Length of stay in the emergency department and its influencing factors among the older adult patients: a retrospective cross-sectional study from routine data in Germany

Introduction

The rapid demographic shift in the aging population has raised concerns about older adults’ increased risks of their poor functional conditions (1,2). It has been observed that longevity and old-age dependency ratios are related to health (3-5). Due to rising longevity and consistently low birth rates, German society has been aging quickly (6). Concerns have been raised, particularly in national healthcare policy, over the aging population’s rapid growth and the resulting significant health and long-term care demands from their old age dependency (5,6).

The older adults represent a large proportion of the patients visiting the emergency department (ED), and they are more likely to stay longer (2,7,8). Older adults more often than younger patients present with atypical signs, symptoms, and multi-morbidity; they therefore represent complex multiple chronic conditions and require more extensive investigations (9,10). With older adults contributing disproportionately to ED visits, older adults need a multidimensional approach in ED care services (11,12). Older adults who visited the ED are considered their healthcare related to the high risk of poor health care outcomes, high cost, and high need (2,13). This development calls for ways to better care for this vulnerable group and manage the rising need for geriatric experts and emergency services.

Providing healthcare services in the ED settings has been an effort to measure the quality of care. Principally, the EDs provide care for the patients based on symptoms and severity (14). The standard tool, such as the Manchester Triage System (MTS), a structured and validated treatment prioritization system, is popular for the initial admission of severity among emergency patients (15,16). The routine data from the Electronic Health Records (EHRs) provide healthcare providers and researchers with the relevant information for evaluating the impact of healthcare interventions and quality of care improvement (17).

One of the key benchmarking indicators in improving the quality of care is length of stay in the ED (ED-LoS) (18,19). Prolonged ED-LoS can have negative impact on health outcomes and is associated with increased risks of prolonged hospital stays, morbidity, mortality, medical expenses, and reduced patient and staff satisfaction (20-23). Otto et al. [2022] highlighted the ED-LoS as quality indicators in their earlier research led by a researcher from the university in Saxony-Anhalt, Germany (19). According to their study, the ED-LoS were linked to variables that were relevant to patient, disease, and organization (19). Additionally, according to Wallstab et al. [2022], 16 hospitals in their study reported an average ED-LoS less than 120 minutes, and eight hospitals reported an average ED-LoS more than 240 minutes (24). The findings from the population that comprised all ED visits are reported in these earlier studies. These investigators identified issues related to aging population in Germany, which may have contributed to longer stay in the ED among older adults (19,24).

Evidence-based practice is crucial competency for healthcare providers, resulting in better patient outcomes (25). Based on the literature and the requirement for preliminary findings to guide practical management in the current rapid structural population change, we desired to select patient and disease-related factors for this study. This retrospective cross-sectional study was conducted at the hospital located in the city with the highest aging population aged 65 years and older in Germany (26). This study aims to identify the influencing factors of the ED-LoS from the variables of interest, including age, gender, visiting pathway, mode of transportation, severity level, pain level, hospital admission, and specialty medicine. The findings will be used to enhance the quality of care by modifying routine care services and reducing the ED-LoS for older adults. We present this article in accordance with the STROBE reporting checklist (available at https://jphe.amegroups.com/article/view/10.21037/jphe-24-117/rc).

Methods

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments, and the Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation). The requests for permission to conduct the study in hospital were made by the research team. Patients and the public were not involved in this study. Data were gathered from the routine data of hospital database and processed anonymously in an Excel database before performing the data analysis.

Study design

A retrospective cross-sectional study from routine data was conducted at a regional hospital in Germany, where the number of older adults has significantly increased (27). This hospital has been qualified for the criteria of emergency structures for comprehensive emergency care with the Emergency Level III, the highest level regulated by the Federal Joint Committee (G-BA), Germany (28). The older adults who visited the ED between 2022 and 2023, a time when coronavirus disease 2019 (COVID-19) prevention was being unlocked, were the study’s population. A total of 76,045 patients visited the ED; of those, 42.15% (n=32,053) were older adults.

Data collection

Data was collected from the EHRs throughout a 2-year period, from January 1, 2022, to December 31, 2023. Age, gender, visiting pathway, mode of transportation, severity level, specialty medicine, hospital admission, pain scores, time (in minutes) between ED arrival and initial assessment, time between ED arrival and first visit by physician, and time between ED arrival and ED discharge were extracted from data of older adults who visited the ED. Before performing the data analysis, the information extracted was then anonymously managed in a database. In this study, the variables of interest included age, gender, visiting pathway, mode of transportation, severity level, pain level, hospital admission, and specialty medicine.

Age was defined as the interval of time that passed between the date of birth and the ED visit among the older adults. They were categorized into three groups: youngest-old aged 65–74 years, middle-old aged 75–84 years, and oldest-old aged 85 years and older (29,30). Older adults in this study were classified as either male or female (31). The severity referred to five urgency categories of patients’ clinical priority as determined by the MTS, the triage systems that are used in German practice (32). These systems consist of the urgency categories corresponding to the maximum waiting time for first contact with a physician, including immediate (red), very urgent (orange): evaluation within 10 minutes, urgent (yellow): evaluation within 30 minutes, standard (green): evaluation within 90 minutes, and non-urgent (blue): evaluation within 120 minutes (33). In this study, severity was also categorized into two groups: urgent care need (red, orange, and yellow) and non-urgent care need (green and blue).

Visiting pathway referred to the referral types of the older adults who visited the ED. In this study, the visiting pathway is categorized into two groups: older adults who were referred to the ED by healthcare professionals; the older adults were referred to the ED by statutory sources through emergency physicians and emergency medical services without an emergency physician, and walk-in older adults; older adults and/or family decided to visit the ED by themselves. Mode of transportation referred to the vehicle that transported older adults to ED. Transportation was categorized into ambulance and private vehicles.

Pain level, which was scored on an 11-point numerical rating scale from 0 to 10, was the older adults’ assessment of their level of pain during the triage. There are four levels of pain: 0 indicates “no pain”, 1–5 indicates “mild pain”, 6–7 indicates “moderate pain”, and 8–10 indicates “severe pain” (34).

Specialty medicine referred to the first physician assigned to older adults from the hospital’s available medical departments for standard care services in the ED of the study hospital is referred to specialty medicine. Additionally, if an older adult needed consultation with the interdisciplinary care team (ICT) after presenting in the ED with imprecise signs and symptoms for a particular medical specialty, the designation “consulted ICT” was assigned.

The whole length of time older adult patients spent in the ED from the time they arrived until they left are termed as ED-LoS. ED-LoS was categorized into two groups: ED-LoS ≤240 minutes and ED-LoS >240 minutes (prolonged ED-LoS) (35).

Statistics and data analysis

The computer software, Jamovi version 2.6.17, was used for statistical analyses (36,37) of the older adults who received healthcare services in the ED of the study hospital. The dataset included 32,053 visits of the older adult population, then 3,029 records with 2.5 standard deviations (SDs) or more away from the mean were defined as outliers (38), and they were deleted. Descriptive statistics were used to determine frequencies, mean and SD, missing data, and proportions of variables and subgroups of interest in a dataset consisting of 29,024 records. A single imputation technique was applied to deal with missing data. Missing data of numerical variables were replaced with the mean, while missing data of categorical variables were replaced with the most frequent category (39).

Logistic regression is commonly used to determine the relationship of a binary outcome with one or more predictors, which may be either categorical or continuous (40). The usability of binomial logistic regression to analyze large datasets allows researchers to understand the relationship between multiple predictors and a binary variable, and to avoid confounding effects by analyzing the association of all predictors together (41). In this study, logistic regression was conducted for univariate and multivariate models. In the multivariate model, all the predictor variables are entered simultaneously (40), to identify the associations between the ED-LoS in binary form (≤240 vs. >240 minutes) and a set of independent variables including age, gender, visiting pathway, mode of transportation, severity level, pain level, hospital admission, and medical specialty. The level of significance at a two-tailed P value with less than 0.05 was considered. The odds ratios (ORs) and their 95% confidence intervals (CIs) were calculated for estimating the odds of prolonged ED-LoS for the interested category of factors compared to the reference category. For our dependent variable, the desirable is to obtain a prolonged ED-LoS; the ED-LoS with 240 minutes and less is our reference category. For our independent variables, we considered which category has the lowest code value as the reference category, except for the specialty medicine; ophthalmology is our desired reference category because the other two categories presented extremely low proportions for this analysis. To interpret the ORs, if the 95% CI for an OR does not include 1.0, then the OR is statistically significant at the 5% level (42).

Sample size

Bujang et al. [2018] conducted a study to estimate the minimum sample size and proposed sample size guidelines for logistic regression based on an observational study with a large sample size (43). These authors recommended the rules of thumb with a minimum sample size of 500 and the formula n = 100 + 50i, where i referred to the number of independent variables (43). They confirmed these rules through the study with a large observational dataset of 70,899 records, and the findings revealed that statistics can represent the parameters in the targeted population (43).

In this study, a set of predictors includes eight independent variables; therefore, a minimum sample size is 500 (n_min = 100 + 50 × 8 = 500). As the larger sample size, the more insightful information, and a large sample size provides models with more accuracy for large data (44), the sample size for statistical analyses in this study was 1,000 records with automatic random selection from computer software. Randomization was suggested as a technique to control confounding in addition to logistic multivariate analysis (41).

Results

The dataset included 32,053 visits of the older adult population, then 3,029 records with 2.5 SDs or more away from the mean were defined as outliers (38), and they were deleted. The dataset of 29,024 records was used for descriptive statistical analysis. Characteristics of older adults visiting the ED were reported in Table 1. Older adults with a mean age of 78.9 years (SD =7.9 years), 51.8% (n=15,025) of them were female.

The proportions of referred older adults were higher than those of walk-in older adults in the oldest-old (63.8%, n=3,552 vs. 36.2%, n=2,018) and the middle-old older adults (55.9%, n=4,710 vs. 44.1%, n=3,723). The referred patients were older than walk-in older adults (mean age ± SD, 80.2±7.7 vs. 77.7±7.9 years). Of those 11,149 referred older adults, 98.2% (n=10,944) were transported by ambulance. Of 16,807 older adults transported by ambulance, 60.9% of them were hospitalized. The proportion of hospitalization of referred older adults was approximately two times higher than that of walk-in older adults (33%, n=6,802 vs. 14.4%, n=2,971). The highest proportion of hospitalized was revealed in the oldest-old older adults (54.2%, n=4,109 out of 7,576), followed by middle-old (51.4%, n=6,120 out of 11,915) and youngest-old (46.1%, n=4,393 out of 9,533).

More than one-third of the older adults were treated by internal medicine specialists (39.1%, n=11,361), followed by orthopedics (19.8%, n=5,734), urology (8.9%, n=2,544), neurology (8.4%, n=2,443), and others. The findings also indicated that 0.4% (n=119) of the older adults needed consultations with ICT. Less than half of the older adults (44.7%, n=12,951) needed urgent care. Of 21,960 older adults were assessed their pain level in the triage, 65.1% of older adults were recorded with mild pain (65.1%, n=14,301), 7% (n=1,527) with moderate pain, and only 1% (n=216) with severe pain. The descriptive findings related to severity and pain level were reported in Table 2.

ED-LoS

The older patients needed a longer time for the initial assessment, a shorter time for seeing the first physician, and a longer ED-LoS than the younger patients. Table 3 demonstrates the length of time spent in the emergency room. The average ED-LoS was 3 hours and 13 minutes (mean =193 minutes, SD =94.9 minutes). The average ED-LoS was longest for the oldest-old older adults (3 hours and 22 minutes), followed by the middle-old older adults (3 hours and 15 minutes) and the youngest-old older adults (3 hours and 4 minutes). As a result, 31.6% (n=9,158) of the older adults had prolonged ED-LoS, and they were older than those having ED-LoS less than 4 hours (mean age ± SD, 79.5±7.8 vs. 78.6±7.9 years).

Prolonged ED-LoS and its associated factors

The proportions of the older adults who experienced ED-LoS longer than 4 hours in the oldest-old adults (34.1%, n=2,582 out of 7,576) was higher than the middle-old and youngest-old adults (32.6%, n=3,889 out of 11,915, and 28.2%, n=2,687 out of 9,533, respectively). The results indicated that prolonged ED-LoS was more common in females than in males (32.9%, n=4,950 out of 15,025 vs. 30.1%, n=4,208 out of 13,999) and in hospitalization (37.6%, n=5,502 out of 14,622) than in those who were discharged from the ED (25.4%, n=3,656 out of 14,402). Compared to the non-urgent group, older adults with urgent care needs had a higher percentage of prolonged ED-LoS (33.1%, n=4,284 out of 12,951 vs. 30.4%, n=4,869 out of 16,003).

Additionally, older adults who were transported by ambulance had a higher percentage than those who visited the ED by their own vehicles (35%, n=5,888 out of 16,807 vs. 26.9%, n=3,211 out of 11,923). Those who were referred to the ED by healthcare providers had a larger percentage of prolonged ED-LoS than those who directly visited the ED (33.4%, n=3,730 out of 11,174 vs. 23.8%, n=2,248 out of 9,450). In comparison to older adults with no pain (35.4%, n=2,097 out of 5,916), moderate pain (35.2%, n=537 out of 1,527), and severe pain (38.9%, n=84 out of 216), those with mild pain had the lowest percentage of prolonged ED-LoS (32.4%, n=4,640 out of 14,301). Tables 4,5 demonstrate the findings on the frequencies of prolonged ED-LoS in older adults.

Figure 1 highlights the prevalence of extended ED-LoS in older adults by medical specialty per 10,000 per year. The largest prevalences were seen in older adults treated by internal medicine, neurology, neurosurgery, and orthopedics, respectively. Considering the combination of hospital admission and medical specialty, the percentage of prolonged ED-LoS by the total number of hospitalizations in each medical specialty was observed. The results indicated that older adults prescribed by orthopedics for hospitalization had the largest percentages of prolonged ED-LoS (42.6%, n=749 out of 1,759), followed by internal medicine (41.1%, n=3,437 out of 8,368) and surgery (35%, n=222 out of 635) (see Figure 2).

Figure 1 Prevalence of prolonged ED-LoS by specialty medicine (per 1,000 older adults). ED-LoS, length of stay in the emergency department; ICT, interdisciplinary care team.

Figure 2 Proportions of prolonged ED-LoS in older adults who were prescribed for hospitalization by medical specialty. ED-LoS, length of stay in the emergency department; ICT, interdisciplinary care team.

The findings from the binary logistic regression analysis are reported in Table 6. Identifying the potential variables associated with the ED-LoS among the older adults, variables with significant P value less than 0.05 indicated a relationship between the variable and the ED-LoS. The results from univariate analysis indicated significant differences odds of prolonged ED-LoS in older adults associated with the variables included age, pain level, hospital admission, visiting pathway, transport vehicle, and medical specialty (P<0.05). The odds of prolonged ED-LoS were 1.02 times higher (OR =1.02; 95% CI: 1.00–1.04; P=0.048) in older adults with older age compared to those with younger age. The odds of prolonged ED-LoS were significantly decreased in older adults who reported mild pain compared to those who reported no pain (OR =0.45; 95% CI: 0.33–0.62; P<0.001).

The results indicated an increased odds of prolonged ED-LoS in older adults who were prescribed for hospitalization compared to those who were discharged from the ED (OR =2.19; 95% CI: 1.67–2.88; P<0.001), in older adults who were referred by healthcare providers compared to those who directly visited the ED (OR =1.85; 95% CI: 1.36–2.25; P<0.001), and in older adults who were transported by ambulance compared to those who visited the ED by their private vehicles (OR =1.17; 95% CI: 1.27–2.23; P<0.001).

The increased odds for prolonged ED-LoS were identified; they were higher in the older adults treated by neurology (OR =8.27; 95% CI: 2.73–25.07; P<0.001), internal medicine (OR =8.18; 95% CI: 2.88–23.21; P<0.001), surgery (OR =6.14; 95% CI: 1.77–21.23; P=0.004), neurosurgery (OR =5.63; 95% CI: 1.72–18.41; P=0.004), and orthopedics (OR =4.23; 95% CI: 1.45–12.30; P=0.008) compared to those who treated by ophthalmology specialty.

Multivariate analysis was conducted, and variance inflation factor (VIF) scores were reported as values below two for all predictors, indicating no multicollinearity. The overall model was significant, ꭓ²[17] =104, P<0.001, with between 9.86% and 13.8% of the variance in the odds of prolonged ED-LoS explained by the predictor set, comprising hospitalization (compared to discharge), mild pain level (compared to no pain), and medical specialty including surgery, internal medicine, neurology, neurosurgery, and orthopedics (compared to ophthalmology) in a range from Cox and Snell to Nagelkerke =0.0986–0.138. The model of predicting prolonged ED-LoS presents 70.3% accuracy, 19.9% sensitivity, and 93.7% specificity. It means that the model correctly predicts prolonged ED-LoS 70.3% of the time, identifies 93.7% of actual cases of prolonged ED-LoS, and correctly identifies 19.9% of cases without prolonged ED-LoS.

The results from multivariate analysis revealed that the ED-LoS among older adults was potentially confounded by factors such as age, visiting pathway, and mode of transportation. The adjusted results, the ORs of prolonged ED-LoS were significantly reported across pain level, hospital admission, and specialty medicine. Compared to older adults without pain, those who reported mild pain had a 0.5-fold decreased chance of experiencing prolonged ED-LoS [adjusted OR (aOR) =0.52; 95% CI: 0.37–0.73; P<0.001].

The odds of prolonged ED-LoS for the older adults prescribed for hospitalization were 1.5 times greater than the odds of prolonged ED-LoS for those discharged from the ED (aOR =1.46; 95% CI: 1.64–2.01; P=0.02). The odds of prolonged ED-LoS were higher for older adults treated by neurology (aOR =5.00; 95% CI: 1.59–15.69; P=0.006), internal medicine (aOR =4.74; 95% CI: 1.62–13.89; P=0.005), surgery (aOR =4.66; 95% CI: 1.32–16.50; P=0.02), neurosurgery (aOR =4.18; 95% CI: 1.25–13.96; P=0.02), and orthopedics (aOR =3.41; 95% CI: 1.15–10.11; P=0.03) compared to those treated by ophthalmology (see Table 6).

Discussion

The results in this study revealed that the average ED-LoS of the older adult patients in the study hospital was also in accordance with earlier research conducted in Germany, and the standard indicators of the ED performance in Europe. On the other hand, prolonged ED-LoS was revealed in nearly one-third of the older adults. The predictor set included hospitalization, mild pain level, and medical specialty including surgery, internal medicine, neurology, neurosurgery, and orthopedics (compared to ophthalmology) were significant variables of longer ED-LoS. Additionally, ED-LoS among the older adults was potentially confounded by factors such as age, visiting pathway, and mode of transportation.

The ED-LoS is one of the well-known performance measures of care quality (45), and our target category of the ED-LoS among the older adults is the prolonged ED-LoS; the ED-LoS is defined as more than 4 hours (35). The findings of this study can be used to modify and to fulfill the routine services in the processes of receiving care in the ED. Despite the advantages of using routine data offering powerful insight, there can be limitations in the way information is recorded in large amounts of data and missing information. Handling the missing data in logistic analysis, traditional mean substitution technique was conducted to complete a dataset, but it may cause biased estimates of the findings (43,44,46). Therefore, analysis and generalized findings can be limited in this study. With a good understanding of the statistical principles of multiple imputation and significant computational resources, the approach with this robust method is recommended (40).

The findings indicated challenges encountered in prolonged ED-LoS among the older adults and indicated the potential associated factors of their prolonged ED-LoS. A prolonged ED-LoS was frequently observed in older adults, which contributed to their poorer health outcomes (7,20,47-49). The demand for specialized and complex care and treatment has been linked to the rising ED-LoS among older adults (7,29,47,50). Therefore, operational adjustments and service management, especially for older adult patients, require evidence-based solutions to reduce prolonged ED-LoS. The occurrence of prolonged ED-LoS of the patients in concern was partly presented in earlier research conducted in the German ED as a performance measure for emergency care service quality improvement (19,24).

Visiting the ED is seen as a sentinel condition for older adults who are experiencing declines in their health and quality of life; they are more likely to have complex medical conditions with complex care demands and are at a higher risk of experiencing bad outcomes during their visit (51). One of the possible factors linked to ED-LoS in older adults, according to our study, is mild pain, one of the most common conditions that affect the impact of activities of daily living and overall well-being among older adults (52). Longer length of ED stay contributes to ED crowding, which is in turn associated with adverse outcomes including acute pain (20). An older adult’s ED-LoS can be greatly impacted by their level of pain. Inadequate pain treatment can induce a longer stay, delayed ambulation, a higher risk of delirium, and functional impairment; then all these elements work together to increase the length of time spent in the ED (20). According to Fry et al. [2016], older adults with cognitive impairment may encounter much more challenges in obtaining appropriate pain care in EDs if there is no standardized pain assessment screening tool available (53).

According to this study, the proportion of older adult patients admitted to hospitals, referred to the ED by healthcare professionals, and transported by ambulance rises with age. In this study, over 60% of older adults who were referred to the ED by healthcare professionals were hospitalized. Berger et al. [2024] stated in the German Hospital Report that 25% of patients of all age groups accessed to the emergency care through the on-call medical service of the statutory health insurance associations, and 75% through direct visits to the EDs (54). Usually, referred patients essentially obtain an initial affirmation of their significant need for medical care from qualified healthcare professionals for additional specialized investigation and treatments; their complicated medical conditions may result in ED-LoS. According to data from 2019, Gries et al. [2022] reported that the rescue and emergency medical service, emergency physician, helicopter rescue service, and community-based physicians were the common sources of referral to the ED of the university hospital in Saxony, Germany (55). Their results indicated that, in comparison to patients who are self-referred, patients of all other referral types, particularly patients referred to the ED by helicopter rescue service, had a noticeably higher chance of being admitted to the hospital (55).

Hong et al. [2022] conducted 3-year statewide research in South Korea, the findings indicated that patients referred by medical professionals were older than those visited the ED directly, and over 60% of them were hospitalized (56). Referred patients in South Korea tended to have a higher degree of illness, higher admission, and longer ED-LoS than direct-visit patients (56). Since patients had to wait for inpatient beds to be allocated, the increasing demand for hospitalization may have also resulted in longer ED-LoS (56). In turn, a study in Saxony, Germany by Gries et al. [2022], 37.4% of referred patients who visited the ED were admitted to the hospital, and one in six self-referred patients required further inpatient medical care (55). When compared to self-referrals and referrals from community-based physicians, the inpatient admission rate is greater for patients referred to the ED by the emergency rescue service (55). However, this is possible that diverse national policies are the cause of the different results. Moreover, it is challenging to accurately measure the number of patients who come to the ED on their own initiative or who do not necessarily need treatment there because this data is not currently documented with enough specificity (57).

High percentage of older adults in this study had prolonged ED-LoS, even though their average ED-LoS was less than 4 hours. Older adults’ ED-LoS increased considerably with age; oldest-old patients demonstrated the longest ED-LoS in comparison to those in other age groups. Longer ED-LoS may result from comorbidities and complexity among older adults throughout the registration, initial ED clinician encounter, waiting for disposition, admission decision, and discharge processes (47). According to a study conducted in South Korea by Lee et al. [2018], oldest-old patients spent more time in the ED (30). In the same way, Otto et al. [2022] reported that whereas patients 61 years of age and older the average time spent up to around 227–240 minutes in the German ED, patients between the ages of 18 and 60 spent an average of less than 3 hours (around 174–204 minutes) (19).

In a concept analysis study of the long ED-LoS, the authors provided the definitions of prolonged ED-LoS among patients with cut-offs wide-ranging between 4 and 48 hours (49). Currently, the ED-LoS at the study hospital should not be more than 4 hours, in accordance with the standard indicators of the ED performance in Europe (48). According to Wallstab et al. [2022], eight hospitals in Germany had an average ED-LoS of more than 240 minutes, whereas 16 hospitals had an average of less than 2 hours (24). These average times, however, were not limited to the aging population (24). George et al. [2006] compared data from 1990 and 2004 to provide a historical perspective on the effect of the UK’s aging population on the ED-LoS (12). The results indicated that the proportion of patients with an ED-LoS of less than 2 hours was higher in 1990 (82.8%) than in 2004 (53.2%) (12). A historical perspective on the impact of the growing aging population on the ED-LoS in the UK was illustrated by the authors through a comparison of data from 1990 and 2004 (12). These authors stated that it is unable to return the target of the ED-LoS with less than 2 hours (12). Furthermore, in the review study, Ogliari et al. [2022] found that older adults were more likely to experience the ED-LoS for longer than 4 hours (20). Time targets are likely more appropriate when they are targeted at and modified for a particular population (49). Therefore, the ED-LoS targets for older adult patients should be reviewed.

Durations of ED-LoS associated with diverse variables, including older patients, gender, middle triage, comorbidity, delirium incidence, hospitalization, consultation rate, higher patient-to-physician and patient-to-nurse ratios, shift work, timing in the observation unit, overcrowding, laboratory and imaging utilization, and inefficient organization, have been reported as associated factors of long ED-LoS durations. (8,21,49,50,58-61). Otto et al. [2022] linked the influencing variables of the ED-LoS to patient, disease, and organization (19). Changing this kind of variable requires resource consumption, time, and many processes (25,62). Therefore, this type of variable was not included in this study.

After visiting the ED, older adults, particularly those with complex medical conditions, are more likely to require hospitalization, and their admission to the hospital often prolongs their ED-LoS (20). A prolonged ED stay for older admitted patients is associated with an increased risk of in-hospital adverse events (63). According to a study conducted in Germany, by Otto et al. [2022], the ED-LoS depended on admission status, triage level, age, and presentation complaint (19). These authors indicated that walk-in patients had a lower ED-LoS than healthcare provider referrals (19), which is in accordance with our findings. They also indicated that hospitalized patients spent 64 minutes longer in the ED than non-admitted patients (19). Even though the ED-LoS was longer for patients who were referred to the ED by healthcare professionals, patients were satisfied if the initiative for the referral came from their physicians (19). Considering the German referral system, Rosemann et al. [2006] mentioned the general practice as the initiator, who mainly refers patients to consultants for clear diagnosis, reduces diagnostic uncertainty, and further appropriate treatments at the hospital (64).

The findings in this study depicted the higher proportions of prolonged ED-LoS compared to the older adults with ED-LoS less than 4 hours in internal medicine, neurology, neurosurgery, surgery, and orthopedic specialties. These mentioned specialties undoubtedly deal with complexity and a large volume of patients, and the patients need time for investigations and consultations. In the study by Rojsaengroeng et al. [2023], patients who required consultations with internal medicine physician indicated a prolonged ED-LoS approximately 1.5 times more than the patients who did not (65).

In many nations, including Germany, outpatient treatments with operations and non-operation have been increased significantly to avoid unnecessary inpatient treatments (54). The main reason older adults should not be admitted to hospitals is that they are more likely to have adverse outcomes that could make it difficult for them to go to nursing homes, post-acute care facilities, or rehabilitation centers (51). In Germany, the reasons to avoid hospital admission are not only increasing potential care at the ED and decreasing the risk of a nosocomial infection (hospital-acquired infection), but also dealing with the staff shortage and financial constraints (54). Surgical and non-surgical procedures are diagnosed and treated by medical disciplines like neurosurgery, orthopedics, and surgery. It usually takes time for patients receiving care and treatment from these disciplines to undergo their specific investigations and therapies. Orthopedic physicians typically handle patients of all ages who arrive at the emergency room with musculoskeletal issues, including fractures, in the study hospital. Therefore, the results revealed that the older persons who were prescribed by this medical specialty for hospitalization had a high extended ED-LoS.

The requirement of consultations and investigations because of acuity and complexity influences the time used in the ED (8,59). The complexity of frail older adults is not addressed by triage alone in the initial assessment, even though it is crucial to maximize the care of critically ill or injured patients who present to the emergency room (51). According to Hesselink et al. [2019], ED-based consultant geriatrician and streaming of care by dedicated staff in the ED could reduce the ED-LoS among the older adults (66). Integrating geriatric assessment into the patient flow within the ED to identify specific geriatric syndromes allows healthcare providers to address significant older adult care needs and make better clinical decisions (51). The identification of seniors at risk (ISAR) is a useful screening tool for selecting older adults, and this tool can be administered by a trained nurse after triage, without any further workload for the ED staff (13).

To maximize the care of older adults in the ED, Sanon et al. [2019] presented a concept paper about the Acute Care for Elder (ACE) model (51). These authors highlighted the crucial roles of the geriatric interdisciplinary team in the ACE model (51). By this model, the ED nurse performed universal triage screening and provided a self-report screening tool, while the transitional care nurse or geriatric nurse practitioner provided more focused geriatric assessments in the ED and facilitated the coordination required for complex transitional care planning of older adults being discharged from the ED (51). Therefore, adoption of the ISAR, a validated geriatric screening tool for direct consultation with geriatricians and/or geriatric nurse practitioners, should benefit health outcomes and decrease ED-LoS among older adults.

Concerning on the evidence-based solutions for dealing and reducing the ED-LoS, according to a study in the Netherlands by van der Veen et al. [2018], accelerating the processes of laboratory and/or radiology testing and consulting, decision-making, and discharge could reduce the ED-LoS (35). Therefore, effective patient flow within-ED and strengthening collaboration among the ED staff are meaningful for optimization of emergency service structures and processes of admission and discharge (67,68). According to Leggio et al. [2022], integrating an Interprofessional Vertical Flow Track with Pivot Triage may enhance patient flow and shorten ED-LoS; this intervention involves prioritizing patient care based on the severity of their condition (triage) and then guiding them through a dedicated, streamlined pathway with a focus on interprofessional collaboration (69).

The utilization of emergency care services and the enhancement of care quality in ED by decreasing the ED-LoS among the aging population are the focus of this study. ED-care professionals should take consideration of the age-based ED-LoS target times. Based on these findings, and literatures of approved implementations, the ED-LoS target time for older adults may be achievable in less than 4 hours. It is advised that older adults’ ED-LoS be routinely monitored and benchmarked. It is advised that the ED’s patient flowchart be modified for older adult patients to include brief warning elements pertaining to predictors of the prolonged ED-LoS. To determine whether older adults will be assigned directly with a geriatrician and/or a multidisciplinary geriatric care team, the ED nurse should administer the ISAR following triage. Furthermore, adopting the ACE model could have a meaningful effect on optimizing the standard of care for older adults.

Limitations

This research has limitations. Although there was little information on gender-diverse older adults from routine statistics, we are concerned about how older adults identify and express their gender and how these differences impact health outcomes and health care services. Furthermore, organizational factors such as bed occupancy, admission-discharge ratio, patient volume, and patients per professional, and other potential variables reported in earlier studies in various contexts were not selected. Consequently, ED-LoS among the older adults in this study could be potentially confounded by these mentioned factors.

Conclusions

The findings indicated challenges encountered in prolonged ED-LoS of the older adults. A set of predictors, comprising hospitalization, mild pain level, and medical specialty, including internal medicine, surgery, neurology, neurosurgery, and orthopedics significantly impacted to the ED-LoS among the older adults. To decrease the ED-LoS, integrating a patient-flowchart within the ED particular in medical specialties associated with prolonged ED-LoS should be modified for older adult patients. Additionally, identifying older adult patients who require direct consultation with a geriatrician might be facilitated by the ED nurse doing the ISAR following triage into routine care services in the ED. Furthermore, adopting the ACE model, modifying the ED-LoS targets for aging population, and the ED-LoS benchmarking could have a meaningful effect on optimizing the standard of care and sustainably decreasing the ED-LoS among older adults. It is essential that ED staff are being educated about various facets of health and illness in older adults.

Acknowledgments

We wish to thank our colleagues working at the ED for their support and cooperation. We are grateful to Dr. Wichai Chattanawaree (MD, Geriatric Medicine), Faculty of Medicine, Siriraj Hospital, Mahidol University, for suggestions on the finding report. We appreciate the support of Assistant Professor Dr. Patcharin Sungwan, School of Nursing, University of Phayao, in proofreading this work.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jphe.amegroups.com/article/view/10.21037/jphe-24-117/rc

Peer Review File: Available at https://jphe.amegroups.com/article/view/10.21037/jphe-24-117/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jphe.amegroups.com/article/view/10.21037/jphe-24-117/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments and the Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation). The requests for permission to conduct the study in hospital were made by the research team. Patients and the public were not involved in this study. Data were gathered from the routine data of hospital database and processed anonymously in an Excel database before performing the data analysis.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. World Health Organization. Ageing and Health. 2025. Available online: https://www.who.int/news-room/fact-sheets/detail/ageing-and-health
2. Ankuda CK, Freedman VA, Covinsky KE, et al. Population-Based Screening for Functional Disability in Older Adults. Innov Aging 2021;5:igaa065. [Crossref] [PubMed]
3. DESTATIS-Federal Statistical Office of Germany. Population: Deaths, life expectancy. 2024. Available online: https://www.destatis.de/EN/Themes/Society-Environment/Population/Deaths-Life-Expectancy/_node.html
4. Eurostat-Statistics Explained. Healthy life years statistics. 2024. Available online: https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Healthy_life_years_statistics#Healthy_life_years_at_age_65
5. Rouzet D, Sánchez AC, Renault T, et al. Fiscal challenges and inclusive growth in ageing societies. Paris: OECD Economic Policy Papers; 2019. Available online: https://doi.org/10.1787/c553d8d2-en
6. OECD. Pensions at a Glance 2015: OECD and G20 indicators. Paris: OECD Publishing; 2015. Available online: https://doi.org/10.1787/pension_glance-2015-en
7. Burgess L, Ray-Barruel G, Kynoch K. Association between emergency department length of stay and patient outcomes: A systematic review. Res Nurs Health 2022;45:59-93. [Crossref] [PubMed]
8. Casalino E, Wargon M, Peroziello A, et al. Predictive factors for longer length of stay in an emergency department: a prospective multicentre study evaluating the impact of age, patient’s clinical acuity and complexity, and care pathways. Emerg Med J 2014;31:361-8. [Crossref] [PubMed]
9. Hwang SY, Park EH, Shin ES, et al. Comparison of factors associated with atypical symptoms in younger and older patients with acute coronary syndromes. J Korean Med Sci 2009;24:789-94. [Crossref] [PubMed]
10. Ukkonen M, Jämsen E, Zeitlin R, et al. Emergency department visits in older patients: a population-based survey. BMC Emerg Med 2019;19:20. [Crossref] [PubMed]
11. Pilotto A, Custodero C, Maggi S, et al. A multidimensional approach to frailty in older people. Ageing Res Rev 2020;60:101047. [Crossref] [PubMed]
12. George G, Jell C, Todd BS. Effect of population ageing on emergency department speed and efficiency: a historical perspective from a district general hospital in the UK. Emerg Med J 2006;23:379-83. [Crossref] [PubMed]
13. Salvi F, Morichi V, Grilli A, et al. Screening for frailty in elderly emergency department patients by using the Identification of Seniors At Risk (ISAR). J Nutr Health Aging 2012;16:313-8. [Crossref] [PubMed]
14. Brammen D, Greiner F, Kulla M, et al. AKTIN - The German Emergency Department Data Registry - real-time data from emergency medicine : Implementation and first results from 15 emergency departments with focus on Federal Joint Committee’s guidelines on acuity assessment. Med Klin Intensivmed Notfmed 2022;117:24-33. [Crossref] [PubMed]
15. Christ M, Grossmann F, Winter D, et al. Modern triage in the emergency department. Dtsch Arztebl Int 2010;107:892-8. [Crossref] [PubMed]
16. German Federal Joint Committee (G-BA). Regulations of the Federal Joint Committee on a tiered system of emergency structures in hospitals pursuant to Section 136c (4) of the Fifth Book of the German Social Code (SGB V). 2020. Available online: https://www.g-ba.de/richtlinien/103/
17. Clarke GM, Conti S, Wolters AT, et al. Evaluating the impact of healthcare interventions using routine data. BMJ 2019;365:l2239. [Crossref] [PubMed]
18. Schuur JD, Hsia RY, Burstin H, et al. Quality measurement in the emergency department: past and future. Health Aff (Millwood) 2013;32:2129-38. [Crossref] [PubMed]
19. Otto R, Blaschke S, Schirrmeister W, et al. Length of stay as quality indicator in emergency departments: analysis of determinants in the German Emergency Department Data Registry (AKTIN registry). Intern Emerg Med 2022;17:1199-209. [Crossref] [PubMed]
20. Ogliari G, Coffey F, Keillor L, et al. Emergency department use and length of stay by younger and older adults: Nottingham cohort study in the emergency department (NOCED). Aging Clin Exp Res 2022;34:2873-85. [Crossref] [PubMed]
21. Ayenew T, Gedfew M, Fetene MG, et al. Prolonged length of stay and associated factors among emergency department patients in Ethiopia: systematic review and meta-analysis. BMC Emerg Med 2024;24:212. [Crossref] [PubMed]
22. Belayneh AG, Temachu YZ, Messelu MA, et al. Prolonged length of stay and its associated factors at adult emergency department in amhara region comprehensive specialized hospitals, northwest Ethiopia. BMC Emerg Med 2023;23:34. [Crossref] [PubMed]
23. Payne K, Risi D, O’Hare A, et al. Factors that contribute to patient length of stay in the emergency department: A time in motion observational study. Australas Emerg Care 2023;26:321-5. [Crossref] [PubMed]
24. Wallstab F, Greiner F, Schirrmeister W, et al. German emergency department measures in 2018: a status quo based on the Utstein reporting standard. BMC Emerg Med 2022;22:5. [Crossref] [PubMed]
25. Newhouse RP, Dearholt S, Poe S, et al. Organizational change strategies for evidence-based practice. J Nurs Adm 2007;37:552-7. [Crossref] [PubMed]
26. DESTATIS-Federal Statistical Office of Germany. Older people in Germany and the EU. 2016. Available online: https://www.destatis.de/EN/Themes/Society-Environment/Population/Current-Population/Publications/Downloads-Current-Population/brochure-older-people-eu-0010021169004.pdf?__blob=publicationFile&v=4
27. Central Information Office [Zentrales Informationsbüro (ZIB)] - Dessau-Rosslau. Senioren. 2024. Available online: https://verwaltung.dessau-rosslau.de/soziales-bildung/senioren.html
28. German Federal Joint Committee (G-BA). Gestuftes System von Notfallstrukturen in Krankenhäusern. 2018. Available online: https://www.g-ba.de/themen/bedarfsplanung/notfallstrukturen-krankenhaeuser/
29. European Union. Ageing Europe: Looking at the lives of older people in the EU. 2020 edition. Luxembourg: Publications Office of the European Union; 2020. Available online: https://ec.europa.eu/eurostat/documents/3217494/11478057/KS-02-20-655-EN-N.pdf/9b09606c-d4e8-4c33-63d2-3b20d5c19c91?t=1604055531000
30. Lee SB, Oh JH, Park JH, et al. Differences in youngest-old, middle-old, and oldest-old patients who visit the emergency department. Clin Exp Emerg Med 2018;5:249-55. [Crossref] [PubMed]
31. Heidari S, Babor TF, De Castro P, et al. Sex and Gender Equity in Research: rationale for the SAGER guidelines and recommended use. Res Integr Peer Rev 2016;1:2. [Crossref] [PubMed]
32. Gräff I, Goldschmidt B, Glien P, et al. The German Version of the Manchester Triage System and its quality criteria--first assessment of validity and reliability. PLoS One 2014;9:e88995. [Crossref] [PubMed]
33. Heller AR, Juncken K. Primary care in the central emergency department. Anästh Intensivmed 2020;61:168.
34. Boonstra AM, Stewart RE, Köke AJ, et al. Cut-Off Points for Mild, Moderate, and Severe Pain on the Numeric Rating Scale for Pain in Patients with Chronic Musculoskeletal Pain: Variability and Influence of Sex and Catastrophizing. Front Psychol 2016;7:1466. [Crossref] [PubMed]
35. van der Veen D, Remeijer C, Fogteloo AJ, et al. Independent determinants of prolonged emergency department length of stay in a tertiary care centre: a prospective cohort study. Scand J Trauma Resusc Emerg Med 2018;26:81. [Crossref] [PubMed]
36. Jamovi (version 2.6). 2024. Available online: https://www.jamovi.org
37. R Core Team. R: A Language and environment for statistical computing (version 4.4). 2024. Available online: https://cran.r-project.org
38. Pollet TV, van der Meij L. To remove or not to remove: the impact of outlier handling on significance testing in testosterone data. Adapt Human Behav Physiol 2017;3:43-60.
39. Kang H. The prevention and handling of the missing data. Korean J Anesthesiol 2013;64:402-6. [Crossref] [PubMed]
40. Ranganathan P, Pramesh CS, Aggarwal R. Common pitfalls in statistical analysis: Logistic regression. Perspect Clin Res 2017;8:148-51. [Crossref] [PubMed]
41. Jager KJ, Zoccali C, Macleod A, et al. Confounding: what it is and how to deal with it. Kidney Int 2008;73:256-60. [Crossref] [PubMed]
42. Norton EC, Dowd BE, Maciejewski ML. Odds Ratios-Current Best Practice and Use. JAMA 2018;320:84-5. [Crossref] [PubMed]
43. Bujang MA, Sa’at N, Sidik TMITAB, et al. Sample Size Guidelines for Logistic Regression from Observational Studies with Large Population: Emphasis on the Accuracy Between Statistics and Parameters Based on Real Life Clinical Data. Malays J Med Sci 2018;25:122-30. [Crossref] [PubMed]
44. Gumpili SP, Das AV. Sample size and its evolution in research. IHOPE Journal of Ophthalmology 2022;1:9-13.
45. Farimani RM, Karim H, Atashi A, et al. Models to predict length of stay in the emergency department: a systematic literature review and appraisal. BMC Emerg Med 2024;24:54. [Crossref] [PubMed]
46. Chevret S, Seaman S, Resche-Rigon M. Multiple imputation: a mature approach to dealing with missing data. Intensive Care Med 2015;41:348-50. [Crossref] [PubMed]
47. Adams G, Hoffman JW, Goldberg EM. It’s time for an overhaul of emergency department operations for people living with dementia. J Public Health Emerg 2024;8:12.
48. Allen T, Walshe K, Proudlove N, et al. Measurement and improvement of emergency department performance through inspection and rating: an observational study of emergency departments in acute hospitals in England. Emerg Med J 2019;36:326-32. [Crossref] [PubMed]
49. Andersson J, Nordgren L, Cheng I, et al. Long emergency department length of stay: A concept analysis. Int Emerg Nurs 2020;53:100930. [Crossref] [PubMed]
50. Sweeny A, Keijzers G, O’Dwyer J, et al. Predictors of a long length of stay in the emergency department for older people. Intern Med J 2020;50:572-81. [Crossref] [PubMed]
51. Sanon M, Hwang U, Abraham G, et al. ACE Model for Older Adults in ED. Geriatrics (Basel) 2019;4:24. [Crossref] [PubMed]
52. Amo-Agyei S, Maurer J. Pain and subjective well-being among older adults in the developing world : A comprehensive assessment based on the WHO Study on Global Ageing and Adult Health. Econ Hum Biol 2024;54:101406. [Crossref] [PubMed]
53. Fry M, Chenoweth L, Arendts G. Assessment and management of acute pain in the older person with cognitive impairment: A qualitative study. Int Emerg Nurs 2016;24:54-60. [Crossref] [PubMed]
54. Berger E, Rödiger H, Busse R. Reform der Notfallversorgung und des Rettungswesens. In: Klauber J, Wasem J, Beivers A, et al. editors. Krankenhaus-Report 2024: Strukturreform. Berlin, Heidelberg: Springer; 2024:209-22.
55. Gries A, Schrimpf AM, Dercks NV. Hospital Emergency Departments—Utilization and Resource Deployment in the Hospital as a Function of the Type of Referral. Dtsch Arztebl Int 2022;119:640-6. [Crossref] [PubMed]
56. Hong SI, Kim JS, Kim YJ, et al. Characteristics of Patients Who Visited Emergency Department: A Nationwide Population-Based Study in South Korea (2016-2018). Int J Environ Res Public Health 2022;19:8578. [Crossref] [PubMed]
57. Köster C, Wrede S, Herrmann T, et al. Ambulante Notfallversorgung. Analyse und Handlungsempfehlungen. Göttingen: AQUA—Institut für angewandte Qualitätsförderung und Forschung im Gesundheitswesen GmbH; 2016. Available online: https://www.vdek.com/presse/publikationen/_jcr_content/par/publicationelement/file.res/Aqua-Gutachten-Notfallversorgung.pdf
58. Ba-Aoum M, Hosseinichimeh N, Triantis KP, et al. Statistical analysis of factors influencing patient length of stay in emergency departments. International Journal of Industrial Engineering and Operations Management 2023;5:220-39.
59. Yoon P, Steiner I, Reinhardt G. Analysis of factors influencing length of stay in the emergency department. CJEM 2003;5:155-61. [Crossref] [PubMed]
60. Elder NM, Mumma BE, Maeda MY, et al. Emergency Department Length of Stay Is Associated with Delirium in Older Adults. West J Emerg Med 2023;24:532-7. [Crossref] [PubMed]
61. Sankham P, Utriyaprasit K, Tankumpuan T, et al. Predictors of older patients´ stay in the emergency department. Journal of Thailand Nursing and Midwifery Council 2022;37:125-45.
62. Rousseau DM, Ten Have S. Evidence-based change management. Organizational Dynamics 2022;51:100899.
63. Ackroyd-Stolarz S, Read Guernsey J, Mackinnon NJ, et al. The association between a prolonged stay in the emergency department and adverse events in older patients admitted to hospital: a retrospective cohort study. BMJ Qual Saf 2011;20:564-9. [Crossref] [PubMed]
64. Rosemann T, Wensing M, Rueter G, et al. Referrals from general practice to consultants in Germany: if the GP is the initiator, patients’ experiences are more positive. BMC Health Serv Res 2006;6:5. [Crossref] [PubMed]
65. Rojsaengroeng R, Sri-on J, Kongkaew N, et al. Factors associated with the prolongation of emergency department length of stay before and during the COVID-19 pandemic in a university hospital in Thailand. J Public Health Emerg 2023;7:26.
66. Hesselink G, Sir Ö, Schoon Y. Effectiveness of interventions to alleviate emergency department crowding by older adults: a systematic review. BMC Emerg Med 2019;19:69. [Crossref] [PubMed]
67. Paling S, Lambert J, Clouting J, et al. Waiting times in emergency departments: exploring the factors associated with longer patient waits for emergency care in England using routinely collected daily data. Emerg Med J 2020;37:781-6. [Crossref] [PubMed]
68. Samadbeik M, Staib A, Boyle J, et al. Patient flow in emergency departments: a comprehensive umbrella review of solutions and challenges across the health system. BMC Health Serv Res 2024;24:274. [Crossref] [PubMed]
69. Leggio WJ, Carhart E, Bruckner AA, et al. Reduced Emergency Department Length of Stay and Proportion of Patients Who Left Without Being Seen Following Implementation of an Interprofessional Vertical Flow Track With Pivot Triage: A Retrospective Pre-/Postintervention Evaluation. Adv Emerg Nurs J 2022;44:136-43. [Crossref] [PubMed]