- Email: [email protected]
Acute cognitive impairment in elderly ED patients: Etiologies and outcomes
Acute Cognitive Impairment in Elderly ED Patients: Etiologies and Outcomes JAMES L. WOFFORD, MD, MS,* LAURA R. LOEHR, MD,t EARL SCHWARTZ, MD:I: Despite the common occurrence of acute cognitive impairment in elderly emergency department (ED) patients, there is much uncertainty regarding the evaluation and management of this syndrome. We performed a retrospective cohort study of all patients 60 years of age and older transported by emergency medical services (EMS) to hospital EDs in Forsyth County, North Carolina, during 1990 specifically for evaluation of acute cognitive impairment. Five percent (227 of 4,688) of EMS transports during this time period were for the purpose of evaluation of acute cognitive impairment. Compared with community-dwelling patients (n -- 105), nursing home patients (n = 47) had a higher prevalence of final ED diagnoses indicative of infection (42.5% v 13.3%) and a lower prevalence of diagnoses indicative of cerebrovascular disease (10.6% v 22.9%) as the etiology of cognitive impairment. The rates of hospitalization and mortality were 74.3% and 28.9%, respectively. The projected aging of the US population and the high prevalence of this syndrome among elderly patients make better understanding of this syndrome essential for ED providers. (Am J Emerg Meal 1996;14:649-653. Copyright © 1996 by W.B. Saunders Company)
cognitive impairment in hospitalized patients, 7,8 few studies of cognitive impairment in the ED setting are available. A major problem with research efforts on this subject is that of defining the target syndrome. The nomenclature of cognitive impairment is confusing and variable. 9,m Furthermore, objective quantitative measures of cognitive impairment are not often used in the ED setting. As a result, clinicians fail to recognize and document the presence of cognitive impairment. Tracing clinical decisions becomes difficult because written acknowledgement of the syndrome is quickly displaced by diagnoses that are more familiar. The purpose of this study was to identify and follow a cohort of elderly ED patients with acute cognitive impairment as the sole reason for ED presentation. Examination of the subsequent ED and hospital course offers data on the etiology, disposition, and outcome of this syndrome.
METHODS The evaluation and treatment of acute cognitive impairment in the geriatric patient is a task that commonly confronts providers in the hospital emergency department (ED). Management of this syndrome is acknowledged to be more difficult for elderly than for younger patients, ~ for several reasons. First, chronic cognitive impairment is common among elderly p e r s o n s Y Discerning whether there is an acute mental status change in a chronically demented patient is difficult, especially when the patient is not familiar to the provider. Second, cognitive impairment among elderly patients frequently results from acute illnesses unrelated to the central nervous system. 4-6 Possible etiologies include infection, medication side effects, and cardiovascular disorders, all of which are c o m m o n among elderly patients. A third reason for the difficulty in management o f acute cognitive impairment in the elderly ED patient is that research into this common geriatric syndrome is not common. While a substantial amount of research has focused on
From the *Department of Internal Medicine, and SDepartment of Emergency Medicine, Bowman Gray School of Medicine of Wake Forest University, Winston-Salem, NC, and the #Department of Internal Medicine, St. Louis University, St. Louis, MO. Manuscript received July 10, 1995, returned September 5, 1995; revision received November 6, 1995, accepted November 16, 1995. Presented at the annual meeting of the American Geriatrics Society, May 1995. Address reprint requests to Dr Wofford, Department of Internal Medicine, Bowman Gray School of Medicine of Wake Forest University, Medical Center BIvd, Winston-Salem, NC 27157. Key Words: Aging, emergency services, cognitive disorder, delirium, geriatrics. Copyright © 1996 by W.B. Saunders Company 0735-6757/96/1407-000855.00/0
Patients were identified using computerized data on EMS transports from a semi-urban North Carolina county, Forsyth County, during the year 1990. Data were available for all patients 60 years of age and older who were transported by the Forsyth County EMS from January 1, 1990, to December 31, 1990. These data included information on patient age, race, gender, location of residence, reason for transport, and destination, all of which were determined by EMS personnel at the time of transport. Data were restricted to nonconvalescent transports to one of two full-service hospitals, one 700-bed tertiary care teaching hospital with a level 1 trauma center and one 1,000-bed community, primarily nonacademic hospital. These two hospitals, the only ones in the county that provide ED services, are both staffed by attending and resident physicians. Reasons for transport recorded at the time of transport by EMS personnel were coded into one of 45 categories, in accordance with requirements of the state EMS supervisory office. Each transport was coded as a single diagnosis. Piloting of the data collection form revealed that for patients with the EMS diagnosis of stroke, seizure, or syncope, EMS diagnoses were consistent to a large degree with the final ED and hospital diagnosis. Because assigning a diagnosis to these patients seemed clear even before ED evaluation, patients with these diagnoses were not considered eligible for this study. For patients with no specific reason for transport according to EMS data (n = 1,086), EMS encounter forms were manually reviewed by two authors (JLW, LRL) to identify patients transported to EDs specifically for the evaluation and management of acute cognitive impairment. EMS documentation denoting decreased level of consciousness or confusion without other obvious problems identified patients who were transported specifically for evaluation of acute cognitive impairment and who were considered eligible for this study. Patients were excluded from the study if ED or hospital records were not available or complete. Patient characteristics abstracted from EMS and ED records included medical history, medication history, final ED diagnosis, and disposition. Particular attention was given to previous diag649
650
AMERICAN JOURNAL OF EMERGENCY MEDICINE • Volume 14, Number 7 • November 1996
noses and medications that reflected underlying cognitive impairment (ie, chronic dementia, organic brain syndrome, multi-infarct dementia). Details of the EMS management focused on medical therapies beyond intravenous line placement or oxygen administration. In the case of hospitalization, information on the hospital course, final hospital diagnosis, length of stay, and disposition was abstracted from the hospital records. Because information on follow-up was not reliably available in the hospital chart for patients discharged from the ED, we did not attempt to collect follow-up data on patients who were not hospitalized. Because nursing home and community-dwelling patients are populations known to have important differences, these two groups were compared with respect to demographics, history of chronic cognitive impairment, psychoactive medication use, hospitalization, and mortality. Based on the final written ED diagnosis, the etiology of the mental status change was categorized as either cerebrovascular, cardiac, infection, metabolic/toxic effect, medication-related, trauma, psychiatric, simple syncope, other causes, or unknown/descriptive. Since the patient's final ED diagnosis frequently reflected uncertainty on the part of the ED provider and the potential for multiple etiologies, a patient may have been placed into more than one category of etiology. Diagnoses classified as cerebrovascular included the written ED diagnosis of intracranial bleeding, cerebrovascular accident, or possible transient ischemic attacks. Cardiac causes included those due to congestive heart failure, cardiac dysrhythmia, or myocardial infarction. The diagnoses of pneumonia, urinary tract infection, sepsis, and "rule out sepsis" were included in the category of infection. The category "toxic/ metabolic" included the diagnoses of acute alcohol ingestion and electrolyte disturbances, while medication-related diagnoses included those thought secondary to prescribed medications (whether related to withdrawal or overdose). When the final ED diagnosis either reflected simply a symptom complex (ie, acute mental status changes) or repeated the patient's medical history (ie, history of dementia) with no suggestion of specific etiology of the acute cognitive impairment, the etiology was classified as descriptive/ unknown. Chi-square tests for independence of association and Fisher exact tests were used in comparing differences in characteristics between nursing home and community dwelling residents. All analyses were performed using Epi-Info (Atlanta, GA).
TABLE 1. Comparison of Study Population with All EMS Patients
Age Mean (+ISD) Range Gender--female %
(n) Race--white % (n) Nursing home % (n) Hospital distribution % (n)
After Exclusions (n = 153)
76.4 (16.1) 60-103
78.2 (9.5) 60-101
78.7 (10.1) 60-101
61.1 (2,865) 74.1 (3,473) 14.3 (671)
67.8 (154) 72.6 (165) 29.9 (68)
66.0 (101) 65.4 (100) 28.8 (44)
74.9 (3,488)
78.8 (179)
74.3 (113)
not available (n = 49) or ED records showed that the patient had markedly different diagnoses than those given by EMS (n = 25). The remaining 153 patients whose records were subjected to further chart review were less often of white race than those excluded from the final analysis, but there were no meaningful differences in mean age, gender, nursing home status, or hospital distribution (Table 1). Special EMS action beyond oxygen therapy and intravenous line placement occurred in 3.3% (5 of 153) of transports for acute cognitive impairment. Four patients received intravenous glucose administration, one patient intravenous atropine, and one intravenous lidocaine. Table 2 shows the study population to have a substantial proportion of known chronic psychoneurologic impairment and psychoactive medication use. Fifty-three patients (34.6%) were receiving at least one psychoactive medication (benzodiazepine, neuroleptic, or antidepressant). Nursing home patients were more often on neuroleptic medications than TABLE 2, Characteristics of Patients With Acute Cognitive
Impairment by Residence Type
RESULTS From a total of 5,999 EMS documented transports for elderly patients during calendar year 1990, 1,311 transports were excluded on the basis of location outside Forsyth County ( n - - 4 1 9 ) , destination other than an acute care hospital (n = 823), convalescent nature of the transport (n = 40), and missing demographic data (n = 29). After the above exclusions, a total of 4,688 patients remained for analysis. Records of the 393 patients transported with EMS diagnoses of stroke (n = 221), seizure (n = 81), and syncope (n = 91) were not subjected to further chart review. Table 1 compares the demographics of the 227 patients who were transported for evaluation of acute cognitive impairment with those of all elderly patients transported during calendar year 1990. Transports for acute cognitive impairment were more often from the nursing home than transports for all diagnoses, but there were no meaningful differences in gender, race, mean age, or hospital distribution. Of the 227 patients with reasons for transport suggestive of acute cognitive impairment, 74 patients were excluded from further analysis because complete records were
All EMS Patients (n = 4,688)
Patients With Acute Cognitive Impairment (n = 227)
Age Gender--°/° female Race--% white Chronic psychoneurologic impairment--% (n) Seizure disorder Dementia* Parkinson's Chronic use of psychoactive medication-% (n) Benzodiazepine Neuroleptic
Antidepressant Hospitalized
Total Patients (n = 153)
Nursing Home Patients (n = 47)
CommunityDwelling Patients (n = 105)
78.7 (10.1) 66.0 (101 ) 64.4 (100)
82.7 (7.6) 72.3 (34) 74.5 (35)
76.8 (10.2) 61.9 (65) 59.0 (62)
6.5 (10) 19.6 (30) 1.3 (2)
42.6 (12)l 42.6 (20)14.3 (2)
13.7 (21) 14.3 (22) 6.5 (10) 74.3 (110)
2.1 19.1 4.3 76.6
3.8 (4)
(1)1(9) (2) (36)
9.5 (10) 0 (0)
17.1 (18) 12.4 (13) 7.6 (8) 70.4 (74)
*Dementia = Alzheimer's or multi-infarct dementia, organic brain syndrome. I P < .05, Chi-square for comparison of nursing home and communitydwelling patients.
WOFFORD ET AL • ACUTE COGNITIVE IMPAIRMENTS IN ELDERLY ED PATIENTS
651
community-dwelling patients (19.2% v 12.4%), whereas community dwellers were more often on benzodiazepines (17.1% v 2.1%). Patients with acute cognitive impairment were hospitalized at a rate of 74.3% (110 of 153) with little difference in rates between nursing home and community-dwelling patients (Table 2). With regard to the disposition of hospitalized patients, 34.0% (37) were discharged to the community, 35.1% (38) were discharged to a nursing home, 28.9% (32) died, and 2.1% (3) were transferred to another hospital. Among patients who were hospitalized, the mean length of stay of 28 days (_+ 5). Infection, metabolic/toxic, and cerebrovascular were the most common categories of final ED diagnoses (Table 3). However, a substantial number of ED diagnoses were unknown or had purely descriptive causes. For nursing home patients, the ED diagnosis of infection as the cause of cognitive impairment was three times as common as that among community-dwelling patients (42.5% v 13.3%), whereas the diagnosis of cerebrovascular disease was more common among community dwellers (22.9% v 10.6%). A comparison of ED diagnoses with final hospital diagnoses found agreement within major etiologic categories in 75.6% (62 of 82) of patients for whom diagnoses were comparable. There was major disagreement in diagnosis categories in the case of 11 patients (Table 4). In the case of 9 patients the absence of a specific etiology from the written ED diagnosis prevented comparison of the two diagnoses.
TABLE4. Comparison of Final ED Diagnosis and Hospital
Acute brain syndrome and ataxia secondary to CVA
Acute confusion state, metabolic versus toxic
DISCUSSION
Urosepsis, COPD with 002
Mental status changes secondary to hypoxia
The syndrome of acute cognitive impairment accounts for over 500,000 ED visits by geriatric patients each year in the United States. 11 As familiar as the syndrome is to ED providers, there is still much uncertainty about its evaluation and management. In a recent survey of ED residency program directors, the majority reported that the evaluation and management of acute mental status changes was more difficult for older than younger patients.1 At least part of this difficulty is due to the paucity of systematic studies on the subject. The three published studies on cognitive impairment in the ED showed that 10% to 40% of elderly patients, TABLE3. Causes of Acute Cognitive Impairment By Residence Type--% (n)
Infection Metabolic/toxic Cerebrovascular Unknown cause or descriptive Other Trauma Cardiac Medication related Syncope Psychiatric
Total
Nursing Home Patients With ACI (n = 47)
CommunityDwelling Patients With ACI (n = 105)
26.1 (40) 22.9 (35) 20.2 (31)
42.5 (20) 17.0 (8) 10.6 (5)
13.3 (14) 19.0 (20) 22.9 (24)
15.6 (24) 7.2 (11) 6.5 (10) 5,9 (9) 4.6 (7) 4.6 (7) 1.3 (2)
8.5 (4) 6.4 (3) 0 (0) 2.1 (1) 4.3 (2) 2.1 (1) 2.1 (1)
17.8 (18) 7.6 (8) 7.6 (8) 4.8 (5) 3.8 (4) 2.9 (3) 0.1 (1)
Diagnosis for Discrepant Diagnoses Final ED Diagnosis
Final Hospital Diagnosis
Acute seizure secondary to noncompliance
Digoxin toxicity
CVA
Psychotic depression, febrile
illness CVA
Dementia
Combative behavior, rule out drug toxicity
Mental status changes secondary to hypoxia
Dilantin toxicity
Hypothermia, Sick euthyroid v central hypothyroid
Urinary retention, DM out of control
Urinary retention secondary to amitriptyline, probable new CVA
Acute CVA, Acute CHF
Probable CHF
Bilobar pneumonia
New CVA
Rule out sepsis
Left atrial thrombus, delirium secondary to emboli
retention
ABBREVIATIONS: CVA, cerebrovascular accident; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; UTI, urinary tract infection.
regardless of their reason for coming to the ED, had significant cognitive impairment and that standardized mental status testing identified those at high-risk. 12-14Our study provides a different perspective on this syndrome by examining not all elderly patients, but those who present specifically for evaluation of cognitive impairment. The causes of cognitive impairment in this study were as myriad as has been previously suggested. 4,5,1° The high prevalence of infection as the etiology for acute cognitive impairment among nursing home patients is consistent with previous studies that show a high rate of infection among nursing home transfers. ~5,16In contrast, the high prevalence of cerehrovascular causes as the etiology of acute cognitive impairment is somewhat surprising. In general, acute cerebrovascular disease is common among elderly ED patients. However, previous studies have shown that stroke is rarely a cause of the acute confusional state. ~7,18 Especially since patients with clear symptoms of stroke were excluded from this study, this finding suggests that the prevalence of acute cerebrovascular disease as a cause of cognitive impairment may be overestimated by ED providers. The high prevalence of elderly patients with previous chronic cognitive impairment or who were on psychoactive medications in this study is noteworthy but not surprising. Underlying dementia is one of the strongest and most
652
AMERICAN JOURNAL OF EMERGENCY MEDICINE • Volume 14, Number 7 • November 1996
consistent risk factors for delirium. 7 Likewise, medications represent a common cause of confusional states, and polypharmacy can significantly increase that risk. 19 Determining whether psychoactive medication for elderly patients is appropriate for the treatment of cognitive impairment or is a factor in causing cognitive impairment is a common clinical dilemma. 20-22 The hospitalization rate of 74.3% for this cohort of ED patients is relatively high compared with general studies of elderly ED patients. 23,24 Naughton et a112 showed that the risk of hospitalization was associated with the presence of cognitive impairment. In contrast, Gerson et a113showed that cognitive impairment did not affect the chance of being admitted to the hospital. Although the relationship between the level of cognitive impairment and the rate of subsequent hospitalization is unclear for elderly patients who may be in the ED for reasons other than cognitive impairment, the high rate of hospitalization in this study of patients transported specifically for evaluation of cognitive impairment is in agreement with the recommendations that delirious geriatric patients be hospitalized for observation. 25,26 The mortality rate of 28.9% in this study, consistent with most previous estimates, further emphasizes the danger of this syndrome. 10,27 Conclusions from past research studies of acute confusional states have varied greatly because of different diagnostic criteria, methods, and settings. 2s Recent prospective studies of delirium among hospitalized geriatric patients have identified risk factors (vision impairment, severe illness, underlying cognitive impairment, decreased renal function, abnormal sodium, fever, psychoactive drug use) that may be useful for future research into this ED syndrome. 7,s However, isolating cognitive impairment as the target condition in the ED setting may be the bigger problem for this area of research. As an example, Lowenstein et a129 showed that mental status change was the documented reason for presentation in 6.8% of elderly ED patients. However, an additional 18.6% of patients had falls, dehydration, or failure of self care as the reason for ED presentation, all syndromes that are commonly associated with acute cognitive impairment. Future research should consider that (1) acute cognitive impairment may present in ways that often obscure the main cause of illness, and (2) determining the chronology of psychoneurologic events among elderly patients can be difficult. This variability in presentation of cognitive impairment emphasizes the need for prospective evaluation of mental status in all elderly ED patients. Several cautions are in order in the interpretation of these findings. First, the uniqueness of our study population cautions against its generalizability. Although identification through computerized EMS records offers the advantage of an independent judgement by EMS personnel regarding the patient's presenting symptoms, it does not identify all patients who presented to EDs with acute cognitive impairment during this time period. In our opinion, this disadvantage is outweighed by the fact that this method identifies a cohort of patients with acute cognitive impairment as the major presenting symptom before ED provider documentation is influenced by other clinical findings. A second difference in our study population is the inclusion of all forms of cognitive impairment that were not clearly related
to stroke, seizure, or syncope. Because any cause of mild cognitive impairment can lead to more severe impairment, the entire spectrum of cognitive impairment was included in our study. Since our choice of populations has certain unique features, replication in other settings is warranted. Other caveats in interpreting our findings include our dependence on provider documentation and the difficulty in determining a definite diagnosis in every case. With the projected aging of the US population and the increasing interest in cognitive impairment, a better understanding of this syndrome should be important to ED providers. Although our study offers insights into acute cognitive impairment in the elderly patient, more and better information is needed to overcome the difficulties in management of this common ED syndrome.
REFERENCES 1. Jones JS, Rousseau EW, Schropp MA, Sanders AB: Geriatric training in emergency medicine residency programs. Ann Emerg Med 1992;21:825-829 2. Jorm AF, Korten AE, Henderson AS: The prevalence of dementia: A quantitative integration of the literature. Acta Psychiatr Scand 1987;76:465-479 3. Preston GAN: Dementia in older adults: Prevalence and institutionalization. J Gerontol 1986;41:261-267 4. Francis J: Delirium in older patients. J Am Geriatr Soc 1992;40: 829-838 5. Purdie FR, Honigman B, Rosen P: Acute organic brain syndrome: A review of 100 cases. Ann Emerg Med 1981;10:455-461 6. Fields SD, MacKenzie R, Charlson ME, Perry SW: Reversibility of cognitive impairment in medical inpatients. Arch Intern Med 1986; 146:1593-1596 7. Francis J, Martin D, Kapoor WN. A prospective study of delirium in hospitalized elderly. JAMA 1990;263:1097-1101 8. Inouye Sk, Viscoli CM, Horwitz RI, et al: A predictive model for delirium in hospitalized elderly medical patients based on admission characteristics. Ann Intern Med 1993; 119:474-481 9. Pousada L: Common neurologic emergencies in the elderly population. Clin Geriatr Med 1993;9:577-589 10. Lipowski Z J: Delirium in the elderly patient. N Engl J Med 1989;320:578-582 11. Strange GR, Chan EH, Sanders AB: Use of emergency departments by elderly patients: Projections from a multicenter data base. Ann Emerg Med 1992;21:819-824 12. Naughton BJ, Moran MB, Kadah H, et al: Delirium and other cognitive impairment in older adults in an emergency department. Ann Emerg Med 1995;25:751-755 13. Gerson LW, Counsell SR, Fontanarosa PB, Smucker WD: Case finding for cognitive impairment in elderly emergency department patients. Ann Emerg Med 1994;23:813-817 14. Lewis LM, Miller DK, Morley JE, et al: Unrecognized delirium in ED geriatric patients. Am J Emerg Med 1995; 13:142-145 15. Tresch DD, Simpson WM, Burton JR: Relationship of Iongterm and acute care facilities: The problem of patient transfer and continuity of care. J Am Geriatr Soc 1985;33:819-826 16. irvine PW, Van Buren N, Crossley K: Causes of hospitalization of nursing home residents: The role of infection. J Am Geriatr Soc 1984;32:103-107 17. Norris JW, Hachinski JW: Misdiagnosis of stroke. Lancet 1982;1:328-331 18. Bernabis AR, Sila CA, Cristea RL: Mental status changes and stroke. J Gen Intern Med 1994; 19:485-487 19. Bowen JD, Larson EB: Drug-induced cognitive impairment: Defining the problem and finding solutions. Drugs Aging 1993;3:349357 20. Semla TP, Palla K, Poddig B, Brauner DJ: Effect of the Omnibus Reconciliation Act 1987 on antipsychotic prescribing in nursing home residents. J Am Geriatr Soc 1994;42:648-652
WOFFORD ET AL • ACUTE COGNITIVE IMPAIRMENTS IN ELDERLY ED PATIENTS
21. Wilcox SM, Himmelstein DU, Woolhandler S: Inappropriate drug prescribing for the community-dwelling elderly. JAMA 1994;272: 292-296 22. Semla TP, Cohen D, Pavreza G, et al: Drug use patterns of persons with Alzheimer's Disease and related disorders living in the community. J Am Geriatr Soc 1993;41:408-13 23. Baum SA, Rubenstein LZ: Old people in the emergency department: Age-related differences in emergency department use and care. J Am Geriatr Soc 1987;35:398-404 24. Ettinger WH, Casani JA, Coon PJ, et al: Patterns of use of the emergency department by elderly patients. J Geronto11987;42:638642
653
25. Tueth MJ: Diagnosing psychiatric emergencies in the elderly. Am J Emerg Med 1994;12:364-369 26. Sanders AB: Recognition of cognitive problems in older adults by emergency medicine personnel. Ann Emerg Med 1995;25:831833 27. Rabins PV, Folstein MF: Delirium and dementia: Diagnostic criteria and fatality rates. Br J Psychiatry 1982;140:149-153 28. Stewart RB, Hale WE: Acute confusional states in older adults and the role of polypharmacy. Ann Rev Health 1992; 13:415-430 29. Lowenstein SR, Creszenzi, Kem DC, Steel K: Care of the elderly in the emergency department. Ann Emerg Med 1986;15:528535
cognitive impairment in hospitalized patients, 7,8 few studies of cognitive impairment in the ED setting are available. A major problem with research efforts on this subject is that of defining the target syndrome. The nomenclature of cognitive impairment is confusing and variable. 9,m Furthermore, objective quantitative measures of cognitive impairment are not often used in the ED setting. As a result, clinicians fail to recognize and document the presence of cognitive impairment. Tracing clinical decisions becomes difficult because written acknowledgement of the syndrome is quickly displaced by diagnoses that are more familiar. The purpose of this study was to identify and follow a cohort of elderly ED patients with acute cognitive impairment as the sole reason for ED presentation. Examination of the subsequent ED and hospital course offers data on the etiology, disposition, and outcome of this syndrome.
METHODS The evaluation and treatment of acute cognitive impairment in the geriatric patient is a task that commonly confronts providers in the hospital emergency department (ED). Management of this syndrome is acknowledged to be more difficult for elderly than for younger patients, ~ for several reasons. First, chronic cognitive impairment is common among elderly p e r s o n s Y Discerning whether there is an acute mental status change in a chronically demented patient is difficult, especially when the patient is not familiar to the provider. Second, cognitive impairment among elderly patients frequently results from acute illnesses unrelated to the central nervous system. 4-6 Possible etiologies include infection, medication side effects, and cardiovascular disorders, all of which are c o m m o n among elderly patients. A third reason for the difficulty in management o f acute cognitive impairment in the elderly ED patient is that research into this common geriatric syndrome is not common. While a substantial amount of research has focused on
From the *Department of Internal Medicine, and SDepartment of Emergency Medicine, Bowman Gray School of Medicine of Wake Forest University, Winston-Salem, NC, and the #Department of Internal Medicine, St. Louis University, St. Louis, MO. Manuscript received July 10, 1995, returned September 5, 1995; revision received November 6, 1995, accepted November 16, 1995. Presented at the annual meeting of the American Geriatrics Society, May 1995. Address reprint requests to Dr Wofford, Department of Internal Medicine, Bowman Gray School of Medicine of Wake Forest University, Medical Center BIvd, Winston-Salem, NC 27157. Key Words: Aging, emergency services, cognitive disorder, delirium, geriatrics. Copyright © 1996 by W.B. Saunders Company 0735-6757/96/1407-000855.00/0
Patients were identified using computerized data on EMS transports from a semi-urban North Carolina county, Forsyth County, during the year 1990. Data were available for all patients 60 years of age and older who were transported by the Forsyth County EMS from January 1, 1990, to December 31, 1990. These data included information on patient age, race, gender, location of residence, reason for transport, and destination, all of which were determined by EMS personnel at the time of transport. Data were restricted to nonconvalescent transports to one of two full-service hospitals, one 700-bed tertiary care teaching hospital with a level 1 trauma center and one 1,000-bed community, primarily nonacademic hospital. These two hospitals, the only ones in the county that provide ED services, are both staffed by attending and resident physicians. Reasons for transport recorded at the time of transport by EMS personnel were coded into one of 45 categories, in accordance with requirements of the state EMS supervisory office. Each transport was coded as a single diagnosis. Piloting of the data collection form revealed that for patients with the EMS diagnosis of stroke, seizure, or syncope, EMS diagnoses were consistent to a large degree with the final ED and hospital diagnosis. Because assigning a diagnosis to these patients seemed clear even before ED evaluation, patients with these diagnoses were not considered eligible for this study. For patients with no specific reason for transport according to EMS data (n = 1,086), EMS encounter forms were manually reviewed by two authors (JLW, LRL) to identify patients transported to EDs specifically for the evaluation and management of acute cognitive impairment. EMS documentation denoting decreased level of consciousness or confusion without other obvious problems identified patients who were transported specifically for evaluation of acute cognitive impairment and who were considered eligible for this study. Patients were excluded from the study if ED or hospital records were not available or complete. Patient characteristics abstracted from EMS and ED records included medical history, medication history, final ED diagnosis, and disposition. Particular attention was given to previous diag649
650
AMERICAN JOURNAL OF EMERGENCY MEDICINE • Volume 14, Number 7 • November 1996
noses and medications that reflected underlying cognitive impairment (ie, chronic dementia, organic brain syndrome, multi-infarct dementia). Details of the EMS management focused on medical therapies beyond intravenous line placement or oxygen administration. In the case of hospitalization, information on the hospital course, final hospital diagnosis, length of stay, and disposition was abstracted from the hospital records. Because information on follow-up was not reliably available in the hospital chart for patients discharged from the ED, we did not attempt to collect follow-up data on patients who were not hospitalized. Because nursing home and community-dwelling patients are populations known to have important differences, these two groups were compared with respect to demographics, history of chronic cognitive impairment, psychoactive medication use, hospitalization, and mortality. Based on the final written ED diagnosis, the etiology of the mental status change was categorized as either cerebrovascular, cardiac, infection, metabolic/toxic effect, medication-related, trauma, psychiatric, simple syncope, other causes, or unknown/descriptive. Since the patient's final ED diagnosis frequently reflected uncertainty on the part of the ED provider and the potential for multiple etiologies, a patient may have been placed into more than one category of etiology. Diagnoses classified as cerebrovascular included the written ED diagnosis of intracranial bleeding, cerebrovascular accident, or possible transient ischemic attacks. Cardiac causes included those due to congestive heart failure, cardiac dysrhythmia, or myocardial infarction. The diagnoses of pneumonia, urinary tract infection, sepsis, and "rule out sepsis" were included in the category of infection. The category "toxic/ metabolic" included the diagnoses of acute alcohol ingestion and electrolyte disturbances, while medication-related diagnoses included those thought secondary to prescribed medications (whether related to withdrawal or overdose). When the final ED diagnosis either reflected simply a symptom complex (ie, acute mental status changes) or repeated the patient's medical history (ie, history of dementia) with no suggestion of specific etiology of the acute cognitive impairment, the etiology was classified as descriptive/ unknown. Chi-square tests for independence of association and Fisher exact tests were used in comparing differences in characteristics between nursing home and community dwelling residents. All analyses were performed using Epi-Info (Atlanta, GA).
TABLE 1. Comparison of Study Population with All EMS Patients
Age Mean (+ISD) Range Gender--female %
(n) Race--white % (n) Nursing home % (n) Hospital distribution % (n)
After Exclusions (n = 153)
76.4 (16.1) 60-103
78.2 (9.5) 60-101
78.7 (10.1) 60-101
61.1 (2,865) 74.1 (3,473) 14.3 (671)
67.8 (154) 72.6 (165) 29.9 (68)
66.0 (101) 65.4 (100) 28.8 (44)
74.9 (3,488)
78.8 (179)
74.3 (113)
not available (n = 49) or ED records showed that the patient had markedly different diagnoses than those given by EMS (n = 25). The remaining 153 patients whose records were subjected to further chart review were less often of white race than those excluded from the final analysis, but there were no meaningful differences in mean age, gender, nursing home status, or hospital distribution (Table 1). Special EMS action beyond oxygen therapy and intravenous line placement occurred in 3.3% (5 of 153) of transports for acute cognitive impairment. Four patients received intravenous glucose administration, one patient intravenous atropine, and one intravenous lidocaine. Table 2 shows the study population to have a substantial proportion of known chronic psychoneurologic impairment and psychoactive medication use. Fifty-three patients (34.6%) were receiving at least one psychoactive medication (benzodiazepine, neuroleptic, or antidepressant). Nursing home patients were more often on neuroleptic medications than TABLE 2, Characteristics of Patients With Acute Cognitive
Impairment by Residence Type
RESULTS From a total of 5,999 EMS documented transports for elderly patients during calendar year 1990, 1,311 transports were excluded on the basis of location outside Forsyth County ( n - - 4 1 9 ) , destination other than an acute care hospital (n = 823), convalescent nature of the transport (n = 40), and missing demographic data (n = 29). After the above exclusions, a total of 4,688 patients remained for analysis. Records of the 393 patients transported with EMS diagnoses of stroke (n = 221), seizure (n = 81), and syncope (n = 91) were not subjected to further chart review. Table 1 compares the demographics of the 227 patients who were transported for evaluation of acute cognitive impairment with those of all elderly patients transported during calendar year 1990. Transports for acute cognitive impairment were more often from the nursing home than transports for all diagnoses, but there were no meaningful differences in gender, race, mean age, or hospital distribution. Of the 227 patients with reasons for transport suggestive of acute cognitive impairment, 74 patients were excluded from further analysis because complete records were
All EMS Patients (n = 4,688)
Patients With Acute Cognitive Impairment (n = 227)
Age Gender--°/° female Race--% white Chronic psychoneurologic impairment--% (n) Seizure disorder Dementia* Parkinson's Chronic use of psychoactive medication-% (n) Benzodiazepine Neuroleptic
Antidepressant Hospitalized
Total Patients (n = 153)
Nursing Home Patients (n = 47)
CommunityDwelling Patients (n = 105)
78.7 (10.1) 66.0 (101 ) 64.4 (100)
82.7 (7.6) 72.3 (34) 74.5 (35)
76.8 (10.2) 61.9 (65) 59.0 (62)
6.5 (10) 19.6 (30) 1.3 (2)
42.6 (12)l 42.6 (20)14.3 (2)
13.7 (21) 14.3 (22) 6.5 (10) 74.3 (110)
2.1 19.1 4.3 76.6
3.8 (4)
(1)1(9) (2) (36)
9.5 (10) 0 (0)
17.1 (18) 12.4 (13) 7.6 (8) 70.4 (74)
*Dementia = Alzheimer's or multi-infarct dementia, organic brain syndrome. I P < .05, Chi-square for comparison of nursing home and communitydwelling patients.
WOFFORD ET AL • ACUTE COGNITIVE IMPAIRMENTS IN ELDERLY ED PATIENTS
651
community-dwelling patients (19.2% v 12.4%), whereas community dwellers were more often on benzodiazepines (17.1% v 2.1%). Patients with acute cognitive impairment were hospitalized at a rate of 74.3% (110 of 153) with little difference in rates between nursing home and community-dwelling patients (Table 2). With regard to the disposition of hospitalized patients, 34.0% (37) were discharged to the community, 35.1% (38) were discharged to a nursing home, 28.9% (32) died, and 2.1% (3) were transferred to another hospital. Among patients who were hospitalized, the mean length of stay of 28 days (_+ 5). Infection, metabolic/toxic, and cerebrovascular were the most common categories of final ED diagnoses (Table 3). However, a substantial number of ED diagnoses were unknown or had purely descriptive causes. For nursing home patients, the ED diagnosis of infection as the cause of cognitive impairment was three times as common as that among community-dwelling patients (42.5% v 13.3%), whereas the diagnosis of cerebrovascular disease was more common among community dwellers (22.9% v 10.6%). A comparison of ED diagnoses with final hospital diagnoses found agreement within major etiologic categories in 75.6% (62 of 82) of patients for whom diagnoses were comparable. There was major disagreement in diagnosis categories in the case of 11 patients (Table 4). In the case of 9 patients the absence of a specific etiology from the written ED diagnosis prevented comparison of the two diagnoses.
TABLE4. Comparison of Final ED Diagnosis and Hospital
Acute brain syndrome and ataxia secondary to CVA
Acute confusion state, metabolic versus toxic
DISCUSSION
Urosepsis, COPD with 002
Mental status changes secondary to hypoxia
The syndrome of acute cognitive impairment accounts for over 500,000 ED visits by geriatric patients each year in the United States. 11 As familiar as the syndrome is to ED providers, there is still much uncertainty about its evaluation and management. In a recent survey of ED residency program directors, the majority reported that the evaluation and management of acute mental status changes was more difficult for older than younger patients.1 At least part of this difficulty is due to the paucity of systematic studies on the subject. The three published studies on cognitive impairment in the ED showed that 10% to 40% of elderly patients, TABLE3. Causes of Acute Cognitive Impairment By Residence Type--% (n)
Infection Metabolic/toxic Cerebrovascular Unknown cause or descriptive Other Trauma Cardiac Medication related Syncope Psychiatric
Total
Nursing Home Patients With ACI (n = 47)
CommunityDwelling Patients With ACI (n = 105)
26.1 (40) 22.9 (35) 20.2 (31)
42.5 (20) 17.0 (8) 10.6 (5)
13.3 (14) 19.0 (20) 22.9 (24)
15.6 (24) 7.2 (11) 6.5 (10) 5,9 (9) 4.6 (7) 4.6 (7) 1.3 (2)
8.5 (4) 6.4 (3) 0 (0) 2.1 (1) 4.3 (2) 2.1 (1) 2.1 (1)
17.8 (18) 7.6 (8) 7.6 (8) 4.8 (5) 3.8 (4) 2.9 (3) 0.1 (1)
Diagnosis for Discrepant Diagnoses Final ED Diagnosis
Final Hospital Diagnosis
Acute seizure secondary to noncompliance
Digoxin toxicity
CVA
Psychotic depression, febrile
illness CVA
Dementia
Combative behavior, rule out drug toxicity
Mental status changes secondary to hypoxia
Dilantin toxicity
Hypothermia, Sick euthyroid v central hypothyroid
Urinary retention, DM out of control
Urinary retention secondary to amitriptyline, probable new CVA
Acute CVA, Acute CHF
Probable CHF
Bilobar pneumonia
New CVA
Rule out sepsis
Left atrial thrombus, delirium secondary to emboli
retention
ABBREVIATIONS: CVA, cerebrovascular accident; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; UTI, urinary tract infection.
regardless of their reason for coming to the ED, had significant cognitive impairment and that standardized mental status testing identified those at high-risk. 12-14Our study provides a different perspective on this syndrome by examining not all elderly patients, but those who present specifically for evaluation of cognitive impairment. The causes of cognitive impairment in this study were as myriad as has been previously suggested. 4,5,1° The high prevalence of infection as the etiology for acute cognitive impairment among nursing home patients is consistent with previous studies that show a high rate of infection among nursing home transfers. ~5,16In contrast, the high prevalence of cerehrovascular causes as the etiology of acute cognitive impairment is somewhat surprising. In general, acute cerebrovascular disease is common among elderly ED patients. However, previous studies have shown that stroke is rarely a cause of the acute confusional state. ~7,18 Especially since patients with clear symptoms of stroke were excluded from this study, this finding suggests that the prevalence of acute cerebrovascular disease as a cause of cognitive impairment may be overestimated by ED providers. The high prevalence of elderly patients with previous chronic cognitive impairment or who were on psychoactive medications in this study is noteworthy but not surprising. Underlying dementia is one of the strongest and most
652
AMERICAN JOURNAL OF EMERGENCY MEDICINE • Volume 14, Number 7 • November 1996
consistent risk factors for delirium. 7 Likewise, medications represent a common cause of confusional states, and polypharmacy can significantly increase that risk. 19 Determining whether psychoactive medication for elderly patients is appropriate for the treatment of cognitive impairment or is a factor in causing cognitive impairment is a common clinical dilemma. 20-22 The hospitalization rate of 74.3% for this cohort of ED patients is relatively high compared with general studies of elderly ED patients. 23,24 Naughton et a112 showed that the risk of hospitalization was associated with the presence of cognitive impairment. In contrast, Gerson et a113showed that cognitive impairment did not affect the chance of being admitted to the hospital. Although the relationship between the level of cognitive impairment and the rate of subsequent hospitalization is unclear for elderly patients who may be in the ED for reasons other than cognitive impairment, the high rate of hospitalization in this study of patients transported specifically for evaluation of cognitive impairment is in agreement with the recommendations that delirious geriatric patients be hospitalized for observation. 25,26 The mortality rate of 28.9% in this study, consistent with most previous estimates, further emphasizes the danger of this syndrome. 10,27 Conclusions from past research studies of acute confusional states have varied greatly because of different diagnostic criteria, methods, and settings. 2s Recent prospective studies of delirium among hospitalized geriatric patients have identified risk factors (vision impairment, severe illness, underlying cognitive impairment, decreased renal function, abnormal sodium, fever, psychoactive drug use) that may be useful for future research into this ED syndrome. 7,s However, isolating cognitive impairment as the target condition in the ED setting may be the bigger problem for this area of research. As an example, Lowenstein et a129 showed that mental status change was the documented reason for presentation in 6.8% of elderly ED patients. However, an additional 18.6% of patients had falls, dehydration, or failure of self care as the reason for ED presentation, all syndromes that are commonly associated with acute cognitive impairment. Future research should consider that (1) acute cognitive impairment may present in ways that often obscure the main cause of illness, and (2) determining the chronology of psychoneurologic events among elderly patients can be difficult. This variability in presentation of cognitive impairment emphasizes the need for prospective evaluation of mental status in all elderly ED patients. Several cautions are in order in the interpretation of these findings. First, the uniqueness of our study population cautions against its generalizability. Although identification through computerized EMS records offers the advantage of an independent judgement by EMS personnel regarding the patient's presenting symptoms, it does not identify all patients who presented to EDs with acute cognitive impairment during this time period. In our opinion, this disadvantage is outweighed by the fact that this method identifies a cohort of patients with acute cognitive impairment as the major presenting symptom before ED provider documentation is influenced by other clinical findings. A second difference in our study population is the inclusion of all forms of cognitive impairment that were not clearly related
to stroke, seizure, or syncope. Because any cause of mild cognitive impairment can lead to more severe impairment, the entire spectrum of cognitive impairment was included in our study. Since our choice of populations has certain unique features, replication in other settings is warranted. Other caveats in interpreting our findings include our dependence on provider documentation and the difficulty in determining a definite diagnosis in every case. With the projected aging of the US population and the increasing interest in cognitive impairment, a better understanding of this syndrome should be important to ED providers. Although our study offers insights into acute cognitive impairment in the elderly patient, more and better information is needed to overcome the difficulties in management of this common ED syndrome.
REFERENCES 1. Jones JS, Rousseau EW, Schropp MA, Sanders AB: Geriatric training in emergency medicine residency programs. Ann Emerg Med 1992;21:825-829 2. Jorm AF, Korten AE, Henderson AS: The prevalence of dementia: A quantitative integration of the literature. Acta Psychiatr Scand 1987;76:465-479 3. Preston GAN: Dementia in older adults: Prevalence and institutionalization. J Gerontol 1986;41:261-267 4. Francis J: Delirium in older patients. J Am Geriatr Soc 1992;40: 829-838 5. Purdie FR, Honigman B, Rosen P: Acute organic brain syndrome: A review of 100 cases. Ann Emerg Med 1981;10:455-461 6. Fields SD, MacKenzie R, Charlson ME, Perry SW: Reversibility of cognitive impairment in medical inpatients. Arch Intern Med 1986; 146:1593-1596 7. Francis J, Martin D, Kapoor WN. A prospective study of delirium in hospitalized elderly. JAMA 1990;263:1097-1101 8. Inouye Sk, Viscoli CM, Horwitz RI, et al: A predictive model for delirium in hospitalized elderly medical patients based on admission characteristics. Ann Intern Med 1993; 119:474-481 9. Pousada L: Common neurologic emergencies in the elderly population. Clin Geriatr Med 1993;9:577-589 10. Lipowski Z J: Delirium in the elderly patient. N Engl J Med 1989;320:578-582 11. Strange GR, Chan EH, Sanders AB: Use of emergency departments by elderly patients: Projections from a multicenter data base. Ann Emerg Med 1992;21:819-824 12. Naughton BJ, Moran MB, Kadah H, et al: Delirium and other cognitive impairment in older adults in an emergency department. Ann Emerg Med 1995;25:751-755 13. Gerson LW, Counsell SR, Fontanarosa PB, Smucker WD: Case finding for cognitive impairment in elderly emergency department patients. Ann Emerg Med 1994;23:813-817 14. Lewis LM, Miller DK, Morley JE, et al: Unrecognized delirium in ED geriatric patients. Am J Emerg Med 1995; 13:142-145 15. Tresch DD, Simpson WM, Burton JR: Relationship of Iongterm and acute care facilities: The problem of patient transfer and continuity of care. J Am Geriatr Soc 1985;33:819-826 16. irvine PW, Van Buren N, Crossley K: Causes of hospitalization of nursing home residents: The role of infection. J Am Geriatr Soc 1984;32:103-107 17. Norris JW, Hachinski JW: Misdiagnosis of stroke. Lancet 1982;1:328-331 18. Bernabis AR, Sila CA, Cristea RL: Mental status changes and stroke. J Gen Intern Med 1994; 19:485-487 19. Bowen JD, Larson EB: Drug-induced cognitive impairment: Defining the problem and finding solutions. Drugs Aging 1993;3:349357 20. Semla TP, Palla K, Poddig B, Brauner DJ: Effect of the Omnibus Reconciliation Act 1987 on antipsychotic prescribing in nursing home residents. J Am Geriatr Soc 1994;42:648-652
WOFFORD ET AL • ACUTE COGNITIVE IMPAIRMENTS IN ELDERLY ED PATIENTS
21. Wilcox SM, Himmelstein DU, Woolhandler S: Inappropriate drug prescribing for the community-dwelling elderly. JAMA 1994;272: 292-296 22. Semla TP, Cohen D, Pavreza G, et al: Drug use patterns of persons with Alzheimer's Disease and related disorders living in the community. J Am Geriatr Soc 1993;41:408-13 23. Baum SA, Rubenstein LZ: Old people in the emergency department: Age-related differences in emergency department use and care. J Am Geriatr Soc 1987;35:398-404 24. Ettinger WH, Casani JA, Coon PJ, et al: Patterns of use of the emergency department by elderly patients. J Geronto11987;42:638642
653
25. Tueth MJ: Diagnosing psychiatric emergencies in the elderly. Am J Emerg Med 1994;12:364-369 26. Sanders AB: Recognition of cognitive problems in older adults by emergency medicine personnel. Ann Emerg Med 1995;25:831833 27. Rabins PV, Folstein MF: Delirium and dementia: Diagnostic criteria and fatality rates. Br J Psychiatry 1982;140:149-153 28. Stewart RB, Hale WE: Acute confusional states in older adults and the role of polypharmacy. Ann Rev Health 1992; 13:415-430 29. Lowenstein SR, Creszenzi, Kem DC, Steel K: Care of the elderly in the emergency department. Ann Emerg Med 1986;15:528535