Stranded in the emergency department: an analysis of boarding trends in older adults in the United States

Abstract

Introduction

The rapidly ageing population and multimorbidity are associated with increased emergency department (ED) visits by older adults. In the ED, older adults have higher risk of hospitalization, functional and cognitive decline, and mortality. Boarding, holding admitted patients in the ED awaiting a hospital bed, exacerbates these negative outcomes, which disproportionately affect older adults.

Methods

We conducted a cross-sectional analysis to investigate US boarding trends by age using clinical administrative data from 5 health systems and publicly available NHAMCS data from 2018 to 2024.

Results

Boarding ≥3 h in the ED has increased across diverse hospital types, with oldest adults (85+) facing the greatest risk (System 4: IRR [95% CI] = 1.18 [1.15-1.20], System 5: IRR [95% CI] = 1.20 [1.17-1.23], System 3 [Community Hospital]: IRR [95% CI] = 1.25 [1.19-1.33]). These results were recapitulated at the national level in NHAMCS (IRR [95% CI] = 1.30 [1.05-1.61]).

Discussion

The trend of increased boarding has serious implications for patients, caregivers, and health systems. The 2025 CMS Age-Friendly Hospital Measure offers opportunities to improve processes and procedures to mitigate the negative effects of hospital boarding on older patients. We highlight opportunities to address this challenge, including ongoing quality improvement initiatives, bed prioritization algorithms, and alternate admission pathways.

Introduction

Our ageing population, coupled with greater prevalence of multimorbidity and social risk factors, are associated with a concomitant increase in emergency department (ED) visits by older adults (65+).^1,2 In the ED, older adults are a vulnerable population with higher risk for hospitalizations, functional and cognitive decline, and, ultimately, mortality from prolonged ED visits and inpatient stays.^3,4 These complications are compounded by being a “boarded patient,” defined by the American College of Emergency Physicians as a patient who remains in the ED after the patient has been admitted or placed into observation status at the facility but has not been transferred to an inpatient bed or observation unit.⁵

Recent studies have demonstrated a consistent association between boarding and negative health outcomes in older adults. Specifically, for older patients, boarding has been linked to a greater risk of delirium,^6,7 prolonged inpatient length of stay (LOS),⁸ functional decline,⁹ and mortality during the hospitalization.^10-12 Factors likely associated with these risks include prolonged exposure to the busy and often noisy, overstimulating environment of the ED, lack of specialized training in geriatric-focused care, and insufficient staff attention amidst crowded EDs which can exacerbate cognitive dysfunction.¹³ Critically, delirium is an important predictor of mortality.^14-16 Because of the association with negative patient outcomes, avoiding prolonged boarding is now included as a measure of quality by the Centers for Medicare & Medicaid Services’ (CMS) Age Friendly Hospital Measure (AFHM) that commenced mandatory reporting in 2025.¹⁷ This measure asks hospitals to attest to having protocols in place to reduce either ED LOS to 8 hours or boarding to 3 hours (h) for older adults. While submitting quantitative data for either metric is not currently required, it is anticipated this may transition to a pay-for-reporting program for acute care hospitals.¹⁸ In light of this measure, more robust study of boarding trends and metrics at the hospital level is important.

Despite research showing there are deleterious health effects associated with boarding and initiatives put forward by federal institutions, differences in boarding rates by age in the United States (US) are unclear. Recent studies have shown that boarding has increased nationwide in the last decade in both frequency and duration,¹⁹ and that boarding has increased for patients 65 years and older in line with the overall national trend.²⁰ However, there is limited research describing these trends for older patients as compared to a younger adult population (21-64) at the hospital system-level or within a nationally representative sample of ED visits. Both levels of reporting are important—national trends describe the overall extent of the issue, but patients experience boarding in their local hospital. Moreover, previous studies have primarily relied on a 4-h cutoff or more to define boarding,^19,21 while the CMS AFHM caps boarding for older adults at 3 h. This discrepancy highlights a need for additional research, which describes boarding in agreement with the goals of the CMS AFHM.

Given the current gaps in knowledge and the importance of addressing hospital boarding, this study aims to report age-stratified boarding trends in the US at multiple levels, focusing on older adults, who are at greater risk of harms from boarding. We provide a descriptive baseline of comparative boarding rates using a 3-h cutoff, in line with the new CMS AFHM attestation. We leverage a consortium of health systems participating in the Geriatric Emergency care Applied Research (GEAR) network (R33 AG058926) and data from the National Hospital Ambulatory Medical Care Survey (NHAMCS).²² We hypothesize that older adults (65+) are at greater risk of hospital boarding than younger adults (21-64) given their higher proportionate ED utilization²³ and more complex medical needs due to increased risk of multimorbidity.

Data and methods

Study design, setting, and population

We conducted a repeated cross-sectional analysis, investigating boarding trends in two ways: (1) using aggregate electronic health record (EHR) data from multiple health systems in varied settings (system-level), and (2) using nationally representative survey data of ED visits from the National Center for Health Statistics NHAMCS study (national-level).²²

System-level

System-level data were collected from hospitals participating in the National Institute on Aging GEAR Network.^24,25 This portion of the study leveraged EHR data from 32 EDs across 5 health systems (Systems 1-5) located in the Northeastern, Southern, Southwest, and Midwestern regions of the US. Systems 1-3 contributed data from both Academic Medical Centers (AMCs) and Community Hospitals (CHs), while System 4 and 5 only provided data from their AMCs. AMCs were defined as teaching hospitals, while CHs were all other non-teaching hospitals. Seven out of 32 (21.9%) EDs were AMCs, none were free-standing, 13 were rural (Rural-Urban Commuting Area code of 4 or greater) (40.6%), and average annual ED visit volume ranged from 3400 to 95,000.

The study sample included all adults aged 21 and over who presented to a contributing ED and had a final ED disposition of Admitted (inpatient hospitalization). We excluded patients missing boarding times from the study sample. This was <5% of the total admissions at all systems except one, where 16.1% of patients were missing a boarding time. At this system, all ED observation patients were given an admitted disposition status. Therefore, missingness was driven by ED observation encounters discharged from the ED. Since the encounters were not admitted, these did not meet study inclusion criteria and were removed from the total count of admissions.

Four out of 5 hospital systems provided data from 2018 to 2024, and one system (System 3), provided data from 2019 to 2024. System 3 excluded patients from the population who had not given permission for their data to be used in research (n = 18,577, 6.4% of total admissions).

National-level

To generate nationally representative estimates of boarding, we analyzed ED visit files from publicly available NHAMCS surveys from 2018 through 2022, which was the end of the study.²²

Hospital system data extraction (GEAR network)

Hospital data were extracted to create summary system-level datasets which included total counts for boarding and admission within covariate groupings (eg, stratified by age group, year, sex, race and ethnicity, and acuity.²⁶ These data were shared with a central GEAR researcher who conducted all system-level analysis.

National hospital ambulatory medical care survey (NHAMCS)

NHAMCS is a national survey of about 500 hospitals reporting encounter-level information extracted from medical record review of ED visits.²² Sample design variables included in the public use dataset were used to produce nationally representative estimates of ED visit characteristics using methods recommended by the National Center for Health Statistics²⁷ and implemented with the svy commands in Stata/SE version 17.0.²⁸

Measures

Boarding

In system-level analyses, boarding time was calculated as the difference in time between bed request and ED departure at all hospital systems except Critical Access Hospitals (CAHs) at System 3. For system 3 CAHs, boarding was calculated as the difference in time between disposition determination and ED departure. In national-level analyses with NHAMCS data, boarding was assessed using the “boarded” variable (difference in time between hospital admission order and ED departure in minutes), which is provided for all patients admitted to the same hospital. For analysis, we dichotomized boarding at boarding time >3 h for admitted patients, following the recommended CMS Age-friendly Health System (AFHS) hospital measure.¹⁷

Age categories

To evaluate boarding proportions for older vs younger patients in both system-level and NHAMCS national-level analyses, older adults were split into 3 groups: 65-74, 75-84 and 85+ years of age, and were compared with younger adults 21-64 years of age (reference group) at the time of the ED hospital admission.

Statistical analysis

To investigate whether older adults (by older age group) board at higher rates than younger adults (21-64), we calculated the proportion of admitted patients that boarded 3+ h by age category annually. These proportions are reported at the system-level in Figures 1 and 2.²⁹

Figure 1.

Proportion of total admits boarding 3+ h at AMCs. Proportions reflect the total number of admitted patients who boarded 3+ h over the total number of admitted patients.

Figure 2.

Proportion of total admits boarding 3+ h at CHs and NHAMCS. Proportions reflect the total number of admitted patients who boarded 3+ h over the total number of admitted patients.

Additionally, we utilized multivariate Poisson regression to obtain incident rate ratios for boarding 3+ h, controlling for covariates of year, sex, race and ethnicity, and acuity. Missing values for all variables were replaced with “Other or Unknown.” We also performed a sensitivity analysis of the Poisson regression using a 4-h cutoff to define boarding as commonly utilized in previous boarding studies.^19,21

Given the heterogeneity across hospitals systems in clinical care, resources, staffing models, patient populations, and other factors, we built individual regression models within each hospital system (Systems 1-5) and hospital type (AMC or CH), resulting in 8 health system-hospital type models. For example, at System 1, we developed 2 models: 1 for the single AMC and another for the 6 CHs. For hospital system-hospital type data containing multiple EDs, we included a fixed effect for ED.

All hospital system-level analyses were conducted using R 4.4.0³⁰ and NHAMCS analysis was conducted using Stata/SE version 17.0.²⁸ These analyses are exempt from institutional review board review as the use of aggregate count data for the system-level analyses and public-use data files from NHAMCS do not involve human subjects as defined by federal regulations.

Results

Sample characteristics

Across the 5 health systems, there were a total of 1 168 682 admissions from the ED and 562 946 cases of boarding 3+ hours (48.2%). The sample was predominantly 21-64 years in age (42%-66%) and White (51%-86%). In the NHAMCS sample, 50% of patients were 21%-64% and 67% were White (Table 1).

Table 1.

Sample characteristics.

Characteristic	System 1 (a)	System 2 (b)	System 3 (c)	System 4	System 5	NHAMCS (d)
Annual ED Visit Volume (e)	18,000-95,000	31,000-69,000	3400-78,000	92 000	93 000	141 000 000
Total Admits	521 567	210 910	269 097	167 108	125 909	77 336 842
Number of EDs included	7	2	21	1	1	−-
Age
21-64	243 365 (47%)	139 610 (66%)	113 207 (42%)	93 691 (56%)	61 928 (49%)	39 228 724 (50%)
65-74	101 464 (19%)	37 033 (18%)	61 700 (23%)	34 107 (20%)	25 746 (20%)	16 201 238 (20%)
75-84	98 327 (19%)	23 141 (11%)	57 008 (21%)	25 212 (15%)	22 585 (18%)	13 035 190 (16%)
85+	78 411 (15%)	11 126 (5.3%)	37 182 (14%)	14 098 (8.4%)	15 650 (12%)	8 871 690 (11%)
Year
2018	82 652 (16%)	32 752 (16%)	−-	23 799 (14%)	12 585 (10.0%)	13 786 021 (17%)
2019	82 170 (16%)	32 915 (16%)	43 728 (16%)	24 256 (15%)	14 503 (12%)	14 940 690 (19%)
2020	72 998 (14%)	30 010 (14%)	40 044 (15%)	22 868 (14%)	16 610 (13%)	17 202 160 (22%)
2021	65 092 (12%)	30 684 (15%)	43 271 (16%)	22 958 (14%)	16 962 (13%)	15 570 244 (20%)
2022	62 648 (12%)	28 917 (14%)	42 267 (16%)	22 506 (13%)	19 855 (16%)	15 837 727 (20%)
2023	67 635 (13%)	27 934 (13%)	46 849 (17%)	25 112 (15%)	21 656 (17%)	−-
2024	88 372 (17%)	27 698 (13%)	52 938 (20%)	25 609 (15%)	23 738 (19%)	−-
Sex
Female	267 274 (51%)	100 227 (48%)	127 621 (47%)	87 727 (52%)	62 338 (50%)	39 922 258 (51%)
Male	254 288 (49%)	110 671 (52%)	141 466 (53%)	79 367 (47%)	63 526 (50%)	37 414 584 (48%)
Other or Unknown	5 (<0.1%)	12 (<0.1%)	10 (<0.1%)	14 (<0.1%)	45 (<0.1%)	−-
Race and Ethnicity
White	334 507 (64%)	107 677 (51%)	232 580 (86%)	82 077 (49%)	72 982 (58%)	52 248 946 (67%)
Black or African American	94 236 (18%)	91 635 (43%)	12 803 (4.8%)	57 286 (34%)	18 716 (15%)	13 071 017 (16%)
Hispanic or Latino	72 417 (14%)	4402 (2.1%)	12 272 (4.6%)	21 241 (13%) (f)	7314 (5.8%)	8 174 499 (10%)
Asian	7158 (1.4%)	4054 (1.9%)	5604 (2.1%)	6268 (3.8%)	9944 (7.9%)	−-
American Indian or Alaska Native	1172 (0.2%)	259 (0.1%)	1785 (0.7%)	525 (0.3%)	448 (0.4%)	−-
Native Hawaiian or Pacific Islander	224 (<0.1%)	13 (<0.1%)	323 (0.1%)	678 (0.4%)	238 (0.2%)	−-
Other or Unknown	11 853 (2.3%)	2870 (1.4%)	3730 (1.4%)	20 274 (12%)	16 267 (13%)	3 842 380 (5.0%)
Acuity
1	14 941 (2.9%)	6740 (3.2%)	8067 (3.0%)	9529 (5.7%)	469 (0.4%)	2 556 869 (3.3%)
2	239 100 (46%)	86 579 (41%)	98 648 (37%)	104 161 (62%)	22 249 (18%)	21 014 601 (27%)
3	255 300 (49%)	80 555 (38%)	157 293 (58%)	50 712 (30%)	97 689 (78%)	27 113 131 (35%)
4	10 524 (2.0%)	3794 (1.8%)	4490 (1.7%)	1979 (1.2%)	2231 (1.8%)	1 784 657 (2.3%)
5	447 (<0.1%)	291 (0.1%)	43 (<0.1%)	92 (<0.1%)	3172 (2.5%)	647 659 (0.8%)
Other or Unknown	1255 (0.2%)	32 951 (16%)	556 (0.2%)	635 (0.4%)	99 (<0.1%)	24 219 924 (31%)
Boarded 3+ Hours	246 031 (47%)	114 501 (54%)	42 993 (16%)	102 726 (61%)	56 695 (45%)	12 422 534 (16%)

(a) 1 Academic Medical Center (AMC), 6 CH. (b) 1 AMC, 1 CH. (c) 3 AMCs, 18 CHs across 2 states. (d) NHAMCS values reflect the population level estimates generated from sample weights. (e) Annual ED visits reflect an average over the study period (2018-2024). (f) Ethnicity is a separate variable from race at System 4, total percentages are >100%.

Boarding rates by age groups at AMCs, CHs, and NHAMCS

Across AMCs, boarding trends differed (Figure 1). At Systems 1, 3, and 4, the proportion of total admits who boarded 3 or more hours steeply increased after 2018 and peaked in 2021 and 2022, remaining high into 2024. At Systems 2 and 4, boarding rates reached a minimum in 2019 and 2020 (during the COVID pandemic years), respectively, before increasing again afterwards. At Systems 1, 2, 4, and 5, the maximum boarding rates were approximately 75%, 68%, 81%, and 80%, respectively. At System 3, boarding rates peaked at 23%.

Across all systems, older adult groups experienced increased proportionate boarding at some point during the study period. At Systems 3 and 4, older patients had consistently higher proportions of boarding compared with younger adults. At Systems 1 and 2, older patients had higher boarding rates after their peak years in 2021 and 2022 and at System 5, older patients had higher proportionate boarding earlier in the study period.

At System 1 and 3 CHs and in NHAMCS, the proportion of admitted patients boarding 3+ h increased over the study period (Figure 2). At System 2, boarding rapidly decreased from 2018-2019 before rising in 2021 and peaking in 2023 at approximately 70%. Older adults boarded at rates higher than younger adults at the end of the period at System 3 and at the beginning and end of the period in NHAMCS.

Multivariate poisson regression by age categories

At Systems 4 and 5 AMCs and Systems 1 and 3 CHs, the incidence of boarding was 2%-25% higher for all older age groups as compared to younger adults (21-64). Similarly, at System 1 AMC and in NHAMCS, the oldest patients (85+) had a 4% and 30% increase in the incidence of boarding when compared with younger patients (21-64), respectively. At System 2 CH, the incidence of boarding was 7% lower in the oldest adults compared with the reference population (Table 2). Incident rate ratios for all other covariates are reported in Tables S1 and S2. Sensitivity analyses using a 4-h boarding cutoff found that the incidence of boarding was significantly higher for at least one age group of older adults compared with younger adults at 5 out of 8 health system-hospital types (AMC or CH) (Table S3).

Table 2.

Incident rate ratios of boarding 3+ h by age.

SYSTEM (a)	21-64	65-74	75-84	85+
AMC
System 1	Ref	1.00 (0.98-1.02)	1.01 (1.00-1.03)	1.04 (1.02-1.06)
System 2	Ref	1.02 (1.00-1.04)	1.00 (0.98-1.03)	1.02 (0.99-1.05)
System 3	Ref	1.03 (1.01-1.06)	1.04 (1.01-1.07)	1.02 (0.98-1.06)
System 4	Ref	1.09 (1.07-1.10)	1.12 (1.10-1.14)	1.18 (1.15-1.20)
System 5	Ref	1.09 (1.07-1.12)	1.12 (1.10-1.15)	1.20 (1.17-1.23)
CH
System 1	Ref	1.02 (1.01-1.04)	1.04 (1.02-1.05)	1.05 (1.04-1.07)
System 2	Ref	1.01 (0.97-1.05)	1.00 (0.96-1.05)	0.93 (0.88-0.99)
System 3	Ref	1.16 (1.10-1.22)	1.18 (1.12-1.24)	1.25 (1.19-1.33)
NHAMCS	Ref	1.06 (0.89-1.27)	1.12 (0.94-1.34)	1.30 (1.05-1.61)

Bolded results are statistically significant. (a) AMC is Academic Medical Center; CH is Community Hospital.

Discussion

We analyzed boarding trends over the past 7 years across 5 geographically diverse health systems comprising 7 AMCs and 24 CHs, as well as in a nationally representative dataset (NHAMCS). Boarding trends varied by system, but overall, the proportion of patients who boarded for 3+ h increased after the pandemic for all age groups, in line with a recently published study that leveraged the Epic Cosmos dataset.¹⁹ At 75% (n = 6 of 8) of health system-hospital type groups and in national data, the incidence of boarding 3+ h was 2%-30% higher for at least one group of older patients compared with their younger counterparts. This range shows that although boarding was higher for older adults at most systems, the differences were not equal. The variability in incidence we observed invites further inquiry into the factors that may have precipitated it.

Likely, local factors are contributing to heterogeneity in overall proportionate boarding rates across systems and within the national data set. For instance, a hospital may reserve certain beds for medicine or surgery alone, which allows boarding to occur despite unused capacity and could vary in effect by patient age. Additionally, boarding can result from inefficient admission processes, even in the absence of hospital crowding. There is no standard in when the decision to admit is made (ie, before or after speaking with the admitting clinician, before or after advanced imaging is complete, etc.), which can systemically alter the measured boarding time without impacting the wait time experienced by patients. Additionally, the presence or absence of a centralized bed command center can impact these times, as can local norms around ED-to-inpatient nurse handoff. Identifying which clinical factors are associated with differences in boarding rates across systems, however, is beyond the scope of this study.

Geographic location is another factor which may influence differences in boarding trends overall across EDs, particularly in the context of the COVID-19 pandemic. At several systems, 2020 was an inflection point for boarding. As COVID-19 first spread throughout the US, it impacted different locations at different times. Many EDs experienced significant fluctuations in ED visit volumes and hospital admissions in their respective first months facing the pandemic.³¹ Also, staffing issues, particularly among nurses, vary by region and are known to drive hospital capacity challenges and boarding.³² Bed shortages, hospital closures, and climbing occupancy rates from increasing patient acuity will likely exacerbate this current crisis.³³

The difference across systems in these factors poses a limitation as it suggests that not all boarding measures are the same and thus difficult to compare. To address this limitation, we performed distinct regression models for each health system and hospital type, calculating incidence rate ratios of boarding for older adult age groups as compared to a younger adult reference population. In calculating incidence rate ratios, we generated a normalized estimate for boarding, enabling comparison across systems. Another limitation of this study is that we excluded patients who board in the ED while awaiting transfer to another institution (ie, patients who are transferred to another hospital for admission) as opposed to an inpatient bed at the same hospital. While these patients are generally excluded from the definition of hospital boarding, they nevertheless resemble boarding patients by occupying an ED bed that could otherwise be used to treat additional ED patients. Rural hospitals, which are generally smaller than urban ones (for example, 63% of rural hospitals have fewer than 50 ED beds, vs 25% of urban hospitals),³⁴ are disproportionately affected by such patients waiting to be transferred. Projections in hospital closures and proposed cuts to Medicare and Medicaid reimbursements will likely increase the burden and risk of patients waiting for an inpatient bed.^35,36 It will be critical to also assess and account for such patients when assessing the demand for ED beds and the number of people with long ED stays.

Solutions

Our findings that older adults have increased boarding rates demonstrate the importance of initiatives already in place by CMS to address the negative health consequences of boarding in this population, including elevated risk of hospital mortality¹⁰ and delirium.^6,7,37-39 In August 2024, CMS announced its FY2025 Inpatient Prospective Payment Systems final rule that includes a new Age-Friendly Hospital Measure. Organized around 5 age-friendly domains, this measure assesses hospital commitment to delivering high quality care to patients 65 years and older following the AFHS 4Ms framework—What Matters, Medication, Mentation, and Mobility. One domain specifically highlights the importance of reducing prolonged ED LOS by addressing boarding. Hospitals must attest to having “protocols to reduce the risk of delirium by reducing length of ED stay with a goal of transferring a targeted percentage of older patients out of the ED within 8 h of arrival and/or within 3 h of the decision to admit.”¹⁷ However, this measure is currently limited in its ability to address boarding. Since it can be achieved solely by having a protocol in place to reduce either ED LOS or boarding, boarding is an optional target of intervention. Moreover, it neither specifies a benchmark of what percentage of patients must receive this intervention, nor mandates hospitals to report boarding time data. Shifting hospital attestations to become structural measures, such as those required as part of the CMS hospital inpatient quality reporting program,¹⁸ may financially incentivize acute care hospitals to address boarding, especially when linked to a federal pay-for-reporting program. In addition, the CMS measure developed to directly target boarding, the Emergency Care Access and Treatment, also known as the Equity of Emergency Care Capacity and Quality measure, was recently endorsed by the Partnership for Quality Measurement for use by CMS in national public reporting and quality improvement programs.⁴⁰ This intermediate outcome measure is inclusive of all payers and will therefore have the opportunity to improve care for older adults insured by both traditional Medicare as well as Medicare Advantage. It extends the impact of the structural Age-Friendly Hospital Measure to every eligible ED visit. Future use of both measures in value-based purchasing initiatives, multi-payer alignment efforts, or alternative payment models will be essential to transforming quality measures into solutions rather than descriptions of the problem.

Beyond these policy-level initiatives, there are solutions that health systems and hospitals can implement to address this problem. First, hospitals should take steps to reduce the incidence of hospital boarding among older adults by prioritizing this group in bed placement. Incorporating age as a factor into a risk-based bed management algorithm stands to reduce the disparity in boarding rate between older and younger adults. Akin to surgical smoothing, the practice of rescheduling elective surgeries over a wider time frame to reduce the likelihood that inpatient beds are full at once, risk-based strategies for bed placement prioritize patients at the highest risk of harm from boarding. Given the association of boarding with longer inpatient LOS,^8,41,42 reducing boarding through a modified bed algorithm could in turn decrease the inpatient burden, with further reduced boarding as a potential downstream effect.

Furthermore, health systems should consider addressing boarding by avoiding preventable hospitalizations for older adults. Integrating geriatric-focused care processes that aim to facilitate care transitions, reduce the risk of re-admission, and reduce hospital LOS has been previously demonstrated to reduce health care cost and utilization,^43,44 suggesting a promising role in reducing boarding. Novel programs like hospital at home^45,46 have been proposed as a potential solution for hospital boarding. When paired from the ED, such programs provide patients with alternate options of acute inpatient care and contribute to hospital outflow. Additionally, some studies have found that hospital at home reduces hospitalization cost and healthcare utilization.⁴⁷ Despite its impact on costs and use, other studies have observed that evidence is lacking on whether hospital at home programs will meaningfully impact hospital boarding. They also note that these programs are vulnerable to the same economic factors that drive boarding and that hospitals may be motivated to fill these programs with new surgical patients rather than patients coming from the ED.^46,48 At the same time, both ED volumes and hospital census are expected to continue to grow in the coming years.^49,50 Initiatives like hospital at home and novel observation practices may partially offset that growth, but boarding is still likely to worsen if these programs’ success does not outpace ED volume and hospital census growth.²¹

Conclusion

Admitted patients are remaining stranded for prolonged periods in the ED as boarding continues to worsen. In the present cross-sectional analysis, we found that inpatient boarding in the ED increased across diverse types of hospitals, with older (65-74,75-84) and/or oldest adults (85+) experiencing higher incidences of boarding than younger adults at 6 out of 8 health system-hospital type (AMC or CH) groups and in NHAMCS. The high proportion of older patients boarding in US EDs is troublesome as boarding compounds their risk for delirium and mortality. While these trends appear to vary by region and years, the general trend is an increase in boarding rates which warrants solutions that can be implemented at the levels of hospital, health system, and policy.

Contribution statement

Concept and design: N.S., M.M.J., and U.H. Acquisition, analysis, or interpretation of data: N.S., M.M.J., M.I., R.B.S., D.S.C., and R.M.S. Drafting and revision of the manuscript: All authors. Obtained Funding: U.H. Supervision: U.H. Supplementary material.

Supplementary material

Supplementary material is available at Health Affairs Scholar online.

Funding

This research was funded by the National Institute on Aging of the National Institutes of Health [R33AG058926-08]. Dr. Iscoe's work on this publication was made possible by CTSA Grant Number KL2 TR001862 from the National Center for Advancing Translational Science (NCATS), a component of the National Institutes of Health (NIH). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of NIH.