Abstract
Objectives:
To evaluate the short-term mortality of adult patients presenting to the emergency department (ED) with altered mental status (AMS) as compared to other common chief complaints.
Methods:
Observational cohort study of adult patients (age ≥ 40) who presented to an academic ED over a 1-year period with five pre-specified complaints at ED triage: AMS, generalized weakness, chest pain, abdominal pain, and headache. Primary outcomes included 7- and 30-day mortality. Hazard ratios (HR) were calculated with 95% confidence intervals (CI) using Cox proportional hazards models adjusted for age, acuity level, and comorbidities.
Results:
A total of 9,850 ED visits were included for analysis from which 101 (1.0%) and 295 (3.0%) died within 7 and 30 days, respectively. Among 683 AMS visits, the 7-day mortality rate was 3.2%. Mortality was lower for all other chief complaints, including generalized weakness (17/1170, 1.5%), abdominal pain (32/3609, 0.9%), chest pain (26/3548, 0.7%), and headache (4/840, 0.5%). After adjusting for key confounders, patients presenting with AMS had a significantly higher risk of death within 7 days of ED arrival than patients presenting with chest pain (HR 3.72, 95% CI 2.05 to 6.76, p <.001). Similarly, we found that patients presenting with AMS had a significantly higher risk of dying within 30 days compared to patients with chest pain (HR 3.65, 95% CI 2.49 to 5.37, p <.001), and headache (HR 2.09, 95% CI 1.09 to 4.01, p = 0.026). Differences were not statistically significant for comparisons with abdominal pain and generalized weakness, but confidence intervals were wide.
Conclusion:
Patients presenting with AMS have worse short-term prognosis than patients presenting to the ED with chest pain or headache. AMS may indicate an underlying brain dysfunction (delirium), which is associated with adverse outcomes and increased mortality.
Keywords: altered mental status, delirium, geriatric, emergency medicine
1.0. INTRODUCTION
Altered mental status (AMS) is a common presentation to the emergency department (ED), and it may be present in up to 40% of geriatric patients being evaluated in the acute care setting.[1] The term “AMS” is nonspecific and can be used to describe both alterations in level of consciousness and alterations in cognitive function.[2] In practice, the term is ambiguous and can refer to abnormal behavior, agitation, confusion, somnolence, disorientation among other acute or chronic changes in a patient’s vigilance, mental content, or attentiveness.[2] The differential diagnosis is broad and strongly affected by age. In older adults, for example, AMS frequently represents acute delirium.[3–6]
If the AMS is indeed acute and represents a change from baseline (i.e., does not represent chronic alterations from chronic conditions such as dementia), they will frequently land within the spectrum of acute brain failure that includes delirium, stupor, or coma.[7] All three entities of brain dysfunction are associated with increased morbidity and mortality, often requiring immediate evaluation.[7] Beyond the eventual need of stabilizing these patients through airway management or other interventions, a change in mentation also represents a challenge to emergency physicians in the sense that this “complaint” will often not have a discrete diagnosis at the end of the ED course, which makes prognostication difficult. In this context, clinicians may underestimate the prognostic value of non-specific complaints such as AMS in the ED, which could lead to inappropriate disposition decisions.
In this study, our objective was to evaluate the association between AMS as the main reason for ED visit and mortality within 7 and 30 days. We hypothesized that AMS would be associated with a worse prognosis (increased short-term mortality) than other common chief complaints including chest pain, abdominal pain, weakness, and headache in a population of adults aged 40 years or older.
2.0. METHODS
This manuscript adheres to the STrengthening of the Reporting of OBservational studies in Epidemiology (STROBE) guideline.[8] This study was approved by the institutional review board, and only patients with prior research authorization for medical records review were included. A prior publication evaluating the effect of age (older versus middle-age adults) on outcomes of patients presenting with similar complaints used the same cohort.[8]
2.1. Study design, setting, and participants
This was an observational cohort study of consecutive adult patients aged ≥ 40 years who presented to an academic ED between January 1st, 2017, and December 31st, 2017, with one of five pre-specified chief complaints of interest: AMS, generalized weakness, chest pain, abdominal pain, or headache. We used the cutoff of 40 years to purposefully have a more homogeneous cohort who would be more likely to present with similar chief complaints and who would have a higher risk of our primary outcome of interest (mortality). Our ED is an academic institution with an annual census of 80,000 visits per year, including 23,000 (29%) by patients 65 and older. All data for this study were automatically abstracted from the electronic health record (EHR) using data retrieval tools.
2.2. Chief complaints and baseline characteristics
A chief complaint was defined as the main concern that brought the patient to the ED, as recorded by triage nurses. Upon arrival to the ED, patients or a proxy reported symptoms to nursing staff, who entered these symptoms/complaints as free text in the EHR. The EHR’s chief complaint text field for each ED visit was reviewed by one board-certified emergency physician and one emergency medicine resident to assign one of the five chief complaints of interest or mark the visit as excluded from the study if none were present. Altered mental status was defined as chief complaints in which one of the following keywords were present: “altered mental status”, “mental status change”, “AMS”, “altered level of consciousness”, “confusion”, “confused”, “delirium”, “agitated”, or “agitation,” or obvious misspellings of these keywords. Generalized weakness was defined as chief complaints mentioning “generalized weakness”, “bilateral weakness”, or “weakness” without a specification of laterality. Focal weakness was not included in this category. Chest pain was defined as chief complaints mentioning “chest pain”, “chest discomfort”, “chest burning”, “chest fullness”, “chest tightness”, “chest congestion”, “chest pressure”, or “chest soreness”. Abdominal pain was defined as those with “abdominal pain”, “abdominal bloating”, “abdominal distension”, “abdominal discomfort”, “abdominal cramping”, “abdominal tenderness”, “flank pain”, “epigastric pain”, “LLQ pain”, “LUQ pain”, “RLQ pain”, or “RUQ pain”. Lastly, headache was defined as those with “headache”, “HA”, or “head pain”. When an ED visit had more than one chief complaint recorded, we used the first-listed complaint with the assumption that this would be the main reason for presentation. Patients were categorized in these groups of chief complaints regardless of their final ED diagnosis.
Baseline characteristics of the cohort were extracted from the EHR, including age, sex, Emergency Severity Index (ESI) triage level, and comorbidities. All International Classification of Diseases (ICD) diagnoses starting 3 years prior to the index date up to the index date (date of ED visit) were retrieved from the EHR. Only those diagnoses that appeared twice at least 30 days apart were included. The following comorbidities were ascertained for the calculation of the Charlson Comorbidity Index: myocardial infarction, congestive heart failure, peripheral vascular disease, cerebrovascular disease, dementia, chronic pulmonary disease, ulcer, mild liver disease, diabetes, hemiplegia, moderate/severe renal disease, moderate/severe liver disease, metastatic solid tumor, AIDS, rheumatologic disease, other cancer.[9] The Charlson Comorbidity Index is a composite score that incorporates several comorbidities and estimates patients’ underlying burden of diseases.
2.3. Outcomes
Our outcomes of interest were mortality within 7 and 30 days of the date of arrival in the ED. Patients were followed through their EHR encounters up to 30 days, regardless of admission or discharge status. To evaluate the completeness of follow-up in our cohort, we obtained the date on which the patient was last seen alive, as noted in the EHR. For survival analyses, patients were censored at date of death, date last seen alive in the EHR, or at the end of the period of interest - 7 days or 30 days. There were 9252 visits (93.9% of the cohort) with at least 7 days of follow-up or mortality prior to 7 days, and 9005 (91.4% of the cohort) visits with at least 30 days of follow-up or mortality prior to 30 days.
2.4. Data analysis
Statistical analyses were performed with R software version 3.6.2. Continuous features were summarized as median and interquartile ranges (IQRs) or means and standard deviations (SD) as appropriate based on data distribution. Categorical features were summarized as counts and percentages. Survival curves were plotted, and a log-rank test implemented to determine if risk of mortality differed across chief complaints. To assess the risk of 7- and 30-day mortality by chief complaint, Cox proportional hazards models were fit using death within 7 or 30 days of ED arrival as the outcome. Hazard ratios (HR) were calculated with 95% confidence intervals (CI) to compare the risk of dying between patients presenting with AMS and those with other common chief complaints at ED triage. To evaluate whether age would confound the associations, we calculated HRs after adjusting for age (i.e., adding age as a covariate in the Cox proportional hazard models). Cox proportional models were also fit to adjust for acuity level (through ESI triage), and comorbidities (through the Charlson comorbidity index). A stratified analysis by chief complaint and age was also performed to evaluate 7-day and 30-day mortality across different age groups presenting with different chief complaints.
3.0. RESULTS
3.1. Baseline characteristics
A total of 9,850 ED visits were identified for analysis, including 683 (6.9%) with a chief complaint of AMS, 1,170 (11.9%) with generalized weakness, 3,548 (36.0%) with chest pain, 3,609 (36.6%) with abdominal pain, and 840 (8.5%) with headache. Patients with chief complaints of AMS or generalized weakness were older than those with chest pain, abdominal pain, or headache. The presence of comorbidities also differed across groups. Dementia, for example, was present in 14.1% of those with a chief complaint of AMS, but only 1.8% of those with abdominal pain. (Table 1)
Table 1.
Baseline characteristics of the cohort stratified by chief complaint category.
| Median (IQR) or counts (%) | |||||
|---|---|---|---|---|---|
| Altered Mental Status (N = 683) |
Generalized Weakness (N = 1170) |
Chest Pain (N = 3548) |
Abdominal Pain (N = 3609) |
Headache (N = 840) |
|
| Age and sex | |||||
| Age (years) | 72 (62, 84) | 75 (65, 83) | 63 (53, 75) | 61 (51, 73) | 56 (47, 69) |
| Female | 329 (48.2%) | 551 (47.1%) | 1635 (46.1%) | 1965 (54.5%) | 543 (64.6%) |
| Comorbidities* (diagnoses prior to ED visit) | |||||
| MI | 32 (4.7%) | 57 (4.9%) | 293 (8.3%) | 99 (2.7%) | 11 (1.3%) |
| CHF | 98 (14.4%) | 188 (16.1%) | 414 (11.7%) | 213 (5.9%) | 41 (4.9%) |
| Peripheral vascular disease | 94 (13.8%) | 148 (12.7%) | 376 (10.6%) | 258 (7.2%) | 34 (4.1%) |
| Cerebrovascular disease | 84 (12.3%) | 116 (9.9%) | 193 (4.9%) | 130 (3.6%) | 41 (4.9%) |
| Dementia | 96 (14.1%) | 68 (5.8%) | 67 (1.9%) | 64 (1.8%) | 16 (1.9%) |
| Chronic pulmonary disease | 99 (14.5%) | 78 (9.3%) | 522 (14.7%) | 421 (11.7%) | 78 (9.3%) |
| Gastroduodenal ulcer | 9 (1.3%) | 14 (1.2%) | 20 (0.6%) | 45 (1.3%) | 0 (0.0%) |
| Liver disease | 59 (8.6%) | 51 (4.4%) | 75 (2.1%) | 130 (3.6%) | 13 (1.6%) |
| Diabetes | 177 (25.9%) | 250 (21.4%) | 622 (17.5%) | 405 (11.2%) | 767 (8.5%) |
| Diabetes with organ damage | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) |
| Hemiplegia | 18 (2.6%) | 29 (2.5%) | 24 (0.7%) | 35 (1.0%) | 6 (0.7%) |
| Renal disease | 127 (18.6%) | 218 (18.6%) | 356 (10.0%) | 283 (7.8%) | 46 (5.5%) |
| Metastatic solid tumor | 28 (4.1%) | 100 (8.6%) | 99 (2.8%) | 226 (6.3%) | 34 (4.1%) |
| AIDS | 3 (0.4%) | 2 (0.2%) | 9 (0.3%) | 6 (0.2%) | 3 (0.4%) |
| Rheumatologic disease | 31 (4.5%) | 47 (4.0%) | 128 (3.6%) | 94 (2.6%) | 35 (4.2%) |
| Other cancer | 85 (12.5%) | 245 (20.9%) | 345 (9.7%) | 548 (15.2%) | 85 (10.1%) |
| Triage levels | |||||
| ESI Level 1 | 15 (2.2%) | 2 (0.2%) | 22 (0.6%) | 1 (0.03%) | 4 (0.5%) |
| ESI Level 2 | 329 (48.2%) | 258 (22.1%) | 2877 (81.1%) | 129 (3.6%) | 78 (9.3%) |
| ESI Level 3 | 335 (49.1%) | 889 (76.0%) | 627 (17.7%) | 3438 (95.3%) | 675 (80.4%) |
| ESI Level 4 | 3 (0.4%) | 19 (1.6%) | 21 (0.6%) | 39 (1.1%) | 81 (9.6%) |
| ESI Level 5 | 0 (0.0%) | 0 (0.0%) | 1 (0.03%) | 1 (0.03%) | 1 (0.1%) |
| Missing | 1 (0.1%) | 2 (0.2%) | 0 (0.0%) | 1 (0.03%) | 1 (0.1%) |
MI, myocardial infarction; CHF, congestive heart failure; AIDS, acquired immunodeficiency syndrome.
Comorbidities data was missing for 24 visits of the cohort.
3.2. 7-day mortality
A total of 7 patients (0.07%) died in the ED, and 101 (1.0%) died within 7 days of ED arrival. Among 683 visits with chief complaints of AMS, the 7-day mortality rate was 3.2% (22 deaths, 95% CI 2.0% to 4.8%). Mortality rates were lower for all other chief complaints, including chest pain (0.7%, 95% CI 0.5% to 1.0%) and abdominal pain (0.9%, 95% CI 0.6% to 1.3%). (Table 2) Figure 1 shows the observed survival of our cohort, by chief complaint. A log-rank test identified a significant difference in the risk of mortality within 7 days of ED arrival (p <.001). After adjusting for age, we found that patients presenting with AMS had a significantly higher risk of dying within 7 days than those with abdominal pain (HR 3.17, 95% CI 1.79 to 5.61, p <.001), generalized weakness (HR 2.24, 95% CI 1.19 to 4.23, p <.05), chest pain (HR 3.63, 95% CI 2.01 to 6.56, p <.001), or headache (HR 6.33, 95% CI 2.06 to 19.46, p <.01). When we adjusted for age, acuity level, and comorbidities, confidence intervals for the effect estimates became wide but AMS remained with a statistically significant higher risk of 7-day mortality than chest pain (HR 3.72, 95% CI 2.05 to 6.76, p <.001). (Table 3)
Table 2.
Number of deaths and short-term mortality rates stratified by chief complaints and age groups (40–65 and ≥ 65 years).
| 7-day mortality N of deaths, rate (95% CI) | 30-day mortality N of deaths, rate (95% CI) | |
|---|---|---|
| Altered mental status (N = 683) | 22, 3.2% (2.0%, 4.8%) | 54, 7.9% (6.0%, 10.2%) |
| Age 40–65 (N = 212) | 8, 3.8% (1.6%, 7.3%) | 13, 6.1% (3.3%, 10.3%) |
| Age ≥ 65 (N = 471) | 14, 3.0% (1.6%, 4.9%) | 41, 8.7% (6.3%, 11.6%) |
| Generalized weakness (N = 1170) | 17, 1.5% (0.9%, 2.3%) | 74, 6.3% (5.0%, 7.9%) |
| Age 40–65 (N = 280) | 3, 1.0% (0.2%, 3.1%) | 13, 4.6% (2.5%, 7.8%) |
| Age ≥ 65 (N = 890) | 14, 1.6% (0.9%, 2.6%) | 61, 6.9% (5.3%, 8.7%) |
| Chest pain (N = 3548) | 26, 0.7% (0.5%, 1.0%) | 59, 1.7% (1.3%, 2.1%) |
| Age 40–65 (N = 1916) | 9, 0.5% (0.2%, 0.9%) | 16, 0.8% (0.5%, 1.4%) |
| Age ≥ 65 (N = 1632) | 17, 1.0% (0.6%, 1.7%) | 43, 2.6% (1.9%, 3.5%) |
| Abdominal pain (N = 3609) | 32, 0.9% (0.6%, 1.3%) | 96, 2.7% (2.2%, 3.2%) |
| Age 40–65 (N = 2086) | 12, 0.6% (0.3%, 1.0%) | 34, 1.6% (1.1%, 2.3%) |
| Age ≥ 65 (N = 1523) | 20, 1.3% (0.8%, 2.0%) | 62, 4.1% (3.1%, 5.2%) |
| Headache (N = 840) | 4, 0.5% (0.1%, 1.2%) | 12, 1.4% (0.7%, 2.5%) |
| Age 40–65 (N = 565) | 3, 0.5% (0.1%, 1.5%) | 7, 1.2% (0.5%, 2.5%) |
| Age ≥ 65 (N = 275) | 1, 0.4% (0.01%, 0.2%) | 5, 1.8% (0.6%, 4.2%) |
| Total cohort (N = 9850) | 101, 1.0% (0.8%, 1.2%) | 295, 3.0% (2.7%, 3.4%) |
| Age 40–65 (N = 5059) | 35, 0.7% (0.5%, 1.0%) | 83, 1.6% (1.3%, 2.0%) |
| Age ≥ 65 (N = 4791) | 66, 1.4% (1.1%, 1.8%) | 212, 4.4% (3.9%, 5.1%) |
Figure 1.
7-day survival by chief complaint identified at ED triage. Log-rank test p <.001.
Table 3.
Hazard ratios for altered mental status (AMS) as compared to different chief complaints and their risk of short-term mortality
| Unadjusted HR (95% CI) | P Value | Age-Adjusted* HR (95% CI) | P Value | Age-Acuity-Comorbidity Adjusted† HR (95% CI) | P Value | |
|---|---|---|---|---|---|---|
| 7-day mortality | ||||||
| AMS vs. Abdominal Pain | 3.64 (2.12, 6.27) | <.001 | 3.17 (1.79, 5.61) | <.001 | 0.95 (0.48, 1.89) | .894 |
| AMS vs. Generalized Weakness | 2.25 (1.20, 4.25) | <.05 | 2.24 (1.19, 4.23) | <.05 | 1.28 (0.65, 2.49) | .472 |
| AMS vs. Chest Pain | 4.39 (2.49, 7.74) | <.001 | 3.63 (2.01, 6.56) | <.001 | 3.72 (2.05, 6.76) | <.001 |
| AMS vs. Headache | 6.49 (2.24, 18.83) | <.001 | 6.33 (2.06, 19.46) | <.01 | 1.71 (0.56, 5.19) | .345 |
| 30-day mortality | ||||||
| AMS vs. Abdominal Pain | 3.04 (2.18, 4.24) | <.001 | 2.36 (1.66, 3.34) | <.001 | 1.42 (0.97, 2.10) | .074 |
| AMS vs. Generalized Weakness | 1.29 (0.91, 1.83) | .156 | 1.31 (0.92, 1.86) | 0.133 | 1.08 (0.75, 1.57) | .673 |
| AMS vs. Chest Pain | 4.89 (3.38, 7.08) | <.001 | 3.69 (2.51, 5.41) | <.001 | 3.65 (2.49, 5.37) | <.001 |
| AMS vs. Headache | 5.43 (2.90, 10.16) | <.001 | 4.27 (2.20, 8.29) | <.001 | 2.09 (1.09, 4.01) | .026 |
Adjusted for patient age (age included as a continuous variable in the Cox model).
Adjusted for patient age (continuous), ESI triage level (categorical: does not assume equal change in risk of mortality between each level), and Charlson comorbidity index (severity weighted sum).
3.3. 30-day mortality
A total of 295 (3.0%) patients died within 30 days. Among 683 visits with chief complaint of AMS, there were 54 deaths (30-day mortality rate of 7.9%, 95% CI 6.0% to 10.2%). Mortality rates were lower for all other chief complaints, including chest pain (1.7%, 95% CI 1.3% to 2.1%) and abdominal pain (2.7%, 95% CI 2.2% to 3.2%). (Table 2) Figure 2 shows the observed survival of our cohort within 30 days, stratified by chief complaint. A log-rank test identified a significant difference in the risk of mortality within 30 days of ED arrival (p <.001). After adjusting for age, we found that patients presenting with AMS had a significantly higher risk of dying within 30 days compared to patients with abdominal pain (HR 2.36, 95% CI 1.66 to 3.34, p <.001), chest pain (HR 3.69, 95% CI 2.51 to 5.41, p <.001), and headache (HR 4.27, 95% CI 2.20 to 8.29, p <.001). For the outcome of 30-day mortality, the difference between patients presenting with AMS and those with generalized weakness was not statistically significant (HR 1.31, 95% CI 0.92 to 1.86, p = 0.1333). In the model adjusted for age, acuity level, and comorbidities, AMS remained with a statistically significant higher risk of 30-day mortality than chest pain (HR 3.65, 95% CI 2.49 to 5.37) and headache (HR 2.09, 95% CI 1.09 to 4.01). (Table 3)
Figure 2.
30-day survival by chief complaint identified at ED triage. Log-rank test p <.001.
4.0. DISCUSSION
In this large cohort of ED visits by middle-aged and older adults, those identified as having AMS as their main reason for presentation had worse short-term mortality than patients with other common chief complaints. 7-day mortality risk of those presenting with AMS was interestingly higher regardless of age when compared to those presenting with chest pain, abdominal pain, generalized weakness, or headache. Risk of 30-day mortality was higher with AMS than chest pain and headache even after accounting for age, acuity level, and comorbidities. Our results suggest that AMS is an important prognostic factor. The short-term high-risk nature of this chief complaint may lower thresholds for hospitalization and heighten awareness around the need to identify a cause of the AMS. Patients who present with AMS and are being discharged will require clear communication with family/care partners about the elevated risk associated with this complaint and need for close follow-up with strict return precautions.
AMS can be viewed as a manifestation of acute brain failure and it often represents acute delirium in geriatric ED patients.[3,9,10] In fact, the simple presence of this presenting complaint significantly raises the likelihood of a delirium diagnosis during their ED stay. In a multicenter cohort study in Turkey, Aslaner and colleagues reported that among 822 patients aged ≥ 65 years presenting to the ED with acute AMS, 55.7% were diagnosed with delirium.[3] Han and colleagues in a US-based cohort reported that 82.6% of geriatric ED patients with AMS had delirium. The positive likelihood ratio of AMS for the diagnosis of delirium was 34, having a large effect on posttest probability of delirium.[3] Most recently, a systematic review also found a strong association between AMS and ED delirium with an odds ratio of 13 (moderate certainty evidence with 3 studies and 1,420 patients).[11] Given existing evidence on the relationship between AMS and delirium, we suspect that a sizeable number of patients in our cohort might have had delirium. Numerous studies have found that delirium in the ED is common (approximately 10%),[12] poorly detected,[13,14]and associated with longer hospital length of stay[15,16] and increased mortality.[16,17] The well-known association between delirium and increased mortality[18] could partly explain our findings.
Patients presenting with AMS may not provide accurate histories. Without knowledge of events surrounding the change in mentation, providers must test broadly, hoping to identify the etiology. Lack of history could lead to missed diagnoses, which could also account for the increased mortality identified in this study.
We found a higher prevalence of dementia in patients presenting with AMS (14%) compared to other common chief complaints. Dementia is one of the greatest risk factors for delirium[11] and is associated with poor cognitive reserve, difficulties expressing symptoms and challenges to effective communication in the ED.[19] Furthermore, delirium in a patient with dementia should be treated as an emergency because it accelerates cognitive decline.[20,21]
The risk associated with a chief complaint of AMS may be underappreciated by ED clinicians. The increased risk of short-term mortality identified in this study indicates that patients presenting with AMS must be taken seriously. AMS is a sign of critical illness and must result in downstream action. Unfortunately, there are few treatment strategies specifically for the management of stable ED patients with AMS. Strategies to address delirium and the use of multicomponent approaches have been proposed and may be helpful in the context of AMS.[4,22–24] Treatment for delirium is aimed at treating the underlying cause and so ED providers caring for patients with AMS must search diligently for a cause of the change in mental status.
We will continue to learn more about delirium treatment and safe disposition planning for patients with AMS or delirium. In the meantime, we must at the very least recognize this subgroup of patients may require different care than we currently provide given their increased risk of short-term death. Those presenting with confusion, agitation, delirium and altered consciousness might require closer follow up with primary providers who know them well. If they are living alone they may need short term medical assistance and if being discharged from the ED given detailed discharge instructions, including how to prevent and recognize delirium and other signs of deterioration that should trigger prompt return to the ED. When we have not identified a cause for the delirium, we may need to lower our threshold to admit to the hospital.
Lastly, we found that patients with a chief complaint of generalized weakness had similar 30-day mortality than patients presenting with AMS. Like AMS, weakness is a non-specific symptom and may have prognostic implication for geriatric ED patients. Previous studies have found that patients presenting to the ED with nonspecific complaints are at higher risk of adverse outcomes.[25–27] Interestingly, in a cohort of geriatric patients presenting with nonspecific complaints (mostly generalized weakness), the 30-day mortality rate was 6%,[27] very close to the 30-day mortality rate of 6.3% found in our cohort of patients presenting with generalized weakness. While the presence of AMS or generalized weakness did not increase the probability of bacterial infection in older adults in one study,[27] older adults are particularly susceptible to atypical presentations of severe illnesses which can lead to misdiagnosis. Altered mental status or generalized weakness might be the only presenting symptom of an acute myocardial infarction or infection in a frail older adult, for example.
5.0. LIMITATIONS
This was an analysis of data collected for non-research purposes, and its accuracy relies on the quality of the medical record. Second, this is a single site study in an academic center and its generalizability to other settings might be limited; for example, our admission rates are approximately 30%, and as high as 56% for older adults. In addition, in the models for which we adjusted for age, acuity level, and comorbidities, the confidence intervals for the effect estimates were wide due to our limited sample size. Future larger observational studies evaluating the prognostic ability of chief complaints in the ED will need to consider the need of adjusting for the most important confounders when estimating their sample size. Third, ESI is a crude measure of severity of illness and the possibility of residual confounding cannot be ruled out. Future studies may need to use more comprehensive surrogates for severity of illness. Lastly, we chose to restrict the cohort and compare specific chief complaints as opposed to comparing AMS to all ED chief complaints. This resulted in less heterogeneity in the data but could impact the external validity to a general ED cohort.
6.0. CONCLUSIONS
Patients presenting with AMS have worse short-term prognosis than those with chest pain even after accounting for age, acuity level, and comorbidities. Altered mental status may indicate an underlying brain dysfunction (mostly delirium), which is associated with adverse outcomes including increased mortality. Our findings indicate that patients presenting with a chief complaint of AMS must be taken seriously, and providers must stay vigilant about the care we provide to these patients given their poor short-term prognosis. These findings can help clinicians with prognostication in the ED, especially in patients where a discrete diagnosis is not immediately available.
Authors acknowledgements:
JAS, LOJS, AG, MMJ and FB designed the study. AG, LOJS and AM conducted procurement and cleaning of data. MMJ and FB provided methodological expertise. AG, AM, and LOJS analyzed the data. All authors provided content expertise on the topic. JAS drafted the manuscript and all authors contributed substantially to its revisions. FB was the senior and mentor for this study project.
Funding:
This research was supported in part though the CCaTS Small grant program, part of Mayo Clinic CCaTS grant number UL1TR000135 from the National Center for Advancing Translational Sciences (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 view of NIH.
Footnotes
The authors of this manuscript have no conflict of interest to disclose.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Naughton BJ, Moran MB, Kadah H, Heman-Ackah Y, Longano J. Delirium and Other Cognitive Impairment in Older Adults in an Emergency Department. Ann Emerg Med. 1995;25(6):751–755. doi: 10.1016/S0196-0644(95)70202-4 [DOI] [PubMed] [Google Scholar]
- 2.Cantrill SV, Campbell M, Gallagher E, et al. Clinical Policy for the Initial Approach to Patients Presenting With Altered Mental Status. Ann Emerg Med. 1999;33(February):251–281. [DOI] [PubMed] [Google Scholar]
- 3.Aslaner MA, Boz M, Çelik A, et al. Etiologies and delirium rates of elderly ED patients with acutely altered mental status: a multicenter prospective study. Am J Emerg Med. 2017;35(1):71–76. doi: 10.1016/j.ajem.2016.10.004 [DOI] [PubMed] [Google Scholar]
- 4.Carpenter CR, Hammouda N, Linton EA, et al. Delirium Prevention, Detection, and Treatment in Emergency Medicine Settings: A Geriatric Emergency Care Applied Research (GEAR) Network Scoping Review and Consensus Statement. Acad Emerg Med. 2021;28(1):19–35. doi: 10.1111/acem.14166 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Han JH, Schnelle JF, Ely EW. The relationship between a chief complaint of “altered mental status” and delirium in older emergency department patients. Acad Emerg Med. 2014;21(8):937–940. doi: 10.1111/acem.12436 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Nagaraj G, Burkett E, Hullick C, Carpenter CR, Arendts G. Is delirium the medical emergency we know least about? EMA - Emerg Med Australas. 2016;28(4):456–458. doi: 10.1111/1742-6723.12639 [DOI] [PubMed] [Google Scholar]
- 7.Han JH, Wilber ST. Altered Mental Status in Older Patients in the Emergency Department. Clin Geriatr Med. 2013;29(1):101–136. doi: 10.1016/j.cger.2012.09.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: Guidelines for Reporting Observational Studies. Vol 12.; 2014. doi: 10.1016/j.ijsu.2014.07.013 [DOI] [Google Scholar]
- 9.Quan H, Sundararajan V, Halfon P, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005;43(11):1130–1139. doi: 10.1097/01.mlr.0000182534.19832.83 [DOI] [PubMed] [Google Scholar]
- 10.Inouye SK, Westendorp RGJ, Saczynski JS. Delirium in elderly people. Lancet. 2014;383(9920):911–922. doi: 10.1016/S0140-6736(13)60688-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Oliveira J. e Silva L, Berning MJ, Stanich JA, et al. Risk Factors for Delirium in Older Adults in the Emergency Department: A Systematic Review and Meta-Analysis. Ann Emerg Med. Published online 2021. doi: 10.1016/j.annemergmed.2021.03.005 [DOI] [PubMed] [Google Scholar]
- 12.Barron EA, Holmes J. Delirium within the emergency care setting, occurrence and detection: A systematic review. Emerg Med J. 2013;30(4):263–268. doi: 10.1136/emermed-2011-200586 [DOI] [PubMed] [Google Scholar]
- 13.Lewis LM, Miller DK, Morley JE, Nork MJ, Lasater LC. Unrecognized delirium in ED geriatric patients. Am J Emerg Med. 1995;13(2):142–145. doi: 10.1016/0735-6757(95)90080-2 [DOI] [PubMed] [Google Scholar]
- 14.Sanders AB. Missed delirium in older emergency department patients: A quality-of-care problem. Ann Emerg Med. 2002;39(3):338–341. doi: 10.1067/mem.2002.122273 [DOI] [PubMed] [Google Scholar]
- 15.Saravay SM, Kaplowitz M, Kurek J, et al. How Do Delirium and Dementia Increase Length of Stay of Elderly General Medical Inpatients? Psychosomatics. 2004;45(3):235–242. doi: 10.1176/appi.psy.45.3.235 [DOI] [PubMed] [Google Scholar]
- 16.Han JH, Shintani A, Eden S, et al. Delirium in the emergency department: An independent predictor of death within 6 months. Ann Emerg Med. 2010;56(3):244–252.e1. doi: 10.1016/j.annemergmed.2010.03.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Kennedy M, Enander RA, Tadiri SP, Wolfe RE, Shapiro NI, Marcantonio ER. Delirium risk prediction, healthcare use and mortality of elderly adults in the emergency department. J Am Geriatr Soc. 2014;62(3):462–469. doi: 10.1111/jgs.12692 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Wilson JE, Mart MF, Cunningham C, et al. Delirium. Nat Rev Dis Prim. 2020;6(1). doi: 10.1038/s41572-020-00223-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Han JH, Bryce SN, Ely EW, et al. The effect of cognitive impairment on the accuracy of the presenting complaint and discharge instruction comprehension in older emergency department patients. Ann Emerg Med. 2011;57(6). doi: 10.1016/j.annemergmed.2010.12.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Blass JP. Delirium Accelerates Cognitive Decline in Alzheimer Disease. Vol 2009.; 2009. doi: 10.1016/s0513-5117(09)79130-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Torpilliesi T, Bellelli G, Trabucchi M. Delirium superimposed on dementia is more predictive of adverse outcomes than dementia alone. Int J Geriatr Psychiatry. 2010;25(4):435–436. doi: 10.1002/gps.2342 [DOI] [PubMed] [Google Scholar]
- 22.Schnitker L, Arendts G, Carpenter CR, et al. Prevention of delirium in older adults with dementia: A systematic literature review. J Gerontol Nurs. 2020;46(10):43–54. doi: 10.3928/00989134-20200820-02 [DOI] [PubMed] [Google Scholar]
- 23.Kennedy M, Webb M, Gartaganis S, et al. ED-DEL: Development of a change package and toolkit for delirium in the emergency department. J Am Coll Emerg Physicians Open. 2021;2(2). doi: 10.1002/emp2.12421 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Bond P, Goudie K. Identifying and Managing Patients with Delirium in Acute Care Settings. Vol 27.; 2015. doi: 10.7748/nop.27.9.28.s19 [DOI] [PubMed] [Google Scholar]
- 25.Malinovska A, Nickel CH, Bingisser R. Trajectories of survival in patients with nonspecific complaints. Eur J Intern Med. 2018;55:e17–e18. doi: 10.1016/j.ejim.2018.06.020 [DOI] [PubMed] [Google Scholar]
- 26.Safwenberg U, Terént A, Lind L. The Emergency Department presenting complaint as predictor of in-hospital fatality. Eur J Emerg Med. 2007;14(6):324–331. doi: 10.1097/MEJ.0b013e32827b14dd [DOI] [PubMed] [Google Scholar]
- 27.Wachelder JJH, Stassen PM, Hubens LPAM, et al. Elderly emergency patients presenting with non-specific complaints: Characteristics and outcomes. PLoS One. 2017;12(11). doi: 10.1371/journal.pone.0188954 [DOI] [PMC free article] [PubMed] [Google Scholar]