (ESI 4–5). EMS patients were less often discharged home from the ED (EMS 58.0%, 95% CI 55.4–60.5; other means 81.8%, 95% CI 80.0–83.4) and were more often admitted, transferred, or pronounced deceased. EMS patients were associated with longer hospital length of stay (mean 6.1 days, 95% CI 5.8–6.5) compared to others (mean 4.6 days, 95% CI 4.4–4.9). EMS patients survived to hospital discharge in 96.7% (95% CI 96.1–97.3) of cases, compared to 99.3% (95% CI 99.0–99.5) among those presenting via any other means.
TABLE 2.
Outcome measures stratified by mode of transportation to the emergency department.
| Outcome | Overall | Arrived by EMS | Arrived by other means |
|---|---|---|---|
| Triage level, weighted % (95% CI) | |||
| 1 Immediate | 0.8 (0.6–1.1) | 2.3 (1.9–2.9) | 0.5 (0.3–0.8) |
| 2 Emergent | 9.9 (8.7–11.3) | 19.3 (16.9–21.9) | 7.9 (6.9–9.1) |
| 3 Urgent | 35.9 (32.6–39.2) | 38.8 (35.6–42.2) | 35.2 (31.9–38.7) |
| 4 Semiurgent | 20.3 (18.3–22.4) | 10.7 (9.3–12.3) | 22.3 (20.1–24.7) |
| 5 Nonurgent | 3.0 (2.5–3.6) | 1.4 (1.0–1.9) | 3.3 (2.8–4.0) |
| Missing/unknown | 30.2 (24.4–36.6) | 27.5 (22.3–33.4) | 30.7 (24.7–37.4) |
| ED disposition, weighted % (95% CI) | |||
| Discharge home | 77.6 (75.7–79.3) | 58.0 (55.4–60.5) | 81.8 (80–83.4) |
| Admit to inpatient/observation | 14 (12.7–15.3) | 30.1 (27.6–32.8) | 10.6 (9.6–11.6) |
| Interfacility transfer | 2.6 (2.2–3.0) | 5.3 (4.5–6.3) | 2.0 (1.7–2.4) |
| Died | 0.2 (0.2–0.2) | 1.0 (0.7–1.2) | 0 (0–0) |
| Other/missing | 5.6 (4.6–7.0) | 5.6 (4.6–6.8) | 5.7 (4.5–7.1) |
| Vital status at discharge, weighted % (95% CI) | |||
| Discharged alive | 98.8 (98.6–99.1) | 96.7 (96.1–97.3) | 99.3 (99.0–99.5) |
| Died (ED or inpatient) | 0.4 (0.4–0.5) | 2.0 (1.6–2.4) | 0.1 (0.1–0.2) |
| Missing/unknown | 0.7 (0.5–1.0) | 1.3 (0.9–1.8) | 0.6 (0.4–0.9) |
| Hospital length of stay in days, mean (95% CI) | 5.2 (5.0–5.4) | 6.1 (5.8–6.5) | 4.6 (4.4–4.9) |
| Missing or not applicable (unweighted) | 68,504 | 11,093 | 57,411 |
Abbreviations: CI, confidence interval; ED, emergency department; EMS, emergency medical services.
During 2020, 20.9% of ED patients arrived via EMS, compared to an averaged 17.7% during 2015–2019. A striking difference in the rate of diagnostic services use was noted among all ED patients: 67.8% (95% CI 66.0–69.6) in 2015–2019 compared to 85.0% (95% CI 82.6–87.0) in 2020. The significant difference between those arriving by EMS versus other means in the rate of diagnostic services used persevered in 2020. Mortality rates in 2020 were slightly higher compared to the preceding 5 years, although the difference did not reach statistical significance. Similarly, the greater mortality rate among those arriving via EMS compared to all others in the ED persisted in 2020. Characteristics and outcomes for NHAMCS 2020 are summarized in Table S1.
Arrival by EMS was associated with significantly higher odds of hospital admission (odds ratio [OR] 2.68, 95% CI 2.44–2.94) and in‐hospital death (OR 11.15, 95% CI 7.25–17.15), adjusting for patient age, sex, race or ethnicity, residence type, insurance status, and number of chronic medical conditions (Table 3). Odds of admission to the hospital were higher with increasing age, male sex, and 1 or more chronic medical conditions. Odds of admission were lower for patients of non‐Hispanic Black race. Odds of in‐hospital death were higher with increasing age and male sex.
TABLE 3.
Adjusted odds of the outcomes of admission and death. a
| Outcome | Admission, OR (95% CI) | Death, OR (95% CI) |
|---|---|---|
| Arrival mode | ||
| Arrival by other means | 1.00 (referent) | 1.00 (referent) |
| Arrival by EMS | 2.68 (2.44–2.94) | 11.15 (7.25–17.15) |
| Age (per 1 year increase) | 1.02 (1.02–1.02) | 1.04 (1.03–1.05) |
| Sex | ||
| Female | 1.00 (referent) | 1.00 (referent) |
| Male | 1.30 (1.23–1.37) | 2.38 (1.71–3.31) |
| Race or ethnicity | ||
| Non‐Hispanic White | 1.00 (referent) | 1.00 (referent) |
| Non‐Hispanic Black | 0.76 (0.67–0.85) | 0.98 (0.69–1.38) |
| Hispanic | 1.00 (0.87–1.14) | 0.63 (0.31–1.27) |
| Non‐Hispanic other | 1.07 (0.91–1.26) | 1.55 (0.68–3.50) |
| Patient residence | ||
| Private residence | 1.00 (referent) | 1.00 (referent) |
| Nursing home | 1.08 (0.9–1.29) | 1.30 (0.74–2.30) |
| Homeless/homeless shelter | 1.11 (0.85–1.46) | 0.17 (0.02–1.21) |
| Other | 1.40 (1.11–1.77) | 1.00 (0.44–2.28) |
| Missing/unknown | 1.25 (0.99–1.58) | 1.91 (1.01–3.60) |
| Chronic conditions | ||
| None | 1.00 (referent) | 1.00 (referent) |
| 1 | 1.95 (1.78–2.13) | 0.66 (0.39–1.11) |
| ≥2 | 3.57 (3.23–3.94) | 0.94 (0.62–1.44) |
| Insurance status | ||
| Private insurance | 1.00 (referent) | 1.00 (referent) |
| Medicare/Medicaid | 1.02 (0.93–1.12) | 0.76 (0.52–1.11) |
| Other insurance | 0.61 (0.48–0.76) | 0.59 (0.22–1.57) |
| Self‐pay | 0.80 (0.67–0.97) | 1.72 (0.91–3.24) |
| Other/missing | 1.01 (0.81–1.27) | 1.15 (0.64–2.05) |
Admission included admission to inpatient service (including intensive care or operating room), admission to observation, or interfacility transfer excluding those who died in the ED. Death included death in the ED or during hospitalization.
Abbreviations: CI, confidence interval; ED, emergency department; EMS, emergency medical services; OR, odds ratio.
3.1. Limitations
There are limitations associated with this study. First, NHAMCS relies upon self‐reported data collected from ED staff at participating hospitals, and the validity/accuracy of NHAMCS data has been subject to controversy. 10 , 11 However, this limitation must be balanced with the systematic methods used to collect this data, the approach to sampling that putatively offers national representation, and organized efforts over the past decade to assess and improve data quality. 12 , 13 Nevertheless, we acknowledge that NHAMCS data likely underrepresent large‐volume, high acuity EDs in large medical centers, which underscores the need for improved data registries of this sort. For example, the Clinical Emergency Data Registry administered by the American College of Emergency Physicians could be considered a suitable alternative, although NHAMCS was chosen for this study given its superior overall balance of strengths/limitations for these specific purposes. Second, patients who called EMS but were not transported to the hospital (eg, those who declined transport) were not included in this study given that NHAMCS data are sourced from hospital records and include only patients who ultimately proceed to the ED, therefore precluding any conclusions about this subset of patients. Third, the inability to link NHAMCS data with external sources, such as large prehospital care report or insurance data sets, prevents deeper analysis of prehospital factors or longitudinal assessment of individual patients. Moreover, NHAMCS variables used in this data set, such as resource use (ie, medications, procedures, and diagnostic services) and specific inpatient disposition (eg, inpatient floor versus intensive care unit), provided limited specificity to enable a more granular analysis of the differences between groups. However, we nevertheless found significant results using this coarse analysis and expect that greater differences would be detected using a more sensitive approach informed by finer detail. Fourth, although the use of ESI is widely established, ESI is partially subjective in nature and therefore may include bias, including direct bias due to arrival via EMS. However, we mitigate potential bias in this outcome measure through the use of several other more objective measures.
4. DISCUSSION
In this large, nationally representative study of more than one hundred million weighted ED visits per year, roughly 18% of patients arrived at the ED by EMS. Our findings support the hypothesis that EMS patients present to the ED with greater severity of illness and injury compared to ED patients arriving by other means, evidenced by higher priority ESI triage scores, higher resource use, and longer length of stay. Arrival by EMS was also associated with higher odds of admission and hospital mortality, even after controlling for other patient factors. Key findings were largely preserved in 2020 compared to the preceding 5 years, including differences in ED and inpatient mortality rates between groups. A higher rate of diagnostic services use overall was found in 2020, likely at least partially attributable to the widespread implementation of COVID testing, as well as a higher proportion of ED patients presenting via EMS during 2020.
Findings from the primary analysis, indicating that EMS patients are substantially sicker on average compared to other ED patients, pose a variety of interesting implications. EMS protocols often rely upon expert consensus or evidence that emerges from the ED setting. Given the important differences between the subset of ED patients transported via EMS versus others, evidence‐based guidelines developed using prehospital research focused on EMS patients would apply more directly to the EMS patient population and would therefore be most desirable. Moreover, this study highlights the importance of using the prehospital literature to contextualize and interpret findings from EMS research and should at least call for caution when comparing EMS patient populations to the overlapping, though distinct, general ED population. In addition, the higher degree of acuity among EMS patients should be kept in mind by ED clinicians and administrators, highlighting the importance of EMS‐ED handoff procedures that should be prioritized to maximize effective communication and minimize delays in care.
These findings also intersect with the field of public health. Ongoing public health research on resource‐intensive health care users should consider EMS users as a target for interventions to improve quality of care, efficiency, and cost‐effectiveness. These findings suggest that EMS users tend to present with higher acuity and consume more hospital resources than others, providing a wider target window for improvement compared to undifferentiated ED patients. Moreover, EMS patient data can be sourced from EMS agencies, including citywide municipal agencies that cross the boundaries of hospital systems and insurance providers, likely offering a more comprehensive cross‐section of the patients that compose a given community. These findings also raise the question of the patient‐driven and EMS‐mediated factors that might drive trends in who uses EMS versus other means to present to the ED. Recent reports have discussed the emergence of ride‐share apps as an increasingly popular alternative to ambulances, particularly for lower acuity patients. 14 Similarly, it has been reported that activation of EMS via 9‐1‐1, followed by declining transport and using a ride‐share app or other modes of self‐transport for transportation to the ED, might be a growing trend based on joint‐decision making between patients, EMS, and medical direction that is likely more prevalent among lower acuity patients. 15 These recent changes would contribute to higher concentration of higher acuity patients among those arriving to the ED via EMS.
Beyond the outcome measures, the factors associated with EMS use described in this study using a nationally representative sample largely align with limited prior reports that examined specific regions or populations (eg, geriatric patients or frequent EMS users). Similar to the associations seen here, these studies found that EMS use was associated with older age, public health insurance, greater use of hospital resources, higher triage score severity, and greater rates of hospitalization, as well as increased representation of patients who are male and African American. 3 , 4 , 5 , 16 , 17
Findings from this study are largely consistent with limited prior reports found outside the peer‐reviewed literature. Prior use of NHAMCS data has indicated that over time, older, sicker, and more complex patients tend to be concentrated in EDs, whereas lower risk patients who tend to require simpler care are increasingly being channeled into urgent care, virtual encounters, and other options for acute care. 18 , 19 , 20 Similarly, it has been noted that rates of EMS transport to the ED are increasing, and based on recent review of data from the 2022 Emergency Department Benchmarking Alliance, it seems that EMS patients have higher admission rates than others. 21
Novel aspects of this work include the use of nationally representative data, the direct comparison of adult ED patients by mode of presentation, and the consideration of key sociodemographic factors to compare patient outcomes among users of EMS versus others in the ED. These innovations generate new insights into the important differences in patients who present to the ED via EMS, ranging from sociodemographic factors to hospital resource use to patient outcome measures.
In conclusion, patients who arrive to the ED by EMS differ from other ED patients in multiple important aspects, including key demographic, socioeconomic, and clinical factors that include patient outcomes. These differences should be accounted for when contextualizing new and prior ED‐based study findings to prehospital‐based research, and vice versa. These findings also bear important implications in the domains of ED and EMS operations, as well as public health. Future prehospital‐based research will benefit from a more thorough understanding of how EMS patient samples differ from those in other acute care settings to guide more accurate interpretation of new results as the field continues to grow.
AUTHOR CONTRIBUTIONS
Gregory A. Peters, Scott A. Goldberg, and Rebecca E. Cash conceived the study. All authors assisted in designing the statistical analysis. Rebecca E. Cash obtained and analyzed the data. All authors contributed to interpretation of results. Gregory A. Peters drafted the manuscript and all authors contributed substantially to its revision. Rebecca E. Cash takes responsibility for the paper as a whole.
CONFLICT OF INTEREST STATEMENT
The authors declare no conflicts of interest.
Supporting information
Supporting Information
Biography
Gregory Peters, MD, is a resident physician at the Harvard Affiliated Emergency Medicine Residency in Emergency Medicine at Massachusetts General Hospital and Brigham & Women's Hospital in Boston, Massachusetts.
Peters GA, Goldberg SA, Hayes JM, Cash RE. Patients who use emergency medical services have greater severity of illness or injury compared to those who present to the emergency department via other means: A retrospective cohort study. JACEP Open. 2023;4:e13017. 10.1002/emp2.13017
Funding and support: By JACEP Open policy, all authors are required to disclose any and all commercial, financial, and other relationships in any way related to the subject of this article as per ICMJE conflict of interest guidelines (see www.icmje.org). The authors have stated that no such relationships exist.
Supervising Editor: Karl Sporer, MD
Meeting Presentations: This work was presented at the 2023 National Association of EMS Physicians Annual Meeting in Tampa, FL (January 25–28, 2023).
REFERENCES
- 1. National EMS Agenda Writing Team . Sayre MR, White LJ, Brown LH, McHenry SD. National EMS research agenda. Prehosp Emerg Care. 2002;6(3 suppl):S1‐43. [PubMed] [Google Scholar]
- 2. Goldberg SA, Chadwick E, Makam S, et al. Twenty Years After the EMS research agenda: trends in EMS publications and research funding. Ann Emerg Med. 2023;82(1):94‐100. [DOI] [PubMed] [Google Scholar]
- 3. Shah MN, Bazarian JJ, Lerner EB, et al. The epidemiology of emergency medical services use by older adults: an analysis of the National Hospital Ambulatory Medical Care Survey. Acad Emerg Med. 2007;14(5):441‐447. [DOI] [PubMed] [Google Scholar]
- 4. Ruger JP, Richter CJ, Lewis LM. Clinical and economic factors associated with ambulance use to the emergency department. Acad Emerg Med. 2006;13(8):879‐885. [DOI] [PubMed] [Google Scholar]
- 5. Déziel JD. Ambulance transport to the emergency department: a patient‐selected signal of acuity and its effect on resource provision. Am J Emerg Med. 2019;37(6):1096‐1100. [DOI] [PubMed] [Google Scholar]
- 6. CDC . NAMCS/NHAMCS ‐ Ambulatory Health Care Data Homepage [Internet]. 2023 [cited 2023 Feb 9]. https://www.cdc.gov/nchs/ahcd/index.htm
- 7. McCaig LF. National Hospital Ambulatory Medical Care Survey: 1998 emergency department summary. Adv Data. 2000;10(313):1–23. [PubMed] [Google Scholar]
- 8. CDC . NAMCS/NHAMCS ‐ Data Collection and Processing. [Internet]. 2019 [cited 2023 Mar 22]. https://www.cdc.gov/nchs/ahcd/ahcd_data_collection.htm
- 9. Shah MN, Cushman JT, Davis CO, Bazarian JJ, Auinger P, Friedman B. The epidemiology of emergency medical services use by children: an analysis of the National Hospital Ambulatory Medical Care Survey. Prehosp Emerg Care. 2008;12(3):269‐276. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Cooper RJ. NHAMCS: does it hold up to scrutiny? Ann Emerg Med. 2012;60(6):722‐725. [DOI] [PubMed] [Google Scholar]
- 11. Green SM. Congruence of disposition after emergency department intubation in the National Hospital Ambulatory Medical Care Survey. Ann Emerg Med. 2013;61(4):423‐426.e8. [DOI] [PubMed] [Google Scholar]
- 12. McCaig LF, Burt CW, Schappert SM, et al. NHAMCS: does it hold up to scrutiny? Ann Emerg Med. 2013;62(5):549‐551. [DOI] [PubMed] [Google Scholar]
- 13. Carlson JN, Wang HE. The challenge of analyzing and Interpreting NHAMCS. Ann Emerg Med. 2013;62(1):99‐100. [DOI] [PubMed] [Google Scholar]
- 14. Moskatel L, Slusky D. Did UberX reduce ambulance volume? Health Economicyyy. 2019;28(7):817‐829. [DOI] [PubMed] [Google Scholar]
- 15. Degesys N, Is Uber the New Ambulance? Financial Concerns May Be Driving This Trend for Many Emergency Department Patients [Internet]. ACEP Now. 2017 [cited 2023 Jun 16]. https://www.acepnow.com/article/uber‐new‐ambulance‐financial‐concerns‐may‐driving‐trend‐many‐emergency‐department‐patients [Google Scholar]
- 16. Shah MN, Glushak C, Karrison TG, et al. Predictors of emergency medical services utilization by elders. Acad Emerg Med. 2003;10(1):52‐58. [DOI] [PubMed] [Google Scholar]
- 17. Knowlton A, Weir BW, Hughes BS, et al. Patient demographic and health factors associated with frequent use of emergency medical services in a mid‐sized city. Acad Emerg Med. 2013;20(11):1101‐1111. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Augustine J, Long‐Term Trends in Emergency Department Visits, Patient Care Highlighted in National Reports [Internet]. ACEP Now. 2017 [cited 2023 Jun 22]. https://www.acepnow.com/article/long‐term‐trends‐emergency‐department‐visits‐patient‐care‐highlighted‐national‐reports [Google Scholar]
- 19. Augustine J, The Latest Emergency Department Utilization Numbers Are In [Internet]. ACEP Now. 2019 [cited 2023 Jun 22]. https://www.acepnow.com/article/the‐latest‐emergency‐department‐utilization‐numbers‐are‐in [Google Scholar]
- 20. Augustine J, Emergency Department Usage Trends Before COVID‐19 [Internet]. ACEP Now. 2020 [cited 2023 Jun 22]. https://www.acepnow.com/article/emergency‐department‐usage‐trends‐before‐covid‐19 [Google Scholar]
- 21. Augustine J, A First Look at Emergency Department Data for 2022 [Internet]. ACEP Now. 2023 [cited 2023 Jun 22]. https://www.acepnow.com/article/a‐first‐look‐at‐emergency‐department‐data‐for‐2022 [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting Information