Psychosomatics 39:273-280, June 1998
© 1998 The Academy of Psychosomatic Medine
A Prospective Study of the Impact of Psychiatric Comorbidity on Length of Hospital Stays of Elderly Medical-Surgical Inpatients
George Fulop, M.D., M.S.,
James J. Strain, M.D.,
Marianne C. Fahs, Ph.D.,
James Schmeidler, Ph.D., and
Stephen Snyder, M.D.
Received December 6, 1996; revised June 4, 1997; accepted July 8, 1997. From the Departments of Community Medicine, Geriatric Medicine, Psychiatry (Division of Behavioral Medicine), and Biomathematical Sciences, The Mount Sinai School of Medicine, New York. Address reprint requests to Dr. Fulop, Department of Medical Affairs, Merck-Medco Managed Care, LLC, MS R2–21, 100 Summit Ave., Montvale, NJ 07645.
|
ABSTRACT
|
To determine the difference in length of hospital stay for geriatric medical-surgical inpatients with or without psychiatric comorbidity, the authors prospectively interviewed 467 admissions by using the Structured Clinical Interview for DSM-III-R and the Mini-Mental State Exam. At admission, 208 (44.5%) inpatients had a current psychiatric comorbidity, 51(10.9%) had an anxiety disorder, 88 (18.8%) had a depressive disorder, and 126 (27%) had cognitive impairment. The patients with cognitive impairment had a significantly prolonged hospital stay compared with those without cognitive impairment (14.6 vs. 10.6 days). No difference existed in length of stay for the patients with and without anxiety disorders (11.6 vs. 11.6 days) or depressive disorders (11.0 vs. 11.8 days). In view of the limited resources available for screening elderly medical-surgical inpatients for psychiatric comorbidity, this study suggests the utility of identifying cognitive impairment and targeting it for interventions to reduce the clinical burden and to decrease hospital stays.
Key Words: Length of Stay • Comorbidity • Hospital Stays • Medical-Surgical • Elderly • Geriatrics • Inpatients
|
INTRODUCTION
|
Psychiatric comorbidity (PC), that is, psychiatric conditions coexisting with medical or surgical disorders, is associated with excess medical resource utilization (e.g., length of hospital stay [LOS]), especially in elderly medical-surgical inpatients in the general hospital. Preliminary reports that used retrospective data from 2 hospitals indicated that the co-occurrence of a psychiatric disorder and medical or surgical conditions was associated with a 1.5- to 2-fold prolongation of the patient's LOS.1 These early reports relied on hospital discharge data abstracted from the medical record. Psychiatric diagnostic criteria were not regularly applied by health care personnel during a patient examination, and the diagnoses listed on the discharge abstract were not confirmed by a mental health specialist. The chart notation of a mental symptom or diagnosis by any health care worker was sufficient to have a patient labeled with a psychiatric comorbidity. In these studies, the possibility of either a false-positive or false-negative classification of a PC is attributable to 1) misidentification, 2) underrecognition, or 3) underrecording of psychiatric disorders and symptoms by nonmental health hospital personnel.
Several prospective reports also confirm that increased hospital stays are associated with psychiatric disorders.2–6 For example, in a survey of 457 mixed-age group (under and over 65 years of age) inpatients in Great Britain, Mayou and colleagues found increased LOS for patients with cognitive impairment, but not depressive or substance use disorders.2 However, Mayou and his group did not address the important confound of severity of medical illness. In mixed-age groups of 278 and 455 admissions, respectively, Saravay et al. and Levenson and colleagues both observed that anxiety, depression, and cognitive impairment were associated with increased LOS.3,4 Among 237 medical-surgical admissions over 70 years of age, cognitive impairment (delirium) was associated with increased LOS and mortality.5
The current study builds upon these reports by prospectively studying a multi-ethnic population of geriatric inpatients on medical and surgical wards by using trained raters to administer a structured psychiatric clinical interview and controlling for other factors known to affect LOS (e.g., severity of illness). The main goal of this study was to observe the associated LOS for elderly medical-surgical patients with and without a psychiatric comorbidity (anxiety, depression, or cognitive impairment).
|
METHODS
|
During a 1-year time-frame (1990–1991), 13,295 geriatric (65 years of age or older) patients were admitted to the general medical and surgical wards at the Mount Sinai Hospital (MSH), New York City, representing about one-third of all admissions. Those geriatric admissions who met the following criteria were eligible to participate in the study: 1) minimum hospital stay of 72 hours (to meet the requirements of the hospital's Institutional Review Board (IRB) regarding consent procedures, to account for acute disabling medical illness, and to minimize patient interview burden); 2) permission by the patient and his/her physician to participate in the interview; 3) able to render informed consent for participation; 4) physically able to be interviewed (e.g., not intubated, too weak, demented, or sedated secondary to anesthesia); 5) English- or Spanish-speaking; and 6) not a readmission of a subject previously enrolled. After complete description of the study, all subjects received an informed-consent form and consented to the study interviews. Informed-consent forms and procedures, study-interview instruments, and methods were reviewed and approved by MSH's IRB.
Due to limited resources, we sampled eligible geriatric admissions on 156 randomly selected days for enrollment over a 1-year period (9/90–9/91). Specifically, among 2,842 geriatric admissions to MSH on the sample enrollment days, 561 (19.7% of the geriatric admissions) met all eligibility criteria. In 94 cases (17% of the 561 eligible), either the patient or his/her physician refused to participate in the study, leaving 467 subjects in our study sample.
All subjects were interviewed on the third hospital day by using the following battery: 1) Structured Clinical Interview for DSM-III-R (SCID-R), nonpsychiatric patient version: Overview, Modules A=Mood Disorders and F=Anxiety Disorders)7; 2) Folstein Mini-Mental State Exam (MMSE)8,9; 3) Geriatric Depression Scale (GDS), the score used as a measure of depression severity10,11; and 4) Katz Index of Activity of Daily Living Scale.12 Medical-surgical disease severity was defined by medical-surgical diagnosis-related group (DRG) relative weights to allow comparisons across medical disease categories and to examine the relationship between psychiatric comorbidity and utilization of services.4
All subjects were then categorized into two cohorts: the presence or absence of a PC based on finding cognitive impairment, depressive disorder, or anxiety disorders. Specifically, cognitive impairment was defined as a patient with a MMSE score less than 24.8,9 Depressive disorder was defined by a SCID-R Module A diagnosis of a current episode of major depression, bipolar disorder/depressed, or dysthymia. A current anxiety disorder was diagnosed by meeting criteria for SCID-R Module F disorders: panic, phobic, obsessive-compulsive, or generalized anxiety disorder.7
SCID-R organic exclusion criteria for symptoms—that is, not rating a symptom as present when the etiology may be an underlying medical condition or current medication—were not used, based upon the discussion by Cohen-Cole and Stoudemire of the difficulty of achieving reliability with exclusive criteria.13 In brief, inclusive criteria rate all symptoms based on the phenomenological appearance rather than on pathophysiologic ("organic") etiology and potentially overestimate the presence of psychiatric disorders in the medically ill, whereas exclusive criteria are more restrictive and potentially underestimate the presence of psychiatric conditions. On a practical basis, reliability of determining the etiology of psychiatric symptoms is difficult to achieve—so an inclusive criteria methodology may increase reliability at the expense of overinclusivity.
Statistical analyses were conducted by using the SAS Institute PC SAS Windows Version 6.0814 and included parametric descriptive statistics for all categorical (Pearson's chi-square) and continuous variables (pooled or separate variance-estimate t-tests) and multivariate analyses. All tests of significance were two-sided. Logarithmic transformation of the dependent variable, LOS, was performed to reduce the impact of outliers and the skewing of distributions of hospital stays toward longer LOS. All multivariate analyses are reported in untransformed formats and logarithmic transformed units (LOS—days, and log LOS, respectively).
|
RESULTS
|
The demographics of the study sample are shown in Table 1. Thirty-seven percent of the subjects were African American or Hispanic, and over half were female. The elderly patients with psychiatric comorbidity were significantly more likely to be African American or Hispanic, female, on Medicaid, and less likely to be discharged to home. The psychiatric comorbidity patients also were significantly older, less educated, more functionally impaired (a higher activities of daily living score), and had more concurrent medical-surgical diagnoses.
The categories and distribution of psychiatric comorbidities are presented in Table 2. Nearly half the sample had cognitive impairment, or a SCID-R psychiatric diagnosis of depression or anxiety disorder (diagnostic categories not mutually exclusive). Overall, the subjects had a mean±standard deviation (SD) MMSE score of 25.0±4.8 (MMSE score  24: "no cognitive impairment") and a mean±SD GDS score of 7.9±6.1 (GDS score <11: "normal"). The depressed subjects had a mean±SD GDS score of 15.5±5.4 (indicating mild-to-moderate severity), compared with 6.2±4.8 for the the nondepressed elderly.
Psychiatric comorbidity, as a composite variable composed of any cognitive impairment, anxiety, or depression, was associated with a significantly (P<0.023) longer hospital stay by 2.6 days (Table 3). The log LOS showed a more significant (P=0.001) difference, corresponding to a 22% prolongation in hospital stay. When cognitive impairment, anxiety, and depression were examined separately, only cognitive impairment was significant, with a 4.0-day or 30% prolonged LOS. Further analyses limiting the depressive disorder category to just major depression (excluding dysthymia) or the anxiety-disorders category to just generalized anxiety (excluding panic, phobic, and obsessive-compulsive disorders) did not alter the findings.
The LOS and log LOS of geriatric medical-surgical inpatients were examined in a priori multivariate models predicting hospital stay by the presence or absence of a psychiatric comorbidity while controlling for functional status, disease severity (medical-surgical DRG relative weight4), insurance status (Medicaid vs. no Medicaid, e.g., Medicare only), ethnicity, gender, age; and discharge to home vs. to a facility are presented (Table 4). The composite variable—any psychiatric disorder—had only a trend toward significance (P<0.07) in predicting increased log LOS. In contrast, cognitive impairment was a statistically significant predictor of length of hospital stays (both untransformed and transformed), even after controlling for coexisting anxiety and depressive disorder (Table 4). Functional status and severity of medical illness were statistically significant covariates in all analyses and predicted significant proportions of the distributions of LOS. Discharge to home was significant only in the untransformed LOS analyses.
|
DISCUSSION
|
This study prospectively confirmed the association of a specific psychiatric comorbidity—cognitive impairment—with significantly prolonged hospital stays. The finding is consistent with the majority of prospective studies in the United States and internationally that examined mixed-age group populations.15 The impact of cognitive impairment on LOS remains significant even after accounting for demographic variables, medical acuity, and functional status.
When compared with retrospective reports, the magnitude of the increase in LOS associated with cognitive impairment is not as robust. Whereas a 1.5- to 2-fold increase in LOS was described for the elderly inpatient with dementia or delirium in our preliminary retrospective report, only a 1.22-fold increase in LOS for cognitive impairment was observed prospectively.1 A major reason for the difference in the findings between these two studies may be the exclusion of the more severe cognitively impaired subjects who could not participate in an interview from the current study—those very inpatients who tend to have longer hospital stays. In a retrospective study that relies upon medical record review, nonpsychiatrists may underidentify less severe (mild-to-moderate) cognitively impaired elderly—those who are more likely to have shorter hospital stays—while more accurately identifying and thereby including the more severely impaired, which would result in an overstatement of the magnitude of the increase in LOS.
The absence of an association of anxiety and depression with increased LOS is consistent with all but three studies.15 In the elderly, psychiatric comorbidity other than cognitive impairment may not have a significant additional impact on hospital stay when controlling for the effects of functional status or severity of illness. Anxiety and depression in an elderly inpatient may not delay discharge from the hospital, compared with the burdens posed by cognitive impairment (e.g., diminished compliance, self-care).
This prospective study identified a wide range of depressive morbidity among the less severely medically ill, whereas the evaluation of depression by nonpsychiatrists, who do not use structured interviews in retrospective studies, underreports mild-to-moderate depressive disorders and tends to only identify the more severe depressive states. Therefore, there may be a tendency to overstate the increase in the association of psychiatric comorbidity and LOS in retrospective studies and to understate the association prospectively.1
A prospective screen for minimal threshold—less apparent—categorical diagnoses of anxiety or depression, or for cognitive impairment, without regard to severity or clinical significance, would tend to identify psychiatric conditions that are milder and less likely to prolong hospitalization. It may not be the presence or absence of a psychiatric condition per se that drives prolonged hospital stays. Rather, it may be the severity of the comorbidity and its attendant behavioral disturbances (e.g., noncompliance) that directly affect the delivery of diagnostic procedures, treatments, nursing services, or discharge disposition, all of which may prolong LOS. Future studies need to examine specific delays for diagnostic tests, procedures, acceptance of medications, and whether these aberrant patient behaviors are associated with 1) increased LOS and 2) psychiatric morbidity.
The question remains whether the use of inclusive DSM-III-R criteria, rather than the organic exclusion criteria of the SCID-R, leads to the overdiagnosis of anxiety and depressive disorders.7 A more restrictive set of criteria may have decreased the number of elderly labeled as having a psychiatric comorbidity, but when present, may have had a greater effect on LOS. By using inclusive criteria, this study anticipated DSM-IV,16 in which mood disorders secondary to medical illness or medication were recategorized from organic mental disorders to mood disorders.14
The current study illustrates the challenges confronting the attempt to assess the frail, acutely ill, elderly inpatient for mental symptomatology, since only 19% of the geriatric medical-surgical admissions could be interviewed because of the study's eligibility criteria. Alternative methodologies to address this limitation might include the use of ratings based on observation and informant interviews, rather than interview-based instruments such as the SCID-R, to make it feasible to include all admissions in an epidemiological survey of the acutely ill elderly.
Despite its lack of impact on LOS, depression, more importantly, may negatively affect a geriatric patient's quality of life, equivalent to the impact of diabetes and other chronic illnesses.17 Furthermore, anxiety and depression may have a delayed or greater impact after discharge from the hospital, in outpatient settings or a nursing home, and when the medical severity of illness is less acute. Several reports highlight increased outpatient medical utilization by depressed elderly.18,19 Therefore, although depression was not associated with increased hospital stay in this group of elderly inpatients, medical resource use could be increased in the posthospital discharge phase or during the entire episode of care in the cohort with a psychiatric comorbidity.
Psychiatric interventions may reduce the impact of psychiatric comorbidity on LOS, especially among elderly women. Mumford et al. noted a decrease of 2 days in LOS after a wide range of psychosocial interventions.20 Strain and colleagues demonstrated the cost-effectiveness of a psychiatric screening intervention in elderly hip-fracture patients that reduced hospital stays by 2 days and resulted in a net saving of $167,000 for the 175 study subjects.21 It appears that timely psychiatric case identification and intervention reduce inpatient medical resource use.
By documenting the magnitude of the increase in LOS and highlighting the cognitively impaired geriatric patient as at particular risk for prolonged stay, this study suggests the need to design interventions targeted at recognizing and managing cognitive impairment, with the aim of reducing the psychiatric distress and excess LOS. Angell says that rationing of medical services may be unnecessary if inappropriate and inefficient medical resource utilization is identified and reduced.22 With limited resources available for screening elderly inpatient populations for psychiatric comorbidity, targeting the detection of cognitive impairment for psychosocial interventions is warranted and may reduce the clinical burden for the patient, family, and staff and decrease the length of medical-surgical inpatient stays.
|
ACKNOWLEDGMENTS
|
This paper was presented at the 1997 Annual Meeting of the American Psychiatric Association in San Diego, California, on May 21, 1997.
This study was supported by the National Institute of Mental Health (NIMH), Services Research Branch, FIRST Award (Grant No. R29–43378). The authors thank research assistants and volunteers who assisted in the completion of this project; Charlotte Muller, Ph.D., Professor Emeritus; and Agnes Rupp, Ph.D., Epidemiology and Services Research Branch, NIMH.
|
REFERENCES
|
-
Fulop G, Strain JJ, Vita J, et al: Impact of psychiatric comorbidity on length of hospital stay of medical/surgical inpatients: a preliminary report. Am J Psychiatry 1987; 144:878–882[Abstract/Free Full Text]
-
Mayou R, Hawton K, Feldman E, et al: Psychiatric problems among medical admissions. Int J Psychiatry Med 1991; 21:71–84[Medline]
-
Saravay SM, Steinberg MD, Weinschel B, et al: Psychological comorbidity and length of stay in the general hospital. Am J Psychiatry 1991; 148:324–329[Abstract/Free Full Text]
-
Levenson JL, Hamer RM, Rossiter LF: Relation of psychopathology in general medical inpatients to use and cost of services. Am J Psychiatry 1990; 147:1498–1503[Abstract/Free Full Text]
-
Francis J, Martin D, Kapoor WN: A prospective study of delirium in hospitalized elderly. JAMA 1990; 263:1097–1101[Abstract]
-
Ames D, Tuckwell V: Psychiatric disorders among elderly patients in a general hospital. Med J Aust 1994; 160:671–675[Medline]
-
Spitzer RL, Williams JBW, Gibbon M, et al: The Structured Clinical Interview for DSM-III-R (SCID). I. History, rationale and description. Arch Gen Psychiatry 1992; 49:624–629[Abstract]
-
Folstein M, Folstein SE, McHugh PR: "Mini-Mental state"—a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 9:189–198
-
Escobar JL, Burnam A, Karno M, et al: Use of the Mini-Mental State Examination (MMSE) in a community population of mixed ethnicity. J Nerv Ment Dis 1986; 174:607–614[Medline]
-
Yesavage J, Brink TL, Rose TL, et al: Development and validation of a geriatric depression screening scale: a preliminary report. J Psychiatr Res 1983; 17:37–49
-
Gonzalez F: Caracteristicas tecnicas y modificaciones introducidas en dos escalas de depresion: estudio piloto para la adaptacion y baremacion de la GDS y el BDI a la poblacion geriatric espanola. Psicogeriatria 1988; 4:59–66
-
Katz S, Ford AB, Moskowitz AB, et al: Studies in illness in the aged: The Index of ADL, a standardized measure of biological and psychosocial function. JAMA 1963; 185:914–919
-
Cohen-Cole SA, Stoudemire A: Major depression and physical illness. Psychiatr Clin North Am 1987; 10:1–17
-
SAS Institute: SAS System for Windows 3.1, Release 6.08. Cary, NC, SAS Institute, 1992
-
Saravay SM, Lavin M: Psychiatric comorbidity and length of stay in the general hospital: a critical review of outcome studies. Psychosomatics 1994; 35:233–252[Abstract/Free Full Text]
-
American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th Edition. Washington, DC, American Psychiatric Association, 1994
-
Hays R, Wells K, Sherbourne C, et al: Functioning and well-being outcomes of patients with depression compared with chronic general medical illnesses. Arch Gen Psychiatry 1995; 52:11–19[Abstract]
-
Waxman HM, Carner EA, Blum A: Depressive symptoms and health service utilization among the community elderly. J Am Geriatr Soc 1983; 31:417–420[Medline]
-
Callahan CM, Hui SL, Nienaber NA, et al: Longitudinal study of depression and health services use among elderly primary care patients. J Am Geriatr Soc 1994; 42:833–838[Medline]
-
Mumford E, Schlesinger HS, Glass GV, et al: A new look at evidence about reduced cost of medical utilization following mental health treatment. Am J Psychiatry 1984; 141:1145–1158[Abstract/Free Full Text]
-
Strain JJ, Lyons JS, Hammer JS, et al: Cost offset from a psychiatric consultation-liaison intervention with elderly hip fracture patients. Am J Psychiatry 1991; 148:1044–1049[Abstract/Free Full Text]
-
Angell M: Cost containment and the physician. JAMA 1985; 254:1203–1207[Abstract]
This article has been cited by other articles:
|
|
|
|
 
V. de Albuquerque Citero, P. B. de Araujo Andreoli, L. A. Nogueira-Martins, and S. B. Andreoli
New Potential Clinical Indicators of Consultation-Liaison Psychiatry's Effectiveness in Brazilian General Hospitals
Psychosomatics,
February 1, 2008;
49(1):
29 - 38.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. A. Bourgeois, D. M. Hilty, J. A. Wegelin, and R. E. Hales
Cognitive-Disorder Diagnoses in Inpatient Psychosomatic-Medicine Consultations: Associations With Age and Length of Stay
Psychosomatics,
October 1, 2006;
47(5):
414 - 420.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. A. Bourgeois, W. S. Kremen, M. E. Servis, J. A. Wegelin, and R. E. Hales
The Impact of Psychiatric Diagnosis on Length of Stay in a University Medical Center in the Managed Care Era
Psychosomatics,
October 1, 2005;
46(5):
431 - 439.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
F. A. Smith, J. Querques, J. L. Levenson, and T. A. Stern
Psychiatric Assessment and Consultation
Focus,
April 1, 2005;
3(2):
241 - 251.
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. A. Bourgeois, J. A. Wegelin, M. E. Servis, and R. E. Hales
Psychiatric Diagnoses of 901 Inpatients Seen by Consultation-Liaison Psychiatrists at an Academic Medical Center in a Managed Care Environment
Psychosomatics,
February 1, 2005;
46(1):
47 - 57.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
S. M. Saravay, M. Kaplowitz, J. Kurek, D. Zeman, S. Pollack, S. Novik, S. Knowlton, M. Brendel, and L. Hoffman
How Do Delirium and Dementia Increase Length of Stay of Elderly General Medical Inpatients?
Psychosomatics,
June 1, 2004;
45(3):
235 - 242.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
G. Fulop, J. J Strain, and G. Stettin
Congestive Heart Failure and Depression in Older Adults: Clinical Course and Health Services Use 6 Months After Hospitalization
Psychosomatics,
October 1, 2003;
44(5):
367 - 373.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. M. Prieto, J. Blanch, J. Atala, E. Carreras, M. Rovira, E. Cirera, and C. Gasto
Psychiatric Morbidity and Impact on Hospital Length of Stay Among Hematologic Cancer Patients Receiving Stem-Cell Transplantation
J. Clin. Oncol.,
April 1, 2002;
20(7):
1907 - 1917.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
H. G. Rincon, M. Granados, J. Unutzer, M. Gomez, R. Duran, M. Badiel, C. Salas, J. Martinez, J. Mejia, C. Ordonez, et al.
Prevalence, Detection and Treatment of Anxiety, Depression, and Delirium in the Adult Critical Care Unit
Psychosomatics,
October 1, 2001;
42(5):
391 - 396.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. R. Hopko, D. Lachar, S. E. Bailley, and R. V. Varner
Assessing Predictive Factors for Extended Hospitalization at Acute Psychiatric Admission
Psychiatr Serv,
October 1, 2001;
52(10):
1367 - 1373.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. F. Borus, A. J. Barsky, L. A. Carbone, A. Fife, G. L. Fricchione, and S. L. Minden
Consultation-Liaison Cost Offset: Searching for the Wrong Grail
Psychosomatics,
August 1, 2000;
41(4):
285 - 288.
[Full Text]
|
|
|
|
|
|
|
C. G. Lyketsos, J.-M. E. Sheppard, and P. V. Rabins
Dementia in Elderly Persons in a General Hospital
Am J Psychiatry,
May 1, 2000;
157(5):
704 - 707.
[Abstract]
[Full Text]
|
|
|
Get information about faster international access.
Privacy Policy
Copyright © 1998
Academy of Psychosomatic Medicine.
All rights reserved.
Home
| Search
| Current Issue
| Past Issues
| Subscribe
| All APPI Journals
| Help
| Contact Us
|
TOP
ABSTRACT
INTRODUCTION
