Abstract
Background:
Outpatient management of select patients with low-risk acute pulmonary embolism (PE) has been proven to be safe and effective, yet recent evidence suggests that patients are still managed with hospitalization. Few studies have assessed contemporary real-world trends in discharge rates from U.S. emergency departments (EDs) for acute PE.
Objective:
To evaluate whether the proportion of discharges from EDs for acute PE changed between 2012 and 2020 and which baseline characteristics are associated with ED discharge.
Design:
Serial cross-sectional analysis.
Setting:
U.S. EDs participating in the National Hospital Ambulatory Medical Care Survey.
Patients:
Patients with ED visits for acute PE between 2012 and 2020.
Measurements:
National trends in the proportion of discharges for acute PE and factors associated with ED discharge.
Results:
Between 2012 and 2020, there were approximately 1 635 300 visits for acute PE. Overall, ED discharge rates remained constant over time, with rates of 38.2% (95% CI, 17.9% to 64.0%) between 2012 and 2014 and 33.4% (CI, 21.0% to 49.0%) between 2018 and 2020 (adjusted risk ratio, 1.01 per year [CI, 0.89 to 1.14]). No baseline characteristics, including established risk stratification scores, were predictive of an increased likelihood of ED discharge; however, patients at teaching hospitals and those with private insurance were more likely to receive oral anticoagulation at discharge. Only 35.9% (CI, 23.9% to 50.0%) of patients who were considered low-risk according to their Pulmonary Embolism Severity Index (PESI) class, 33.1% (CI, 21.6% to 47.0%) according to simplified PESI score, and 34.8% (CI, 23.3% to 48.0%) according to hemodynamic stability were discharged from the ED setting.
Limitations:
Cross-sectional survey design and inability to adjudicate diagnoses.
Conclusion:
In a representative nationwide sample, rates of discharge from the ED for acute PE appear to have remained constant between 2012 and 2020. Only one third of low-risk patients were discharged for outpatient management, and rates seem to have stabilized. Outpatient management of low-risk acute PE may still be largely underutilized in the United States.
Primary Funding Source:
None.
A cute pulmonary embolism (PE) is a leading cause of cardiovascular-related mortality and is responsible for nearly 300 000 deaths annually (1, 2). As a result, clinical management for patients with acute PE often necessitates inpatient hospitalization for anticoagulation or other advanced therapies (3, 4). However, in recent years, it has become increasingly clear that outpatient management for select low-risk patients with acute PE is a safe and feasible approach (5–7). To date, multiple randomized controlled trials have been completed and have collectively shown that outpatient management among select patients is associated with low rates of recurrent venous thromboembolism, major bleeding, and mortality compared with hospitalization (8, 9). These findings have resulted in changes in international guidelines published by various clinical societies, including the European Society of Cardiology (ESC) and the American College of Chest Physicians (CHEST) (10, 11). Adoption of these guidelines has the potential to improve patient care and to substantially reduce overall health care spending by limiting costly inpatient admissions (12, 13).
Despite this increasing evidence, recent work has suggested that many patients with low-risk PE are still managed with inpatient hospitalization (14–16). For example, in a limited analysis among community health systems, nearly 85% of patients with low-risk PE were hospitalized despite updates to clinical practice guidelines (17). This observation is particularly important as U.S. health care costs continue to surge and overcrowding in emergency departments (EDs) remains a critical issue (18, 19). Although previous work has shown an increase in outpatient utilization for acute PE, few contemporary studies have focused specifically on this decision-making process within the ED, a frequent location for these triage decisions (20, 21). In addition, little is known about national patterns in discharge rates for acute PE in the United States. As the landscape of PE management continues to rapidly evolve, and now that immediate oral anticoagulation with direct oral anticoagulants has been firmly established as first-line therapy, understanding national management patterns is critical.
In this study, we utilized a nationally representative data set to assess recent real-world national trends in discharge rates for acute PE from U.S. EDs. We specifically sought to evaluate whether the proportion of discharges for low-risk acute PE increased between 2012 and 2020 and to determine whether any baseline characteristics are associated with ED discharge.
Methods
Data Source
We conducted a serial cross-sectional analysis leveraging data from the National Hospital Ambulatory Medical Care Survey (NHAMCS) between 2012 and 2020. The NHAMCS is a nationally representative probability sample that reflects ED visits to nonfederal, nonmilitary, and non–Veterans Health Administration hospitals. The NHAMCS is published annually by the National Center for Health Statistics (NCHS) at the Centers for Disease Control and Prevention. Methods for primary data collection of the NHAMCS have been described in detail elsewhere (22, 23). Briefly, the NHAMCS uses a 3-stage probability design to formulate a representative sample of weighted patient visits to U.S. EDs. A sample of geographically defined areas serve as the primary sampling units where emergency service areas are identified and assigned a random 4-week interval to report on a random sample of ED visits. Survey completion and data reporting are done entirely by trained hospital personnel. Data elements collected by the NHAMCS vary by year but include patient sociodemographic information, hospital information, comorbidities, vital signs (as continuous variables), physician diagnoses, clinical presentations, treatments received, diagnostic tests ordered, and ED dispositions. Appendix Table 1 (available at Annals. org) shows the NHAMCS codes that were used to identify the variables of interest for the present study. In this analysis, there were no missing data for age, sex, race, and ethnicity. However, several comorbidity variables, including coronary artery disease, chronic kidney disease, alcohol use disorder, asthma, osteoporosis, depression, hyperlipidemia, hypertension, obesity, and obstructive sleep apnea, were missing in fewer than 7% of patients as these variables were not coded in the 2012 and 2013 surveys.
A total of 9 survey data files were obtained and used for this analysis. All data sets are publicly available on the NCHS website (www.cdc.gov/nchs/ahcd). All data were deidentified and publicly available, so institutional review board approval was not required for this analysis.
Study Population and Subgroups
Patients of all ages who presented with a diagnosis of acute PE were first identified. Visits for acute PE were identified from any PE-related billing claims based on International Classification of Diseases (ICD) codes on the NHAMCS survey. Appendix Table 1 shows the ICD-9 and ICD-10 codes that were used to identify patients. Claims codes used for this analysis were selected on the basis of prior work (24–26). After identification, patients were excluded if they had an unspecified disposition. Patients were censored from the analysis if they died while in the ED, died on arrival, left before evaluation by staff, left before treatment completion, left against medical advice, or were transferred to another hospital system directly from the ED.
In order to better investigate patients who were at low risk for PE-related death, subgroup analyses were performed using previously established risk stratification tools (27, 28). Patients were stratified on the basis of hemodynamic stability (defined as heart rate <110 beats/min, systolic blood pressure >100 mm Hg, and oxygen saturation >90%), a low-risk Pulmonary Embolism Severity Index (PESI) class (class I or II), and a low-risk simplified PESI (sPESI) score (score of 0). An overview of how these scores were calculated with the survey variables is provided in Appendix Table 1.
Primary Outcomes and Exposure
The primary end point of this analysis was the proportion of ED visits for acute PE that resulted in patient discharge directly from the ED. This variable was defined as an ED visit that did not result in hospitalization and did not have any other coded visit dispositions within the NHAMCS. For patients admitted to an ED observation unit, the final disposition after this placement was utilized for outcome classification. The primary exposure of this analysis was the year of the ED visit. For illustrative purposes, visits were aggregated into 3-year groups and visualized longitudinally with time.
Statistical Analysis
All analyses were conducted using the multistage probability survey design with the associated visit weights to obtain national estimates. Summary statistics were used to compare baseline characteristics. Continuous variables are reported as means ± SEs, and categorical variables are reported as counts with percentages. Baseline covariates were compared temporally using standardized mean differences. All outcomes were reported as percentages with 95% confidence intervals. Confidence intervals were calculated using a logistic regression model to estimate a Wald-type interval.
To investigate temporal trends, survey-weighted quasi-binomial and quasi-Poisson regression models were used to obtain an estimate of the risk ratio with year of the visit included as a continuous linear variable. To investigate variables associated with discharge from the ED, similar survey-weighted quasi-binomial and quasi-Poisson regression models were also used. Finally, to investigate variables associated with oral anticoagulant prescription among discharged patients, similar survey-weighted regression models were used. For all regression analyses, results of both univariable and multivariable models are reported. All multivariable models were adjusted for age, sex, race, ethnicity, and geographic region.
Because this study utilized claims-based data and diagnoses were not able to be thoroughly adjudicated, we performed 2 sensitivity analyses to evaluate the reliability of the findings. First, we performed an analysis that included only patients who received a computed tomography scan in the ED. Second, we performed an analysis restricted to patients who had acute PE as the first coded diagnosis in the NHAMCS survey.
All statistical analyses were done using R, version 4.1.0, in RStudio (R Foundation for Statistical Computing), with the “survey” package for analysis. Figures were plotted using Python, version 3.10 (Python Software Foundation). Statistical significance was set at a 2-tailed P value less than 0.05.
Role of the Funding Source
This study received no funding.
Results
Study Population
Figure 1 illustrates selection of the study population. Between 2012 and 2020, there were an estimated 1 827 207 ED visits for acute PE in the United States. Among those visits, 29 523 (1.62%) were excluded because they resulted in an unspecified disposition. After exclusions, 1 797 684 patients remained. Of the remaining visits, 28 675 (1.57%) were censored from the analytic cohort due to death in the ED, 26 312 (1.44%) due to the patient leaving before treatment completion or against medical advice, and 107 397 (5.88%) due to transfer to another hospital system. The final analytic cohort included a total of 1 635 300 weighted visits for acute PE during the study period.
Figure 1. Patient flow diagram for cohort formation, 2012–2020.
ED = emergency department; PE = pulmonary embolism.
Baseline Characteristics of the Study Population
Table 1 shows baseline characteristics, hospital information, and risk stratification for patients in the study cohort, stratified over time. Overall, the estimated number of ED visits for acute PE increased from 370 000 between 2012 and 2014 to 745 833 between 2018 and 2020. Among the entire cohort, the mean age (±SE) was 57.7 ± 2.19 years, 39.1% (n = 640 092) were male, 73.9% (n = 1 208 183) were White, and 9.18% (n = 150188) were Hispanic or Latino. A total of 17.3% (n = 283212) of patients were treated in a teaching hospital, and 7.15% (n = 112 645) were treated in a rural location.
Table 1.
Patient Demographic Characteristics, Baseline Data, Clinical Presentation, and Risk Stratification
| Variable | 2012–2014 | 2015–2017 | 2018–2020 | Overall | Standardized Mean Difference* | |
|---|---|---|---|---|---|---|
| Crude number of visits | 71 | 82 | 97 | 250 | — | |
| Estimated number of visits | 370 000 | 519 467 | 745 833 | 1 635 300 | — | |
| Demographic characteristics | ||||||
| Mean age (±SE), y | 50.9 ± 4.82 | 56.9 ± 4.12 | 61.8 ± 2.32 | 57.7 ± 2.19 | 0.30 | |
| Male sex, n (%) | 125 742 (34.0) | 243 449 (46.9) | 270 901 (36.3) | 640 092 (39.1) | 0.17 | |
| Race, n (%) | ||||||
| White | 317 594 (85.8) | 392 819 (75.6) | 497 769 (66.7) | 1 208 183 (73.9) | 0.30 | |
| Black | 38 683 (10.5) | 126 648 (24.4) | 221 965 (29.8) | 387 296 (23.7) | 0.33 | |
| Other | 13 723 (3.7) | 0 (0.0) | 26 098 (3.5) | 39 821 (2.4) | 0.19 | |
| Ethnicity, n (%) | ||||||
| Hispanic or Latino | 38 660 (10.4) | 68 407 (13.2) | 43 121 (5.8) | 150 188 (9.2) | 0.17 | |
| Not Hispanic or Latino | 331 340 (89.6) | 451 060 (86.8) | 702 719 (94.2) | 1 485 112 (90.8) | — | |
| Comorbidities, n (%) | ||||||
| Cancer | 48 046 (13.0) | 80 872 (15.6) | 49 905 (6.7) | 178 823 (10.9) | 0.19 | |
| Dementia | 8327 (2.3) | 31 349 (6.0) | 5000 (0.7) | 44 677 (2.7) | 0.21 | |
| Cerebrovascular disease | 0 (0.0) | 58 286 (11.2) | 31 281 (4.2) | 89 568 (5.5) | 0.36 | |
| COPD | 104 848 (28.3) | 104 857 (20.2) | 140 286 (18.8) | 349 992 (21.4) | 0.15 | |
| Congestive heart failure | 44 918 (12.1) | 58 981 (11.4) | 95 830 (12.8) | 199 729 (12.2) | 0.03 | |
| Diabetes mellitus | 81 301 (22.0) | 90 953 (17.5) | 138 797 (18.6) | 311 052 (19.0) | 0.08 | |
| History of VTE | 88 159 (23.8) | 124 299 (23.9) | 232 802 (31.2) | 445 261 (27.2) | 0.11 | |
| History of HIV | 0 (0.0) | 2811 (0.5) | 0 (0.0) | 2812 (0.2) | 0.07 | |
| End-stage renal disease | 9885 (2.7) | 0 (0.0) | 17 574 (2.4) | 27 459 (1.7) | 0.16 | |
| Chronic kidney disease | — | 69 439 (13.4) | 57 057 (7.7) | 136 614 (8.9) | 0.19 | |
| Coronary artery disease | — | 100 619 (19.4) | 95 664 (12.8) | 211 877 (13.8) | 0.18 | |
| Alcohol use disorder | — | 3636 (0.7) | 25 022 (3.4) | 29 879 (1.9) | 0.19 | |
| Asthma | — | 63 584 (12.2) | 115 773 (15.5) | 234 926 (15.3) | 0.10 | |
| Osteoporosis | — | 0 (0.0) | 70 042 (9.4) | 70 042 (4.6) | 0.46 | |
| Depression | — | 80 110 (15.4) | 72 487 (9.7) | 164 593 (10.7) | 0.17 | |
| Hyperlipidemia | — | 116 735 (22.5) | 140 319 (18.8) | 271 481 (17.7) | 0.09 | |
| Hypertension | — | 220 532 (42.5) | 379 072 (50.8) | 638 371 (41.6) | 0.17 | |
| Obesity | — | 108 834 (21.0) | 151 579 (20.3) | 295 731 (19.3) | 0.02 | |
| Obstructive sleep apnea | — | 33 646 (6.5) | 5066 (0.7) | 60 746 (4.0) | 0.32 | |
| Mean total chronic comorbidities (±SE), n | 1.62 ± 0.30 | 2.66 ± 0.27 | 2.34 ± 0.42 | 2.28 ± 0.23 | 0.33 | |
| Geographic region, n (%) | ||||||
| Northeast | 108 718 (29.4) | 112 236 (21.6) | 146 952 (19.7) | 367 906 (22.5) | 0.15 | |
| Midwest | 49 100 (13.3) | 148 719 (28.6) | 142 883 (19.2) | 340 702 (20.8) | 0.26 | |
| South | 90 756 (24.5) | 176 939 (34.1) | 273 566 (36.7) | 541 261 (33.1) | 0.18 | |
| West | 121 426 (32.8) | 81 573 (15.7) | 182 433 (24.5) | 385 432 (23.6) | 0.27 | |
| Rural location, n (%) | 0 (0.0) | 42 753 (8.2) | 69 893 (9.4) | 112 645 (7.2) | 0.31 | |
| Teaching hospital, n (%) | 74 245 (20.1) | 98 084 (18.9) | 110 883 (14.9) | 283 212 (17.3) | 0.09 | |
| Primary payment, n (%) | ||||||
| Private insurance | 74 245 (20.1) | 98 084 (18.9) | 110 883 (14.9) | 432 792 (28.5) | 0.09 | |
| Medicare | 123 114 (35.5) | 186 496 (37.7) | 123 182 (18.2) | 716 585 (47.2) | 0.30 | |
| Medicaid/other state program | 117 409 (33.8) | 200 825 (40.6) | 398 351 (59.0) | 261 670 (17.2) | 0.35 | |
| Other | 45 954 (13.2) | 91 341 (18.5) | 124 376 (18.4) | 106 273 (7.0) | 0.10 | |
| Clinical presentation, n (%) | ||||||
| Chest pain | 111 851 (30.2) | 129 229 (24.9) | 171 469 (23.0) | 412 549 (25.2) | 0.11 | |
| Dyspnea | 220 770 (59.7) | 162 429 (31.3) | 324 387 (43.5) | 707 586 (43.3) | 0.39 | |
| Hemoptysis | 4507 (1.2) | 0 (0.0) | 0 (0.0) | 4507 (0.3) | 0.11 | |
| Altered mental status | 0 (0.0) | 4310 (0.8) | 5597 (0.8) | 9908 (0.6) | 0.09 | |
| Vital sign abnormalities, n (%) | ||||||
| Fever (temperature >100.4 °F) | 3737 (1.1) | 6960 (1.4) | 0 (0.0) | 10 697 (0.7) | 0.12 | |
| Tachycardia (HR >100 beats/min) | 144 874 (42.7) | 143 529 (27.9) | 220 133 (32.6) | 508 530 (33.2) | 0.21 | |
| Hypotension (SBP <80 mm Hg) | 0 (0.0) | 0 (0.0) | 6677 (1.0) | 6677 (0.4) | 0.09 | |
| Tachypnea (RR >20 breaths/min) | 146 873 (40.3) | 134 002 (25.8) | 250 997 (36.4) | 531 872 (33.8) | 0.21 | |
| Oxygen saturation <90% | 67 062 (19.0) | 8089 (1.6) | 25 757 (3.7) | 100 908 (6.5) | 0.41 | |
| Hemodynamically stable, n (%) † | 171 450 (55.2) | 354 450 (70.2) | 393 392 (58.2) | 919 292 (61.7) | 0.21 | |
| PESI class, n (%) | ||||||
| I or II | 195 953 (65.6) | 304 711 (62.0) | 410 234 (61.2) | 910 898 (62.4) | 0.06 | |
| III to V | 102 770 (34.4) | 187 149 (38.0) | 260 111 (38.8) | 550 030 (37.6) | — | |
| sPESI score, n (%) | ||||||
| 0 | 104 124 (28.1) | 263 366 (50.7) | 366 286 (49.1) | 733 776 (44.9) | 0.32 | |
| ≥1 | 265 876 (71.9) | 256 101 (49.3) | 379 547 (50.9) | 901 524 (55.1) | — | |
COPD = chronic obstructive pulmonary disease; HR = heart rate; PE = pulmonary embolism; PESI = Pulmonary Embolism Severity Index; RR = respiratory rate; SBP = systolic blood pressure; sPESI = simplified Pulmonary Embolism Severity Index; VTE = venous thromboembolism.
Reported values are an average of the 3 pairwise between-group standardized mean differences.
Defined as SBP >100 mm Hg, HR <110 beats/min, and oxygen saturation >90%.
Patients presenting with acute PE had high rates of comorbid cardiopulmonary disease, including 21.4% (n = 349 992) with chronic obstructive pulmonary disease (COPD), 12.2% (n = 199 729) with congestive heart failure, 13.8% (n = 211 877) with coronary artery disease, 17.7% (n = 271 481) with hyperlipidemia, and 41.6% (n = 638 371) with hypertension. Included patients had an average (±SE) of 2.28 ± 0.23 chronic comorbidities. Many of the baseline comorbidities remained relatively constant over the 9-year study period, including dementia, congestive heart failure, diabetes, end-stage renal disease, obesity, asthma, history of HIV, and history of venous thromboembolism. Rates of cancer (13.0% between 2012 and 2014 and 6.69% between 2018 and 2020) and COPD (28.3% between 2012 and 2014 and 18.8% between 2018 and 2020) decreased over time.
The most frequent clinical symptom on presentation was dyspnea, which occurred in 43.3% (n = 707 586) of patients, followed by chest pain, which occurred in 25.2% (n = 412 549). Among vital sign abnormalities, fever was recorded in 0.71% (n = 10 697) of patients, tachycardia in 33.2% (n = 508 530), hypotension in 0.43% (n = 6677), tachypnea in 33.8% (n = 531 872), and oxygen desaturation in 6.50% (n = 100 908). In order to better understand PE severity, patients were stratified on the basis of hemodynamic stability, PESI risk class, and sPESI score. In the cohort, 61.7% (n = 919 292) of patients were hemodynamically stable, 62.4% (n = 910 898) had PESI risk class I or II, and 44.9% (n = 733 776) had a sPESI score of 0. Rates of hemodynamic stability (55.2% between 2012 and 2014 and 58.2% between 2018 and 2020) and PESI risk class I or II (65.6% between 2012 and 2014 and 61.2% between 2018 and 2020) remained constant over the study period (Table 1).
National Trends in ED Discharge Rates for Acute PE
Figure 2 shows the estimated rates of discharge for all patients with acute PE presenting to U.S. EDs from 2012 to 2020. Overall, 30.7% (95% CI, 22.3% to 41.0%) of patients with acute PE were discharged from the ED over the 9-year period. Discharge rates seem to have remained constant over time, with rates of 38.2% (CI, 17.9% to 64.0%) between 2012 and 2014 and 33.4% (CI, 21.0% to 49.0%) between 2018 and 2020. There was no increase in ED discharges for acute PE over time (adjusted risk ratio, 1.01 per year [CI, 0.89 to 1.14]; P = 0.85) (Table 2).
Figure 2. National temporal trends in ED discharge rates for all patients with acute PE.
Error bars indicate 95% CIs. ED = emergency department; PE = pulmonary embolism.
Table 2.
Univariable and Multivariable Regression Analysis for Predictors of Discharge Among All Patients With Acute PE in the ED
| Variable | Admitted (n = 1 132 893) |
Discharged (n = 502 407) |
Unadjusted Risk Ratio (95% CI) |
Adjusted Risk Ratio (95% CI)* |
|---|---|---|---|---|
| Year | – | – | 0.99 (0.86–1.15) | 1.01 (0.89–1.14) |
| Demographic characteristics | ||||
| Mean age (±SE), y | 59.4 ± 2.31 | 54.0 ± 4.89 | 0.99 (0.97–1.00) | 0.99 (0.98–1.00) |
| Male sex, n (%) | 496 661 (43.8) | 143 431 (28.5) | 0.62 (0.36–1.06) | 0.65 (0.39–1.10) |
| Race, n (%) | ||||
| White | 825 856 (72.9) | 382 327 (76.1) | 1.12 (0.59–2.14) | 1.24 (0.63–2.43) |
| Black | 273 607 (24.2) | 113 689 (22.6) | 0.94 (0.46–1.90) | 0.85 (0.40–1.81) |
| Other | 33 431 (3.0) | 6391 (1.3) | 0.51 (0.06–4.04) | 0.48 (0.06–3.37) |
| Hispanic or Latino ethnicity, n (%) | 116 422 (10.3) | 33 766 (6.7) | 0.71 (0.24–2.07) | 0.70 (0.24–2.04) |
| Comorbidities, n (%) | ||||
| Cancer | 107 477 (9.5) | 71 346 (14.2) | 1.63 (0.88–3.01) | 1.62 (0.88–3.01) |
| Dementia | 44 677 (3.9) | 0 (0.0) | – | – |
| Cerebrovascular disease | 81 906 (7.2) | 7662 (1.5) | 0.26 (0.03–1.82) | 0.31 (0.04–2.20) |
| COPD | 220 236 (19.4) | 129 755 (25.8) | 1.27 (0.65–2.50) | 1.23 (0.71–2.13) |
| Congestive heart failure | 141 833 (12.5) | 57 897 (11.5) | 0.93 (0.48–1.79) | 1.05 (0.53–2.06) |
| Diabetes mellitus | 193 964 (17.1) | 117 088 (23.3) | 1.29 (0.66–2.52) | 1.65 (0.77–3.55) |
| History of VTE | 257 236 (22.7) | 188 025 (37.4) | 1.59 (0.86–2.94) | 1.55 (0.87–2.74) |
| History of HIV | 2812 (0.2) | 0 (0.0) | – | – |
| End-stage renal disease | 27 459 (2.4) | 0 (0.0) | – | – |
| Chronic kidney disease | 125 231 (11.8) | 11 383 (2.4) | 0.25 (0.04–1.35) | 0.26 (0.04–1.58) |
| Coronary artery disease | 170 925 (16.1) | 40 951 (8.6) | 0.58 (0.26–1.31) | 0.75 (0.29–1.95) |
| Alcohol use disorder | 28 658 (2.7) | 1221 (0.3) | 0.12 (0.01–1.23) | 0.18 (0.01–1.96) |
| Asthma | 159 345 (15.0) | 75 581 (15.9) | 1.04 (0.52–2.06) | 0.90 (0.47–1.71) |
| Osteoporosis | 41 398 (3.9) | 28 644 (6.0) | 1.34 (0.28–6.38) | 1.20 (0.23–6.14) |
| Depression | 109 497 (10.3) | 55 095 (11.6) | 1.09 (0.51–2.30) | 1.15 (0.56–2.35) |
| Hyperlipidemia | 196 562 (18.6) | 74 919 (15.8) | 0.87 (0.38–1.95) | 0.96 (0.40–2.30) |
| Hypertension | 435 482 (41.1) | 202 888 (42.7) | 1.04 (0.58–1.86) | 1.27 (0.73–2.20) |
| Obesity | 243 026 (23.0) | 52 705 (11.1) | 0.52 (0.15–1.70) | 0.5 (0.15–1.61) |
| Obstructive sleep apnea | 60 746 (5.7) | 0 (0.0) | – | – |
| Geographic region, n (%) | ||||
| Northeast | 243 535 (21.5) | 124 371 (24.8) | 1.13 (0.62–2.04) | 1.94 (0.59–6.35) |
| Midwest | 262 349 (23.2) | 78 352 (15.6) | 0.70 (0.30–1.60) | 0.72 (0.31–1.66) |
| South | 398 139 (35.1) | 143 122 (28.5) | 0.80 (0.43–1.49) | 0.84 (0.44–1.58) |
| West | 228 869 (20.2) | 156 563 (31.2) | 1.46 (0.73–2.92) | 1.59 (0.59–4.25) |
| Teaching hospital, n (%) | 172 659 (15.2) | 110 553 (22.0) | 1.34 (0.68–2.66) | 1.47 (0.72–3.01) |
| Primary payment, n (%) | ||||
| Private insurance | 254 224 (24.8) | 178 567 (36.4) | 1.30 (0.78–2.18) | 1.19 (0.75–1.91) |
| Medicare | 507 699 (49.5) | 208 887 (42.5) | 0.91 (0.50–1.65) | 1.27 (0.64–2.52) |
| Medicaid/other state program | 173 792 (16.9) | 87 878 (17.9) | 0.99 (0.55–1.76) | 0.94 (0.50–1.79) |
| Other | 90 681 (8.8) | 15 592 (3.2) | 0.43 (0.09–1.90) | 0.50 (0.13–1.87) |
| Rural location, n (%) | 73 789 (6.8) | 38 856 (8.0) | 1.12 (0.31–4.05) | 1.05 (0.28–3.87) |
| Hemodynamically stable, n (%)† | 599 540 (59.6) | 319 751 (66.0) | 1.20 (0.66–2.17) | 1.15 (0.65–2.04) |
| PESI class I or II, n (%) | 583 457 (59.5) | 327 441 (68.1) | 1.28 (0.73–2.25) | 0.96 (0.43–2.13) |
| sPESI score of 0, n (%) | 490 788 (43.3) | 242 989 (48.4) | 1.15 (0.64–2.06) | 1.06 (0.58–1.93) |
COPD = chronic obstructive pulmonary disease; ED = emergency department; PE = pulmonary embolism; PESI = Pulmonary Embolism Severity Index; sPESI = simplified Pulmonary Embolism Severity Index; VTE = venous thromboembolism.
All multivariable models were adjusted for age, sex, race, ethnicity, and geographic region.
Defined as systolic blood pressure >100 mm Hg, heart rate <110 beats/min, and oxygen saturation >90%.
Two sensitivity analyses were also performed to further validate the trends observed in the primary analysis. Among patients with a diagnosis of acute PE who also had a computed tomography scan during their ED encounter, similar findings were observed with relatively constant rates of discharge over the study duration (adjusted risk ratio, 1.02 per year [CI, 0.87 to 1.19]; P = 0.82) (Appendix Figure 1, available at Annals.org). Conversely, among patients with acute PE as the first coded diagnosis, there was a significant increase in the proportion discharged, from 9.65% (CI, 3.25% to 25.0%) between 2012 and 2014 to 20.6% (CI, 12.3% to 32.0%) between 2018 and 2020 (adjusted risk ratio, 1.24 [CI, 1.06 to 1.47]; P = 0.009) (Appendix Figure 2, available at Annals.org).
Discharge Rates Among Patients With Low-Risk PE
Figure 3 shows temporal trends in discharge rates of patients with PE who were identified as low-risk. When stratified on the basis of hemodynamic stability, rates of discharge of patients with low-risk PE remained constant over the study period (adjusted risk ratio, 0.95 [CI, 0.81 to 1.13]; P = 0.60) (Figure 3, top). From 2012 to 2020, only 35.9% (CI, 23.9% to 50.0%) of patients with a low-risk PESI class and 33.1% (CI, 21.6% to 47.0%) with a low-risk sPESI score were discharged from the ED setting. Among these groups, trends in discharge rates remained relatively constant over the 9 years investigated for patients determined to be at low risk by the PESI score (adjusted risk ratio, 1.00 [CI, 0.84 to 1.18]; P = 0.98) and the sPESI score (adjusted risk ratio, 1.04 [CI, 0.85 to 1.27]; P = 0.67) (Figure 3, middle and bottom).
Figure 3. National temporal trends in ED discharge rates for hemodynamically stable patients (top), patients with PESI class I or II (middle), and patients with a sPESI score of 0 (bottom).
Error bars indicate 95% CIs. ED = emergency department; PE = pulmonary embolism; PESI = Pulmonary Embolism Severity Index; sPESI = simplified Pulmonary Embolism Severity Index.
Predictors of Discharge Among Patients With Acute PE
Table 2 shows the results of the univariable and multivariable regression analysis to assess for clinical variables associated with ED discharge for acute PE. Among patients who were hospitalized, the mean length of stay (±SE) was 6.31 ± 1.01 days. Among ED visits between 2012 and 2020, no clinical variables, including any demographic information (such as age, sex, or race), hospital features (such as teaching hospital status, region, or rural location), or comorbidities (such as chronic kidney disease, congestive heart failure, or COPD), were associated with ED discharge. Likewise, risk stratification frameworks, including hemodynamically stable patients (adjusted risk ratio, 1.15 [CI, 0.65 to 2.04]), low-risk PESI scores (adjusted risk ratio, 0.96 [CI, 0.43 to 2.13]), and low-risk sPESI scores (adjusted risk ratio, 1.06 [CI, 0.58 to 1.93]), were not associated with ED discharge. When the cohort was restricted to patients who were hemodynamically stable, those in a low-risk PESI class, or those with a low-risk sPESI score, no demographic information or hospital features were associated with ED discharge in the multivariable regression analysis (Appendix Table 2, available at Annals.org).
Medications Prescribed During ED Encounters
Among patients who were hospitalized after their ED encounter and had medication information available, heparin products were the most frequently prescribed anticoagulants, used in 68.9% (n = 703 454) of patients. In admitted patients, 8.26% (n = 84 336) received a factor Xa inhibitor. Among discharged patients with medication information available, the most frequently used anticoagulant class was factor Xa inhibitors (43.2% [n = 102 758]), of which apixaban was the most widely prescribed agent (25.7% [n = 61 225]), followed by rivaroxaban (12.7% [n = 30 217]). Among discharged patients, factor Xa inhibitors were prescribed more than 3 times as frequently at teaching hospitals (adjusted risk ratio, 3.53 [CI, 1.48 to 8.38]; P = 0.006). In addition, discharged patients were more likely to receive a factor Xa inhibitor for anticoagulation if they had private insurance (adjusted risk ratio, 5.03 [CI, 1.50 to 16.9]; P = 0.011) and less likely if they had Medicaid (adjusted risk ratio, 0.08 [CI, 0.011 to 0.061]; P = 0.017).
Discussion
In this nationwide study, we investigated the contemporary temporal trends in discharge rates for acute PE within U.S. EDs. The central finding of our analysis was that the proportion of ED visits resulting in discharge remained relatively unchanged in the United States between 2012 and 2020, including within low-risk groups, where nearly two thirds of these visits still result in hospitalization. Importantly, we did not find any baseline patient or hospital characteristics that were predictive of outpatient management for acute PE; however, use of oral anticoagulation at discharge was more frequent at teaching institutions and among patients with private insurance.
Acute PE is a common diagnosis in the ED setting, and its prevalence continues to increase in the United States (24, 29, 30). Previous studies have established the safety of outpatient management in both randomized and nonrandomized populations, and current recommendations endorse its use in appropriately triaged, low-risk patients (10). In addition, updated guidelines for management of PE, including the ESC and CHEST guidelines, recommend outpatient care if certain criteria are met, such as hemodynamic stability, low risk for death based on stratification scores, and a feasible outpatient plan (10, 11). Moreover, inpatient management for acute PE is expensive, with some estimates from more than a decade ago suggesting a cost of nearly $9000 per hospitalization (12, 13). Prior studies have demonstrated the substantial health care cost savings resulting from utilizing outpatient management for acute PE (31).
Despite the recommended changes in practice, rates of adoption of outpatient management for acute PE have varied (32). For instance, a nationwide analysis covering the period from 2007 to 2012 showed an increase in outpatient management of acute PE, from 3.9% to 6.7% (33). Similarly, in a Canadian cohort of patients, ED discharge significantly increased from 8.79% to 14.3% between 2000 and 2009. However, others have found no change in the adoption of outpatient management (15, 20). In a multicenter study among a cohort of 7 community hospitals, temporal trends did not significantly increase between 2012 and 2018, with nearly 85% of visits resulting in admission (17).
This analysis, which leverages a longitudinal nationwide sample of ED visits, found similar trends in a longer and more recent time frame. Consistent with other studies examining changes in the early 2010s, we found that rates of adoption remained largely stable between 2012 and 2020. This plateau in adoption rates means that a large proportion of patients with PE, including low-risk PE, remain hospitalized for treatment despite clinical evidence and guidelines recommending otherwise. These findings were consistent in 2 of our 3 analyses, including the primary cohort analysis and 1 of the sensitivity analyses. Among patients with a first coded diagnosis of acute PE, we observed a small but statistically significant increase in discharge rates, although overall rates remained low, which is consistent with the primary findings.
In our analysis, we also stratified patients by using previously established clinical prediction tools to identify those at low risk. Interestingly, among these low-risk patients, we found that nearly two thirds were still managed with inpatient hospitalization. In fact, when assessed temporally, rates of discharge among low-risk patients remained constant over the 9 years of the analysis. The reason outpatient management has plateaued remains unclear. Comorbidities did not significantly differ among those who were admitted or discharged with acute PE. Furthermore, additional clinical factors, such as the PE risk scores, were not predictive of outpatient care. However, we did find that patients treated at teaching hospitals and those with private insurance had higher rates of oral anticoagulant prescription at discharge, which may partially explain the observed trends. It is important to note that these findings must be interpreted with caution because of the possible influence of multiple testing. Nevertheless, because accessibility to oral anticoagulation may have influenced the patterns we observed in this analysis, we speculate that increasing availability of these agents may support implementation of ED discharge for low-risk patients with PE. Finally, given that the decision to hospitalize patients is multifactorial, several other factors are also likely involved, including provider discomfort with outpatient management, poor clinical systems to allow for short-term follow-up and management of these patients, and medicolegal concerns (34, 35). Further research to more precisely investigate these reasons is warranted.
Our study has several important limitations. First, our analysis leveraged a data set that relied on administrative diagnostic codes for patient identification. Like other nationwide studies, this analysis may be susceptible to misclassification because the reported diagnoses could not be sufficiently adjudicated. Nevertheless, the billing codes utilized for this study have been extensively described in the literature, and 2 sensitivity analyses were performed to further validate our findings (24–26). Second, due to the cross-sectional nature of the data set, our analysis was only able to capture ED visits at 1 instance in time, limiting our ability to examine other important outcomes of interest, including recurrent venous thromboembolism, bleeding, and death. Third, although a strength of the NHAMCS data set was its ability to provide more granular patient-specific information compared with other claims-based data sets, it was not able to fully capture a patient’s complete medical history, and it was impossible to know whether a single patient had multiple ED visits. Thus, it remains possible that the estimated risk stratification scores reported in this analysis may be underestimates of each patient’s true risk. Moreover, right ventricular assessment, burden of thrombotic disease, and other high-risk features, such as pulmonary infarction, were not available in this data set. These additional risk factors have been shown to influence admission decisions for low-risk patients with PE (36). Similarly, determination of hemodynamic stability was based on only 1 set of vital signs, which may not accurately reflect each patient’s true hemodynamic profile. In addition, admission may be driven by non-PE reasons, such as management of comorbidities, which may not have been captured in administrative data. Finally, this analysis included data up to 2020, which coincided with the beginning of the COVID-19 pandemic. Whether the onset of the pandemic influenced the findings in this later period is unclear.
In conclusion, we examined real-world national trends in discharge rates for acute PE in U.S. EDs between 2012 and 2020. Our findings show that outpatient management seems to have remained relatively unchanged during that period and that a considerable proportion of low-risk patients are still managed with inpatient hospitalization. Although no characteristics were predictive of outpatient management for acute PE, we found that use of oral anticoagulation at discharge was more frequent at teaching institutions and among patients with private insurance, which may partially explain these trends. Altogether, these findings suggest that outpatient management of acute PE remains underutilized despite clinical evidence and guideline recommendations. Further investigation of the root causes of ED triage decisions and dedicated interventions to improve appropriate use of outpatient management are warranted.
Grant Support:
Dr. Secemsky is funded in part by grant K23HL150290 from the National Heart, Lung, and Blood Institute.
Appendix
Appendix Figure 1. National rates of ED discharge for patients with acute PE who received a CT scan in the ED.
Error bars indicate 95% CIs. CT = computed tomography; ED = emergency department; PE = pulmonary embolism.
Appendix Figure 2. National rates of ED discharge for patients with a first coded diagnosis of acute PE.
Error bars indicate 95% CIs. ED = emergency department; PE = pulmonary embolism.
Appendix Table 1.
Claims and Survey Codes Used to Identify Variables for Analysis
| Variable | Claims/Survey Codes Used* |
|---|---|
| Diagnosis | |
| Acute pulmonary embolism | ICD-9 codes 415.1, 415.11, 415.19, 415.13 |
| ICD-10 codes I26-, I260, I269 | |
| Clinical findings | |
| Chest pain | 10500, 10501, 10502, 10503, 12650 |
| Dyspnea | 14150, 14200 |
| Hemoptysis | 14701 |
| Altered mental status | 58420, 58400 |
| Syncope | 10300 |
| Teaching hospital | RESINT (patient seen by resident or intern physician) |
| Risk stratification | |
| PESI | Age, sex, CANCER, CHF, COPD, PULSE, BPSYS, RESPR, TEMPF, altered mental status, POPCT |
| sPESI | Age, CANCER, chronic cardiopulmonary disease (CHF or COPD), PULSE, BPSYS, POPCT |
| Medications | |
| Heparin products | Medication category: 261 |
| Unfractionated heparin | 14240, 12216, 03318 |
| Enoxaparin (Lovenox [Sanofi-Aventis]) | 00198, 94117 |
| Factor Xa inhibitors | Medication category: 285 |
| Apixaban (Eliquis [Bristol-Myers Squibb]) | 14002, 13009 |
| Rivaroxaban (Xarelto [Janssen Pharmaceuticals]) | 09958, 11393 |
| Warfarin products | Medication category: 262 |
BPSYS = initial vital signs: blood pressure – systolic; CHF = congestive heart failure; COPD = chronic obstructive pulmonary disease; ICD = International Classification of Diseases; NHAMCS = National Hospital Ambulatory Medical Care Survey; PESI = Pulmonary Embolism Severity Index; POPCT = initial vital signs: pulse oximetry (percent); PULSE = initial vital signs: heart rate per minute; RESPR = initial vital signs: respiratory rate per minute; sPESI = simplified Pulmonary Embolism Severity Index; TEMPF = initial vital signs: temperature (Fahrenheit).
Except where indicated, NHAMCS codes are shown.
Appendix Table 2.
Adjusted Risk Ratios for Predictors of Discharge Among Hemodynamically Stable Patients, Patients With PESI Class I or II, and Patients With sPESI Score of 0 in the ED
| Variable | Adjusted Risk Ratio (95% CI)* |
||
|---|---|---|---|
| Hemodynamically Stable (n = 919 292) | PESI Class I or II (n = 910 898) | sPESI Score of 0 (n = 733 776) | |
| Year | 0.95 (0.81–1.13) | 1.00 (0.84–1.18) | 1.04 (0.85–1.27) |
| Demographic characteristics | |||
| Age | 0.99 (0.98–1.01) | 0.99 (0.97–1.01) | 1.00 (0.98–1.03) |
| Male sex | 0.58 (0.30–1.11) | 0.92 (0.46–1.83) | 0.69 (0.32–1.47) |
| Race | |||
| White | 1.21 (0.49–3.01) | 0.94 (0.42–2.12) | 0.77 (0.32–1.86) |
| Black | 0.82 (0.30–2.22) | 1.05 (0.43–2.58) | 1.27 (0.48–3.34) |
| Other | 0.85 (0.13–5.26) | 1.02 (0.15–6.77) | 1.19 (0.22–6.21) |
| Hispanic or Latino ethnicity | 0.87 (0.30–2.55) | 1.02 (0.27–3.84) | 2.48 (0.79–7.72) |
| Comorbidities | |||
| Cancer | 1.29 (0.53–3.17) | 2.73 (1.63–4.56) | — |
| Dementia | — | — | — |
| Cerebrovascular disease | — | — | — |
| COPD | 1.18 (0.58–2.40) | 2.24 (0.95–5.32) | — |
| Congestive heart failure | 1.06 (0.48–2.32) | 0.25 (0.02–2.38) | — |
| Diabetes mellitus | 2.54 (1.11–5.83) | 1.47 (0.56–3.84) | 1.66 (0.73–3.75) |
| History of VTE | 1.29 (0.58–2.86) | 1.45 (0.68–3.09) | 1.61 (0.69–3.77) |
| History of HIV | — | — | — |
| End-stage renal disease | — | — | — |
| Chronic kidney disease | 0.73 (0.15–3.42) | 1.08 (0.21–5.55) | 1.65 (0.54–5.04) |
| Coronary artery disease | 0.68 (0.27–1.74) | 1.78 (0.60–5.27) | 3.25 (0.85–12.4) |
| Alcohol use disorder | — | 0.09 (0.00–1.47) | — |
| Asthma | 0.48 (0.23–1.00) | 1.02 (0.44–2.37) | 1.09 (0.44–2.66) |
| Osteoporosis | 1.13 (0.22–5.67) | 1.58 (0.28–9.00) | 1.24 (0.23–6.54) |
| Depression | 1.17 (0.43–3.18) | 1.79 (0.70–4.57) | 1.56 (0.40–5.99) |
| Hyperlipidemia | 0.96 (0.37–2.47) | 1.01 (0.30–3.38) | 1.05 (0.36–3.07) |
| Hypertension | 1.53 (0.78–2.99) | 1.13 (0.55–2.28) | 1.36 (0.65–2.82) |
| Obesity | 0.82 (0.27–2.46) | 0.64 (0.21–1.97) | 0.77 (0.28–2.16) |
| Obstructive sleep apnea | — | — | — |
| Geographic region | |||
| Northeast | 3.85 (0.85–17.4) | 1.01 (0.22–4.67) | 1.24 (0.21–7.24) |
| Midwest | 0.54 (0.20–1.45) | 1.35 (0.53–3.39) | 1.00 (0.31–3.27) |
| South | 0.68 (0.29–1.57) | 0.63 (0.26–1.54) | 0.80 (0.31–2.08) |
| West | 2.94 (0.76–11.4) | 1.72 (0.46–6.30) | 1.41 (0.33–6.01) |
| Teaching hospital | 1.45 (0.60–3.54) | 1.20 (0.50–2.86) | 0.57 (0.13–2.45) |
| Primary payment | |||
| Private insurance | 0.96 (0.54–1.68) | 0.85 (0.46–1.57) | 0.74 (0.35–1.52) |
| Medicare | 1.12 (0.48–2.60) | 1.57 (0.72–3.42) | 1.06 (0.43–2.60) |
| Medicaid/other state program | 0.67 (0.29–1.53) | 0.83 (0.38–1.80) | 1.26 (0.53–2.98) |
| Other | 1.13 (0.41–3.14) | 0.72 (0.23–2.28) | 0.57 (0.19–1.64) |
| Rural location | 0.76 (0.14–4.06) | 0.98 (0.20–4.78) | 1.17 (0.17–8.01) |
| Hemodynamically stable† | — | 0.99 (0.47–2.09) | 0.80 (0.32–1.99) |
| PESI class I or II | 0.85 (0.31–2.31) | — | 1.20 (0.15–9.15) |
| sPESI score of 0 | 0.83 (0.41–1.71) | 0.77 (0.36–1.66) | — |
COPD = chronic obstructive pulmonary disease; ED = emergency department; PESI = Pulmonary Embolism Severity Index; sPESI = simplified Pulmonary Embolism Severity Index; VTE = venous thromboembolism.
All multivariable models were adjusted for age, sex, race, ethnicity, and geographic region.
Defined as systolic blood pressure >100 mm Hg, heart rate <110 beats/min, and oxygen saturation >90%.
Footnotes
Disclosures:
Disclosures can be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum=M23-2442.
Reproducible Research Statement:
Study protocol: Not available. Statistical code: Detailed R code is available from Mr. Watson (e-mail, nwatson@bidmc.harvard.edu). Data set: Data from the NHAMCS are publicly available on the NCHS website (www.cdc.gov/nchs/ahcd).
Contributor Information
Nathan W. Watson, Harvard Medical School, and Smith Center for Outcomes Research in Cardiology, Department of Medicine, Beth Israel Deaconess Medical Center, Boston, Massachusetts.
Brett J. Carroll, Harvard Medical School; Smith Center for Outcomes Research in Cardiology, Department of Medicine, Beth Israel Deaconess Medical Center; and Division of Cardiology, Department of Medicine, Beth Israel Deaconess Medical Center, Boston, Massachusetts.
Anna Krawisz, Harvard Medical School; Smith Center for Outcomes Research in Cardiology, Department of Medicine, Beth Israel Deaconess Medical Center; and Division of Cardiology, Department of Medicine, Beth Israel Deaconess Medical Center, Boston, Massachusetts.
Alec Schmaier, Harvard Medical School, and Division of Cardiology, Department of Medicine, Beth Israel Deaconess Medical Center, Boston, Massachusetts.
Eric A. Secemsky, Harvard Medical School; Smith Center for Outcomes Research in Cardiology, Department of Medicine, Beth Israel Deaconess Medical Center; and Division of Cardiology, Department of Medicine, Beth Israel Deaconess Medical Center, Boston, Massachusetts.
References
- 1.Silver MJ, Giri J, Duffy Á, et al. Incidence of mortality and complications in high-risk pulmonary embolism: a systematic review and meta-analysis. J Soc Cardiovasc Angiogr Interv 2023;2:100548. doi: 10.1016/j.jscai.2022.100548 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Wendelboe AM, Raskob GE. Global burden of thrombosis: epidemiologic aspects. Circ Res 2016;118:1340–1347. doi: 10.1161/CIRCRESAHA.115.306841 [DOI] [PubMed] [Google Scholar]
- 3.Duffett L, Castellucci LA, Forgie MA. Pulmonary embolism: update on management and controversies. BMJ 2020;370:m2177. doi: 10.1136/bmj.m2177 [DOI] [PubMed] [Google Scholar]
- 4.Prabhu W, Soukas PA. Pulmonary embolism in 2017: increasing options for increasing incidence. R I Med J (2013). 2017;100:27–32. [PubMed] [Google Scholar]
- 5.Beer JH, Burger M, Gretener S, et al. Outpatient treatment of pulmonary embolism is feasible and safe in a substantial proportion of patients. J Thromb Haemost 2003;1:186–187. doi: 10.1046/j.1538-7836.2003.00005.x [DOI] [PubMed] [Google Scholar]
- 6.Erkens PMG, Gandara E, Wells P, et al. Safety of outpatient treatment in acute pulmonary embolism. J Thromb Haemost 2010;8:2412–2417. doi: 10.1111/j.1538-7836.2010.04041.x [DOI] [PubMed] [Google Scholar]
- 7.Piran S, Le Gal G, Wells PS, et al. Outpatient treatment of symptomatic pulmonary embolism: a systematic review and meta-analysis. Thromb Res 2013;132:515–519. doi: 10.1016/j.thromres.2013.08.012 [DOI] [PubMed] [Google Scholar]
- 8.Bledsoe JR, Woller SC, Stevens SM, et al. Management of low-risk pulmonary embolism patients without hospitalization: the Low-Risk Pulmonary Embolism Prospective Management Study. Chest. 2018; 154:249–256. doi: 10.1016/j.chest.2018.01.035 [DOI] [PubMed] [Google Scholar]
- 9.Aujesky D, Roy PM, Verschuren F, et al. Outpatient versus inpatient treatment for patients with acute pulmonary embolism: an international, open-label, randomised, non-inferiority trial. Lancet. 2011;378:41–48. doi: 10.1016/S0140-6736(11)60824-6 [DOI] [PubMed] [Google Scholar]
- 10.Konstantinides SV, Meyer G, Bueno H, et al. ; ESC Scientific Document Group. 2019 ESC guidelines for the diagnosis and management of acute pulmonary embolism developed in collaboration with the European Respiratory Society (ERS). Eur Heart J 2020;41:543–603. doi: 10.1093/eurheartj/ehz405 [DOI] [PubMed] [Google Scholar]
- 11.Kearon C, Akl EA, Ornelas J, et al. Antithrombotic therapy for VTE disease: CHEST guideline and expert panel report. Chest. 2016;149: 315–352. doi: 10.1016/j.chest.2015.11.026 [DOI] [PubMed] [Google Scholar]
- 12.Fanikos J, Rao A, Seger AC, et al. Hospital costs of acute pulmonary embolism. Am J Med 2013;126:127–132. doi: 10.1016/j.amjmed.2012.07.025 [DOI] [PubMed] [Google Scholar]
- 13.LaMori JC, Shoheiber O, Mody SH, et al. Inpatient resource use and cost burden of deep vein thrombosis and pulmonary embolism in the United States. Clin Ther 2015;37:62–70. doi: 10.1016/j.clinthera.2014.10.024 [DOI] [PubMed] [Google Scholar]
- 14.Lutsey PL, Walker RF, Maclehose RF, et al. Inpatient versus outpatient acute venous thromboembolism management: trends and postacute healthcare utilization from 2011 to 2018. J Am Heart Assoc 2021;10:e020428. doi: 10.1161/JAHA.120.020428 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Mansour S, Alotaibi G, Wu C, et al. Trends in admission rates and in-hospital stay for venous thromboembolism. Thromb Res 2017;156:149–154. doi: 10.1016/j.thromres.2017.06.012 [DOI] [PubMed] [Google Scholar]
- 16.Klil-Drori AJ, Coulombe J, Suissa S, et al. Temporal trends in outpatient management of incident pulmonary embolism and associated mortality. Thromb Res 2018;161:111–116. doi: 10.1016/j.thromres.2017.10.026 [DOI] [PubMed] [Google Scholar]
- 17.Khatib R, Nitti K, McDowell M, et al. Outpatient management of patients presenting with venous thromboembolism: retrospective cohort study at 11 community hospitals. J Thromb Thrombolysis. 2021;52:179–188. doi: 10.1007/s11239-020-02328-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Dieleman JL, Cao J, Chapin A, et al. US health care spending by payer and health condition, 1996–2016. JAMA. 2020;323:863–884. doi: 10.1001/jama.2020.0734 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Richards JR, Derlet RW. Overcrowding in the nation’s emergency departments: complex causes and disturbing effects. Ann Emerg Med 2000;35:63–68. doi: 10.1016/s0196-0644(00)70105-3 [DOI] [PubMed] [Google Scholar]
- 20.Singer AJ, Thode HC, Peacock WF. Admission rates for emergency department patients with venous thromboembolism and estimation of the proportion of low risk pulmonary embolism patients: a US perspective. Clin Exp Emerg Med 2016;3:126–131. doi: 10.15441/ceem.15.096 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Hepburn-Brown M, Irving L, Hammerschlag G. Early decision-making in acute pulmonary embolism: a retrospective clinical audit. Intern Med J 2019;49:481–489. doi: 10.1111/imj.14042 [DOI] [PubMed] [Google Scholar]
- 22.Pendyal A, Rothenberg C, Scofi JE, et al. National trends in emergency department care processes for acute myocardial infarction in the United States, 2005 to 2015. J Am Heart Assoc 2020;9: e017208. doi: 10.1161/JAHA.120.017208 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Berger JS, Ladapo JA. Underuse of prevention and lifestyle counseling in patients with peripheral artery disease. J Am Coll Cardiol 2017;69:2293–2300. doi: 10.1016/j.jacc.2017.02.064 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Smith SB, Geske JB, Kathuria P, et al. Analysis of national trends in admissions for pulmonary embolism. Chest. 2016;150:35–45. doi: 10.1016/j.chest.2016.02.638 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Stein PD, Matta F, Hughes PG, et al. Nineteen-year trends in mortality of patients hospitalized in the United States with high-risk pulmonary embolism. Am J Med 2021;134:1260–1264. doi: 10.1016/j.amjmed.2021.01.026 [DOI] [PubMed] [Google Scholar]
- 26.Raghupathy S, Barigidad AP, Doorgen R, et al. Prevalence, trends, and outcomes of pulmonary embolism treated with mechanical and surgical thrombectomy from a nationwide inpatient sample. Clin Pract 2022;12:204–214. doi: 10.3390/clinpract12020024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Aujesky D, Obrosky DS, Stone RA, et al. Derivation and validation of a prognostic model for pulmonary embolism. Am J Respir Crit Care Med 2005;172:1041–1046. doi: 10.1164/rccm.200506-862OC [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Jiménez D, Aujesky D, Moores L, et al. ; RIETE Investigators. Simplification of the Pulmonary Embolism Severity Index for prognostication in patients with acute symptomatic pulmonary embolism. Arch Intern Med 2010;170:1383–1389. doi: 10.1001/archinternmed.2010.199 [DOI] [PubMed] [Google Scholar]
- 29.Schissler AJ, Rozenshtein A, Schluger NW, et al. National trends in emergency room diagnosis of pulmonary embolism, 2001–2010: a cross-sectional study. Respir Res 2015;16:44–47. doi: 10.1186/s12931-015-0203-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Minges KE, Bikdeli B, Wang Y, et al. National trends in pulmonary embolism hospitalization rates and outcomes for adults aged ≥65 years in the United States (1999 to 2010). Am J Cardiol 2015;116:1436–1442. doi: 10.1016/j.amjcard.2015.07.068 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Bledsoe JR, Woller SC, Stevens SM, et al. Cost-effectiveness of managing low-risk pulmonary embolism patients without hospitalization. The Low-Risk Pulmonary Embolism Prospective Management Study. Am J Emerg Med 2021;41:80–83. doi: 10.1016/j.ajem.2020.12.042 [DOI] [PubMed] [Google Scholar]
- 32.Pokharel Y, Tang F, Jones PG, et al. Adoption of the 2013 American College of Cardiology/American Heart Association cholesterol management guideline in cardiology practices nationwide. JAMA Cardiol 2017;2:361–369. doi: 10.1001/jamacardio.2016.5922 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Stein PD, Matta F, Hughes MJ. National trends in home treatment of acute pulmonary embolism. Clin Appl Thromb Hemost 2018;24: 115–121. doi: 10.1177/1076029616674827 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Roy PM, Moumneh T, Penaloza A, et al. Outpatient management of pulmonary embolism. Thromb Res 2017;155:92–100. doi: 10.1016/j.thromres.2017.05.001 [DOI] [PubMed] [Google Scholar]
- 35.Kline JA, Kahler ZP, Beam DM. Outpatient treatment of low-risk venous thromboembolism with monotherapy oral anticoagulation: patient quality of life outcomes and clinician acceptance. Patient Prefer Adherence. 2016;10:561–569. doi: 10.2147/PPA.S104446 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.O’Hare C, Grace KA, Schaeffer WJ, et al. Adverse clinical outcomes among patients with acute low-risk pulmonary embolism and concerning computed tomography imaging findings. JAMA Netw Open. 2023;6:E2311455. doi: 10.1001/jamanetworkopen.2023.11455 [DOI] [PMC free article] [PubMed] [Google Scholar]