Elderly patients with non-specific complaints at the emergency department have a high risk for admission and 30-days mortality

Chest pain, dyspnea and abdominal pain are all specified symptoms frequently appearing at the Emergency Department (ED) that can be caused by a limited number of possible diagnoses. In those patients having these specific chief-complaints, there are usually protocols and work-up programs that can be followed and support healthcare decisions [1‐3]. However, there are also non-specific complaints (NSC) that appear at the ED such as fatigue, generalized weakness, altered mental status, failure to eat and drink, reduced mobility and falling where ED protocols usually do not apply [4, 5]. Studies have shown that NSC are frequent among patients at the ED and highest among older adults [4, 6]. Older adults often have multiple co-morbidities which can disguise classical signs and symptoms as well as age-related physiological changes such as failure to develop fever and lack of chest pain when having a heart attack which may result in NSC in a situation when deteriorating [7, 8].

Older adults have more complex medical needs that consequently results in longer waiting time at the ED and increased use of resources [9‐13]. The risk of adverse events for elderly patients increases with length of stay (LOS) at the ED [14, 15]. Older patients with NSC often appear in the ED, and even though this can mean worse clinical outcome and longer LOS, they are still a patient category that is usually not prioritized [16‐21]. NSC are one of the most challenging conditions for an ED-physician since there are not any specific protocols to follow and the cause of NSC can be caused by everything from life threatening conditions, lack of home health care or natural aging [4, 20, 22].

The aim of this study was to compare older adults with NSC at the ED with patients with specific chief-complaints such as dyspnea, chest pain and abdominal pain regarding admission rate and 30-day mortality.

A total of 15,528 patients aged ≥ 65 years visited the ED in RH during 2016. Of these, 4927 (32%) were included in the study based on chief-complaint. The top 10 reasons for ED-visits are shown in Table 1. The distribution of chief-complaints was 1599 (32%) patients with chest pain, 1343 (27%) with dyspnea, 1460 (30%) with abdominal pain and 525 (11%) with NSC.

Descriptive statistics of demographic data are shown in Table 2. Patients with NSC had a mean age of 80 years. The patients who sought for dyspnea had a mean age of 80 years, chest pains 77 years and abdominal pain 78 years (p = 0.001). Patients with NSC had a mean LOS of 4,7 h at the ED which was significantly higher compared to chest pain, dyspnea and abdominal pain. Mean bed-days for the whole population was 4.2 days compared to patients with NSC who had a mean LOS of 5.6 days.

Patients with dyspnea and NSC had the highest hospital admission rate 79% vs 70% compared to patients with chest pain (63%) and abdominal pain (61%) (p =  < 0.001) which is shown in Table 2.

The overall 30-days mortality in the total study group was 6% where patients with dyspnea had the highest mortality rate of 10% followed by patients with NSC 9%, abdominal pain 5% and chest pain 2% which is displayed in Table 2.

A multivariate regression analysis for hospital admission adjusted for age, previous health conditions and LOS at the ED had an Odds ratio 1.6 (Confidence Interval [CI] 1.27–2.00) for dyspnea compared to NSC. The analysis for hospital admission is also separated for age, previous health conditions and LOS at the ED, illustrated in Table 3.

Overall Cox regression for 30-days mortality with NSC as reference showed a Hazard ratio of 1.12 (CI 0.80–1.58) for dyspnea, 0.23 (CI 0.15–0.36) and 0.53 (CI 0.36–0.77) for chest pain and abdominal pain respectively. Table 4 shows an adjusted Cox regression model and Hazard ratio for NSC, dyspnea, chest pain and abdominal pain.

Among older adults visiting the ED, those with NSC and dyspnea had the highest age. NSC as the chief-complaint was associated with a higher rate of ED duration > 4 h and readmission within 72 h. Patients with NSC had the longest in-hospital LOS. Older adults with dyspnea in the ED were associated with the highest hospital admission rate followed by NSC which had a higher hospital admission rate compared to patients presenting with chest and abdominal pain. Earlier studies have shown a similar high admission rate for elderly patients with chief complaint dyspnea and chest pain [27]. The high admission rate for geriatric patients may be a sign of the difficulties to distinguish between life-threatening conditions or the sign of chronic geriatric diseases at the ED. The 30-day mortality rate for patients with dyspnea and NSC were significantly higher compared to patients having chest pain or abdominal pain at the ED.

The present study has shown that patients in the ED with NSC have a high average age, which is consistent with other studies [4‐6]. Increasing age entails an increased risk of hospital admission and likewise the occurrence of ≥ 4 comorbidities. LOS at ED ≥ 4 h was associated with a reduced risk of hospitalization which may seem odd. It was likely a certain degree of selection and that long LOS in the ED could be used as an observation time in practice.

In this study, 11% of the study cohort were defined as having NSC as their chief-complaint and in comparison, previous studies have shown that up to 20% of older adults at the ED present with NSC [28]. The difference may be due to the lack of international definition of non-specific complaints. Most commonly fatigue, generalized weakness, altered mental status, failure to eat and drink, reduced mobility and falling are considered NSC but sometimes these symptoms are referred to as the geriatric syndrome at the ED [20]. Patients with NSC have a poor clinical outcome and are often prioritized as low acuity at the ED. Studies show that screening for frailty could be one useful parameter to improve triage for older adults at the ED. Frailty appears to be of more use than chronological age when predicting outcome for geriatric trauma patients [29, 30]. Several studies have shown that a geriatric approach at the ED will shorten LOS at the ED and in-hospital LOS once admitted [31, 32].

The chief complaint with the highest risk for hospital admission was dyspnea followed by NSC. Still, NSC was associated with higher admission risk compared to both chest and abdominal pain and this finding was regardless of age group, degree of comorbidity and time at ED. Older patients with NSC were associated with a significantly higher 30-days mortality rate compared to patients with chest pain and abdominal pain and the mortality rate was comparable to patients with dyspnea, these data’s are comparable with previous studies [4, 33‐35]. Chest pain as chief complaint will be prioritized higher at the ED compared to NSC. This study is in line with earlier studies and shows that patients with NSC have a higher admission rate and mortality rate than patients presenting with chest pain at the ED [36]. In addition, patients with NSC have a longer LOS at the ED, a high admission rate and the highest number of bed-days once admitted. Despite the lower number of patients with NSC in this study the 30-days mortality rate were in line with previous studies [4].

These results demonstrate that NSC in older adults can be difficult to assess for ED staff even though these individuals may be at significant risk for hospital requirements and 30-day mortality. There may be a need to improve routines regarding the handling of this patient group in the ED and previous study have reported there are limitations in existing risk stratification instruments for older adults visiting the ED [37].

Older patients who present with NSC at the ED are associated with a high risk for hospital admission and 30-days mortality. In addition, patients with NSC compared to patients with SC have a longer LOS at the ED, a high admission rate and the highest number of bed-days once admitted. This study indicates that ED staff should be more vigilant when an elderly patient presents with NSC at the ED. Further research is needed to approach how to best care for older patients with NSC to reduce morbidity and mortality.

In the study, it is not possible to evaluate the degree of acuity or morbidity in terms other than the number of comorbidities. Nor can frailty be assessed, which is likely to be of decisive importance in this patient group. Trauma is a common diagnosis in the ED and was not included as a chief complaint in this study. The intention was to identify NSC and compare them with chief complaints that are judged to be relatively serious in the ED and where there are guidelines on how these patients should be managed at ED. Trauma has a wide variety of conditions and was not assessed being not appropriate in this regard.

The terminologies fatigue, altered mental status, falling, generalized weakness and non-specific complaints to define NSC could have been somewhat restrictive and leading to lower acceptance rate. However, these chief-complaint used compiled to a high rate compared to others and the purpose was not to overlap to specific complaint associated with specific diseases such as vertigo and neurology.

It should be emphasized that the results in this study are only associations, and it is not possible to draw any conclusions regarding causality.