lnsermbnal Psychogeriatrics, Vol. 4, No.2.1992 Q 1992 Springer Publishing Company

The Effect of Dementia on Acute Care in a Geriatric Medical Unit Lucia Torian, Emily Davidson, George Fulop, Laura Sell, and Howard Fillit ABSTRACT. Treatmentof dementia costs billions of dollarsin long-term care and community services every year. Dementia also burdens the acute care system and may contribute to financial problems for hospitals serving large numbers of demented elderly. In a specialized geriatric medical unit devoted to acute care of the frail elderly, Alzheimer’s disease and vascular and mixed dementias afflicted 63% of inpatients and were associated with excess consumption of nursing resources, complicationsof treatment, nosocomial infections, lengthy hospitalizations,and financial losses to the hospital. Due in part to the effects of dementia on mobility, continence, and nutrition, demented patients suffered more frequently from life-threatening infections, sepsis, iatrogenic disease, and prolonged hospital stays. Hospital losses were 75% higher for demented patients than for nondemented patients. Dementia affected the majority of acute care patients in this study. However, it was rarely coded as an admitting diagnosis, even though it may have been the proximate cause of the medical morbidity which led to the acute hospitalization. In addition, despite the significant impact of dementia on the hospital course and costs, it was a factor in hospital reimbursementin less than one third of cases. The results indicate that dementia was not considered to be an acute diagnosis, nor was it recognized as a complex medical illness. The impact of dementia on acute hospitalization, including the mechanisms by which dementia prolongs the hospital stay, requires further investigation.

Alzheimer’s disease and vascular and mixed dementias are leading causes of morbidity and mortality in the elderly. Patients with dementia have shortened survival compared to theirnondemented peers. (Diesfeldt et al., 1986; Hassel et al., 1988; Jagger & Clarke, 1988;Li et al., 1991;Martinet al., 1987;Newman &Bland, From the RitterDepartmentof Geriatricsand Adult Development(L. Torian, PhD, E. Davidson, RPAC; H. Fillit, MD). the Department of Psychiatry (G. Fulop. MD), and the Department of Nursing (L. Sell, MSN), Mount SinaiMedical Center; and the Brookdale Centeron Aging, City Universityof New York, New York, U. S. A. (L. Torian, PhD). 23 1

232

L.Torian et al.

1987; Peck et al., 1978). Worldwide deaths attributable to dementia may be underestimated because diagnosis and death certification conventions vary from country to country (Centers for Disease Control, 1991; Flaten, 1989; Martyn & Pippard, 1988). Infections and sepsis are common immediate causes of death in patients withdementia(Chandraetal.,1990;Molsaetal., 1986;Zarianetal., 1989). The pathogenesisof infectionin demented patientsinvolves factorsthat are directly and indirectly related to the disease, such as feeding disorders leading to malnutrition and secondary immunodeficiency (Abraham, 1989; Jeejeebhoy, 1989; Lewis & Bell, 1990); aspiration predisposing to pneumonia (Fischer & Johnson, 1990; Volicer et al., 1989); urinary and fecal incontinence predisposing to urinary tract and wound infections (Schimpff et al., 1989); and immobility predisposing to skin breakdown, wound infections, and impaired ventilation and clearing of respiratory secretions (Hamer, 1989). The cost of caring for demented patients in hospitals, nursing homes, and the community has been estimated in the tens of billions of dollars (Hay & Ernest, 1987;Huang et al., 1988).The high cost of acute care of the demented results from the high cost of treating life-threatening infections and multiple interdependent comorbidities, the predisposition of the demented to nosocomial infections and complications of treatment, and difficulties in arranging for placement or community services which prolong the length of stay. Excess length of stay increases the risk of development of iatrogenic illness (Jahnigen et al., 1982; Patterson, 1986; Saviteer et al., 1988). Extensive literature documentsthe nature of long-term care of the demented. While it is generally accepted that dementia also creates excess burden on acute care resources, no studies have analyzed the effects on clinical outcome and cost of comorbidities,complicationsof treatment, and complications of hospitalization in patients with dementia.

METHOD We studied the nature of acute care for the demented frail elderly admitted to a geriatric unit (GETU)at a major metropolitan medical center. Sociodemographic, clinical, and discharge information for all patients admitted to the GETU in 1988 was recorded on a study protocol that was updated weekly so that new health events and nursing problems, changes in treatment preferences, and changes in mental status could be followed. Activities of Daily Living (ADL) were assessed using a modified scale (Katz, 1983) which included grooming and walking, with a high score indicating increased functional disability. A score of 0 for an individual parameter indicated functionalindependence with respect to the parameter, a score of 1 indicated that some assistance was required; a score of 2 indicated that the patient was completely dependent with mpect to the parameter. All patients were evaluated for dementia using DSM-III-R criteria (American Psychiatric Association, 1987). Patients with Alzheimer’s disease and vascular and mixed dementias were classified as demented. Acuity of illness and nursing workload were measured

Effect of Dementia on Acute Care

233

using the MEDICUS Patient Classificationsystem (Medicus SystemsCorporation, Evanston, Illinois). This factor evaluation tool classifies patients daily into one of five categories based on a nursing assessment of 37 critical indicators. The acuity score measures the relative intensity of nursing care needs. The score can then be translated into hours of direct nursing care required. The Patient Accounts and Medical Records databases were obtained from the respective departments. Data were analyzed on an IBM 3084 processor using the SPSS-X statistical package (SPSS, Inc., Chicago, Illinois). Statistical procedures included the two-tailed Student’s f -test, the chi-square test for contingencytables, and analysis of variance. Nonnormally and nonsymmetrically distributed continuous data were analyzed using the Wilcoxon rank sum test.

RESULTS Summary Data for All GETU Patients The GETU is a 12-bed acute-care unit devoted to treatment of the frail elderly in the East Harlem section of New York City. Data for the 143 patients who were admitted between January 1 and December 31, 1988, and for whom complete information was available, are reported here. The mean age of the patients was 82 years (range 62 to 101). Seventy-eight percent were female. Forty-onepercent were white, 33%black, 19% Hispanic, and 6% Asian. The majority had Medicare as their primary insurance. On admission, 70% of the patients had six acute complicatingdiagnoses, and 92% had four acute complicating diagnoses. Sixty-three percent of the patients had dementia. The most common reason in 1988 for acute admission to the GETU was an infectious disease, typically pneumonia or a urinary tract infection. The next most common diagnostic category was cardiovasculardisease. Patients had an average of two conditions suggesting poor prognosis. These included the following: dementia, renal failure, sepsis, metastatic cancer, acute stroke, congestive heart failure, history of previous cardiac or respiratory arrest,ventilator dependence,and oxacillin-resistant staphylococcus infection. The mean Activities of Daily Living (ADL) score was 8.5. Most patients were dependenton some assistancefor all eight basic activitiesmeasured on the 16-point scale. Twenty-seven percent were totally dependent upon assistance for all basic activities. Only 8% were functionally independent. Nearly 50% were incontinent of urine or urine and feces. Four patients required extended mechanical ventilation. Six patients had oxacillin-resistantstaphylococcusinfections;they remained in the hospital for periods extending to 168 days. Six patients, of whom five were demented, were resuscitated following a cardiac arrest; two survived to discharge. The survivors included one demented and the nondemented patient. The mean total length of stay was 26 days (SO = 29 days; range = 1-198 days), of which 18 were acute days and 8 were alternate level of care days, or “social

234

L.Torinn et al.

hospitalization”days.Patients were placed on alternate level of care when they no longer needed acute hospital care, but could not be discharged due to problems with placement or arranging for home support services. Seventy-one patients (45%) spent one or more days on alternatelevel of care. Eleven percent of all patients were placed on alternative level of care status for a period of time, but were reinstated to acute care when they experienced complications or acquired nosocomial infections. Three such patients cycled in and out of acute and alternate level of care status two and three times. Patients with two or more conditions suggesting poor prognosis spent significantly more acute and alternate level of care days in the hospital than patients with one or zero conditions, suggesting poor prognosis by ANOVA (f= 4.019, p = .024). Fifty percent of the patients were discharged home with no formal support services, 15%were discharged home with a home health aide, 26% were discharged to a nursing home, and 10%died.

Characteristics of the Demented Versus the Nondemented Patients Patients with dementia were about the same average age as patients without dementia. Gender distribution was approximately equivalent in the two groups. Patients with dementia differed significantly from patients without dementia on a variety of measures, however, such as reason for admission, level of functional impairment, acuity of illness, length of stay, cost of care, and discharge status. Patients with dementia were more frequently admitted for treatment of infectious disease than were patients without dementia (Table 1). Seventy percent of patients with dementia who were admitted for treatment of infectious disease were admitted for treatment of bronchopneumonia, of a urinary tract infection, or of sepsis that had originated in the lower respiratory or urinary tract. During their hospital stay, demented patients were also more likely to develop pneumonia as a complicating diagnosis and to become septic than were nondemented patients. All of the patients with oxacillin-resistant staphylococcus infections suffered from dementia. Patients with dementia also were more likely to have a history of respiratory or cardiac arrest than nondemented patients. Patients with dementia who died were most likely to have died of infection or sepsis. Patients with dementia were not more likely to die in the hospital than were patients without dementia; however, they were twice as likely to be discharged to a nursing home. Patients with dementia had a mean ADL score (mean score = 12) that was more than double the mean score of patients without dementia (Table 2). These functional impairments translated into needs for support in the performance of most personal activities during the hospitalization. In general GETU patients required resource-intensive nursing care, as reflected in an average MEDICUS acuity of illness class score of 2.5, which was higher than the average acuity score of patients in the coronary care unit of the hospital, where the mean score during the same period was 2.3. Patients with dementia, whose average acuity of illness score was 2.9, required significantly higher-intensitynursing care than did patients without dementia, whose average acuity of illness score was 2.1 (Table 2). Eighty-eight percent of the patients who were incontinent were demented; 13 of the 14 patients

235

Effect of Dementia on Acute Care

on artificial feeding were demented; 3 of the 4 patients who underwent mechanical ventilation were demented. Patients with dementia had a mean total length of stay that was twice that of patients without dementia-34 days as compared to 17 days (Table 3). Patients with dementia had more than twice as many acute days as patients without dementia, and they also had twice as many alternate level of care days as those without dementia. In addition, patients with dementia were more likely to cycle in and out of alternate level of care and acute status during their admission than were patients without dementia. Although the mean Medicare Diagnosis-Related Group (DRG) weight for patients with dementia was higher than that for patients without dementia, the difference was not statistically significant. In contrast, the mean length of stay for the admitting DRG for patients with dementia was significantly longer than that for the DGR of patients without dementia. In addition, patients with dementia had an actual length of stay that was more than four times as long as the mean length of stay their DRG would predict, while patients without dementia stayed about three times as long as the mean length of stay their DRGs would predict. Treatment of patients with dementia was about 75% more costly than treatment of patients without

TABLE 1. Admitting Diagnosis and Discharge Disposition of GETU Patients With and Without Dementia Patients without Dementia N % Admitting diagnosis Infectious disease Neoplastic disease Metabolic/endocrine Mental illness CNS disease Cardiovascular disease Pulmonary disease Digestive system disease Genitourinary disease Skin disease Musculoskeletal disease Signs and symptoms Injury and poisoning Totals Discharge Status Home self care SNF or HRF Home with home aide Left AMA Died Totals ~~

7 1 5 1

0 19 5 4 1 0 2 6 2 53 31 8 8 0 6 53

Patients with Dementia

N

%

13.2 1.9 9.4 1.9 0.0 35.8 9.4 7.5 1.9 0.0 3.8 11.3 3.8 99.9

31 4 6 6 1 15 4 5

41.1* 4.4 6.1 6.7 1.1 16.7 4.4 5.6 0.0 2.2 3.3 6.7

58.5 15.1 15.I 0.0 11.3

36 30 14 1 9 90

100.0

0

2 3 6 1

90

~

*Infectious disease vs. other admission Chi-square = 10.12, p = 0.015 **Discharge to nursing home vs. other discharge Chi-square = 5.23, p = 0.012.

1.1 1100.0

40.0 33.3** 15.6 1.1 10.0 100.0

L.Tor&

236

et d.

dementia (Table 3). Dementia was listed as the admitting diagnosis for only 3.3% of the demented patients, and was ultimatelycoded in some position on the Medical Records database of only 31% of the demented patients.

DISCUSSION The frail elderly in this study consumed high levels of medical resources, required high-intensity nursing services, and were costly to treat. The presence of dementia in nearly two-thirds of this population had a dramatic impact on the nature of medical care provided, the length of hospital stay, and hospital revenues. Nevertheless, dementia was rarely coded on the Medical Records database,possibly because it was not considered to be an acute diagnosis by physicians or groupers, and possibly because the demented patients in this series had sufficient acute principal

TABLE 2. Characteristics of Patients With and Without Dementia Patients without Dementia

Patients with Dementia

Mean (SD) 81.50 (7.86) 5.14 (4.81) 2.1 (.60)

Mean (SD) 83.15 (7.40) 11.80 (4.92) 2.9 (S4)

p = .220 p = .om1 p = .o001

1.075 (0.712)

1.226 (0.550)

p = .I61

Patient age ADL score MEDICUS acuity of illness class score Mean DRG weight ADL = Activities of Daily Living DRG = Diagnosis Related Group

TABLE 3. Effect of Dementia on Hospital Course

Total length of stay Number of acute days Number of alternate level of care days Mean length of stay for DRG Mean difference between actual length of stay and mean DRG length of stay Net hospital profit or loss**

Patients without Dementia

Patients with Dementia

Mean (SD) 17.12 (2.67) 12.83 (9.73) 4.29 (8.52) 6.59 (2.65) 10.53 (12.33) $-3,331.50 ($-7,286.06)

Mean (SD) 33.55 (35.03) 23.54 (26.82) 10.01 ( 17.94) 7.78 (2.70) 25.29 (34.46) $4,910.44 ($9,034.46)

*Wilcoxon rank sum test **Net Profit or Loss = Hospital operating costs minus revenues

p = .001* p = .005* p = .029* p = ,011 p = .003 p = ,066

Effectof Dementia on Acute Care

237

and complicating diagnoses to overshadow the contribution of dementia to the hospitalization. In striking contrast to the patients without dementia, who required admission primarily for cardiovascular disease, the most common reason for hospital acute care admission in demented patients was an infectious disease. Dementia increases the risk of infectious disease in elderly patients because it precipitates a complex series of interrelated events. Alterations in feeding, continence, and mobility lead to impairments of primary host defenses against infection, such as aspiration, urinary colonization with fecal flora, and impaired skin barrier function. Malnutrition is the leading cause of secondary immunodeficiency. Dementia also predisposes to accidental injuries, poor compliance with medication orders and nutritional advice,nonrecognition or misinterpretationof signs and symptoms,and deficits in health-seekingbehavior. To the extent that any of these risk factors is a consequence of dementia, it can be considered a related comorbidity. Our findings demonstrate that patients whose acute medical admissions are precipitated or complicated by dementia suffer complex, lengthy, and expensive hospitalizations. In many cases, while dementia may be the proximate cause of admission to acute medical care-for example, as a cause of aspiration pneumonia-generally dementia is not recognized as an acute care diagnosis. Previous work has demonstrated that psychiatric comorbidities complicate the course and cost of care in medically ill patients (Fulop et al., 1987;Mayou et al., 1988)and that severity of cognitive impairment is inversely related to survival (Gamsu et al., 1990; Walsh et al., 1990). In addition to the impact on acute care of direct comorbidities secondary to dementia, such as infectious disease, the acute treatment of coexisting independent comorbidities, such as myocardial infarction or renal failure,may be more difficult in the presence of dementia. However, the exact role dementia plays in acute care resource consumption by the frail elderly is not well documented and requires further study. In summary, the medical complications associated with dementia are generally inevitable as the disease progresses to the moderate and severe stages, at which point dementia becomes a complex medical condition whose treatment involves many levels of ambulatory and acute care resources. The absence of dementia as either an admitting or a comorbid diagnosis in the majority of the cases in our geriatric medical unit suggests that the impact of dementia in acute medical care of the frail elderly may be significantly underestimated from the perspective of both hospital course and costs.

REFERENCES Abraham, E. (1989). Cellular and humoral immunity in the critically ill patient. In S. Shoemaker,S. Ayres, A. Grenvik,P. R Holbrook, & W.L. Thompson (Eds.), Textbook of critical care (pp. 1024-1028). Philadelphia: W. B. Saunders. American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders (3rd ed., rev.). Washington, DC:Author. Bums, A., Jacoby,R.,Luthert,P.,&Levy,R. (1990)Causeofdeathin Alzheimer’sdisease. Age and Aging, 19,341-344.

L. Torianet al.

238

Centers for Disease Control. (1991). Mortality from Alzheimer Disease-United Stabs, 1979-1987. Mortality and Morbidity WeeklyReport, 39,785-788. Chandra, V., Bharucha, N. E., & Schoenberg, B. S. (1986). Conditions associated with Alzheimer’s disease at death: Case-control study. Neurology, 36,209-21 1. Diesfeldt,H. F., van Houte, L. R., & Moerkens,R. M. (1986). Duration of survival in senile dementia. Acta Psychiatrica Scandinavia, 73,366-371. Fischer, J., &Johnson,M. A. (1990). Low body weight and weight loss in the aged. Journal of the American Dietetics Association, 1697-1 706. Flaten, T. P. (1989). Mortality from dementia in Norway, 1969-83.Journal of Epidem’ology and Community Health, 43,285-289. Fulop, G., Strain, J. J., Vita, J., Lyons, J. S., &Hammer, J. S. (1987). Impact of psychiatric comorbidity on length of hospital stay for medicWsurgical patients: A preliminary report. American Journal of Psychiatry, 144,878. Gamsu, C. V., McLaren, S. M., Barry, F., & McPherson, F. M. (1990). Prediction of survival in dementia by cognitive tests: A six-year follow-up. British Journal of Clinical Psychology, 29,99-104. Harrier, D. C. (1989) Critical care decisions in trauma. In W. C. Shoemaker,S. Ayres, A. Grenvik, P. R. Holbrook, & W. L. Thompson (Eds.), Textbook of critical care (pp. 1271-1274). Philadelphia: W. B. Saunders. Hassall,C.,Prior,P., & Cross, K. W. (1988). Apreliminary study of excessmortality using a psychiatric case register. Journal of Epidemiology and Community Health, 42, 286-289.

Hay, 1.W., & Emst, R. D. (1987). The economic costs of Alzheimer’s disease. American Journal of Public Health, 77,9. Huang, L. F., Cartwright, W. S., & Ju, T. W. (1988). The economic cost of senile dementia in the United States, 1985. Public Health Report, 103, 1. Jagger, C., & Clarke, M. (1988). Mortality risks in the elderly: Five-year follow-upof a total population. International Journal of Epidemiology, 17,111-1 14. Jahnigen, D., Hannon, C., Laxson, L., & LaForce, F. M. (1982). Iatrogenic illness in hospitalized elderly veterans. Journal of the American Geriatrics Society, 30, 387. Jeejeebhoy, K. N. (1989). Nutrition in critical illness. In W. C. Shoemaker,S. Ayres, A. Grenvik, P. R. Holbrook, & W. L. Thompson (Eds.), Textbook of critical care (pp. 1093-1 118). Philadelphia: W. B. Saunders. Katz, S. (1983). Assessing self-maintenance: Activities in Daily Living, mobility and Instrumental Activities in Daily Living. Journal of the American Geriatrics Society, 129,1229.

Lewis, E. J., & Bell, S. J. (1990). Nutritional assessment of the elderly. In J. E. Morley, Z. Glick, & L. Z. Rubinstein (Eds.), Geriatric nutrition (pp. 73-87). New York: Raven Press. Li,G., Shen, Y.C.,Chen, C. H., Zhau, Y.W., Li, S. R.,et al. (1991). A three-year followup study of age-related dementia in an urban area of Beijing. Acta Psychiatrica Scandinavia,83.99-104. Martin, D. C., Miller, J. K., Kapoor, W., Arena, V. C.,& Boller, F. (1987). A controlled study of survival with dementia. Archives of Neurology, 44,1122-1 126. Martyn, C. N., & Pippard, B. C. (1988). Usefulness of mortality data in determining the geography and time trends of dementia. Journal of Epidemiology and Community Health, 42,2.

Effectof Dementia on Acute Care

239

Mayou, R., Hawton, K., & Feldman, E. (1988). What happens to medical patients with psychiatric disorder? Journal of Psychosomatic Research, 32.54 1. Molsa, P. K., Martilla, R. J., 62 Rinne, U. K. (1986). Survival and cause of death in Alzheimer's disease and multi-infarct dementia.Acta Neurologica Scandinavia, 74, 103-107. Newman, S . C., & Bland, R. C. (1987). Canadiantrends in mortality from mental disorders. Acta Psychiatrica Scandinavia, 76, 1-7. Patterson, C. (1986). Iatrogenic disease in late life. Clinics in Geriatric Medicine, 2,21. Peck, A., Wollock, L., & Rodstein, M. (1978). Mortality of the aged with chronic brain syndrome: Further observationsin a five-year study.Journal of the American Geriatrics Society, 264. Saviteer,S . M., Gamsa,G. P., & Rutala,W. A. (1988). Nosocomialinfections in theelderly: Increased risk per hospital day. American Journal of Medicine, 88,661. Schimpff, S . C., DeJongh, C. A., & Caplan, E. S. (1989). Infections in the critical care patient. In W. C. Shoemaker,S. Ayres, A. Grenvik,P. R. Holbrook, & W.L. Thompson (Eds.), Textbook of critical care (pp. 767-779). Philadelphia: W. B. Saunders. Volicer, L., Seltzer, B., Rheaume, Y., Karner, J., Glennon, M., et d. (1989). Eating difficulties in patients with probable dementia of the Alzheimer type. Journal of Geriatric Psychiatry and Neurology, 2 , 188-195. Walsh, J. S., Welch, H. G., & Larson,E. B. (1990). Survivalof outpatientswith Alzheimertype dementia. Annals of Internal Medicine, 113,429-434. Zarian, D. A., Peter, S. A., Lee, S., & Kleinfeld, M. (1989). The causes and frequency of acute hospitalization of patients with dementia in long-term care facility. Journal of the National Medical Association, 81,4.

Acknowledgment. This work was supported in part by The Florence J. Could Foundation. Correspondence. Direct correspondence to Dr. Lucia Torian, HIV Serosuxvey Program, New York City Department of Health, 346 Broadway, New York, NY

10013, U.S.A.

The effect of dementia on acute care in a geriatric medical unit.

Treatment of dementia costs billions of dollars in long-term care and community services every year. Dementia also burdens the acute care system and m...
687KB Sizes 0 Downloads 0 Views