Psychosomatic Medicine Faster Service from Outside North America
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sloan, D. M.
Right arrow Articles by Schubert, D. S. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sloan, D. M.
Right arrow Articles by Schubert, D. S. P.
Related Collections
Right arrow Consultation Psychiatry
Right arrow Depression
Psychosomatic Medicine 61:21-25 (1999)
© 1999 American Psychosomatic Society

ORIGINAL ARTICLES

A Five-Year Study on the Interactive Effects of Depression and Physical Illness on Psychiatric Unit Length of Stay

Denise M. Sloan, PhD, James Yokley, PhD, Howard Gottesman, MD and Daniel S. P. Schubert, MD, PhD

From the Department of Psychology (D.M.S.), School of Medicine, MetroHealth Medical Center Campus (J.Y., H.G., D.S.P.S.), Case Western Reserve University, Cleveland, Ohio.

Address reprint requests to: Denise M. Sloan, PhD, University of Florida, Health Sciences Center, Box 100165, Gainesville, FL 32610-0165.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHOD
 RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 
OBJECTIVE: Previous research has shown that hospital length of stay among medical patients is significantly increased by comorbid mental illness, in particular depression. However, few studies have examined the length of stay effect of comorbid physical and mental illness among psychiatric patients.

METHOD: The present study examined the effect of comorbid physical and psychiatric illness on hospital length of stay among 2323 psychiatric inpatient admissions over a 5-year period. Patients were grouped into seven diagnostic categories.

RESULTS: Average length of stay was significantly longer for patients with comorbid physical diagnoses (mean = 20.01 days) than for patients with no physical diagnoses (mean = 16.63 days). Analyses of the psychiatric categories revealed that the average length of stay for depressed patients was significantly greater for those with comorbid physical diagnoses (mean = 19.73 days) than for depressed patients with no comorbid physical diagnoses (mean = 13.96 days). No other psychiatric group evidenced a significant increase in length of stay for comorbid physical illness.

CONCLUSIONS: Results suggest that comorbid physical diagnosis increases length of stay among psychiatric patients overall, with increased hospitalization stay for depressed patients, in particular.

Key Words: depression, • physical illness, • comorbidity, • hospital length of stay.

Abbreviations: LOS = length of stay;; ANOVA = analysis of variance;; MCMI-II = Millon Clinical Multiaxial Inventory, Version 2.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHOD
 RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 
In recent years, there has been an emphasis on decreasing medical costs, such as decreasing hospital length of stay, by developing more effective treatment programs. Physical and psychiatric illness comorbidity has been speculated to increase medical costs by both increasing hospital length of stay, as well as increasing hospital readmissions (1). Due to the interest in decreasing medical care costs, there have been several studies that have examined the effect of comorbid physical and psychological illness on hospital medical wards. Few studies, however, have investigated the existence of physical illness among psychiatric patients and the effect comorbidity may have on factors such as hospital LOS.

Effects of LOS for psychiatric comorbidity among a variety of medical patients have been examined, including patients with spinal cord damage (2), burn patients (3), stroke patients (4), and patients in physical diagnosis-related groups (5). Overall, these studies find that psychiatric illness significantly increases hospital LOS for medical patients compared with noncomorbid patients. Furthermore, the severity of physical illness is increased in the presence of psychological illness (6). More recently, researchers have investigated whether specific psychiatric disorders lengthen LOS among medical patients. Several studies have found that depression significantly increases hospital LOS among a variety of medical patients (4, 6–8). In addition to increased hospital LOS, Mayou et al. (9) found that comorbid depression was associated with continuing medical and social problems after discharge. Some studies have also found that cognitive impairment (particularly cognitive impairments associated with dementia (10)) and anxiety (11) increase hospital LOS among medical patients.

Although previous research has examined psychiatric disorders among medical patients, few studies have investigated the effects of comorbid physical illness among psychiatric patients. Examining the effects of physical illness among psychiatric patients, as well as examining which psychiatric groups are most affected by physical illness, would increase our knowledge of how to most effectively treat psychiatric patients. In one of the first studies to address these issues, Schubert et al. (12) investigated the effects of comorbid physical illness and psychiatric disorder on LOS of a hospital psychiatry unit over a 17-month period. Results were that psychiatric patients with comorbid physical illness evidenced significantly longer LOS than patients with no comorbid physical illness. An examination of diagnostic categories revealed that comorbid depression and physical illness resulted in a significant increase (ie, nearly doubling) in LOS over depressed patients without comorbid physical diagnosis. No other psychiatric diagnostic category demonstrated a significant LOS difference between comorbid and noncomorbid groups.

The purpose of the present study was to extend the findings of Schubert et al. (12) by including a reliability measure of the psychiatric diagnoses, which the previous study did not include. In addition, the sample size in the Schubert el al. study prevented a statistical conclusion about psychiatric categories that are less frequent than depression, such as anxiety disorder that has been found to increase LOS on a medical unit (10). Therefore, the present study included a larger sample size so that psychiatric categories, such as anxiety disorders, could be examined. Due to the focus of the present study on primary psychiatric diagnoses, cognitive impairment was not examined. LOS was compared between psychiatric inpatient groups with comorbid physical illness and psychiatric inpatient groups with no comorbid physical illness over a 5-year period between 1987 and 1992. This time period was used due to the minimal influence of managed care policies. Managed care has sharply decreased hospital LOS in recent years; therefore, managed care policies would represent a substantial confounding variable if a more recent time period was examined.

Consistent with previous findings (12), the present study predicted that psychiatric patients with comorbid physical illness would have significantly longer LOS than patients without comorbid physical illness. It was also expected that comorbid physical illness would significantly increase LOS among patients diagnosed with depression and anxiety disorders compared with depressed and anxiety disorder patients with no comorbid physical illness.


   METHOD
TOP
 ABSTRACT
 INTRODUCTION
 METHOD
RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 
The subjects were 2323 consecutive admissions to a psychiatry ward over a 5-year period (1987–1992) at MetroHealth Medical Center. The Medical Center is a county hospital and draws patients primarily from Cleveland, Ohio, who are typically from lower socioeconomic classes.

The hospital computer information systems printed all diagnoses on all subjects over a 5-year period, which were then divided into seven psychiatric disorder categories. Psychiatric disorders such as delirium or dementia were excluded. Psychiatric disorder categories were as follows: 1) psychoses, 2) depressions, 3) personality disorders, 4) anxiety disorders, 5) adjustment disorders, not depressed, 6) bipolar disorders, not depressed, and 7) other disorders (eg, eating disorders, substance abuse/dependency, etc.). Disorder categories were divided additionally into those with and without comorbid physical illness disorders. Patients were placed in diagnostic categories according to their primary mental illness diagnosis. The hospital LOS, in days, was recorded for each patient. Throughout the 5-year period, there were three psychiatrists who worked on the unit for the entire time period. These psychiatrists decided patient diagnoses, treatment formulation, and patient discharges within a treatment team approach. No changes in policy or procedures on the unit occurred during the time period examined in the present study. In addition, all physical illnesses were diagnosed by faculty psychiatrists and psychiatric residents by usual methods, including history, physical examination, laboratory tests, and medical records. Physical illnesses were limited to those appearing in the International Classified of Diseases-9 (ICD 9), thus including both major disorders and minor disorders such as allergic rhinitis.

Table 1 presents demographic characteristics of patients by diagnostic category. The mean age of patients with no comorbid physical diagnosis (mean = 33.2, SD = 11.1) was significantly lower than the mean age of patients who had comorbid psychiatric and physical diagnoses (mean = 42.8, SD = 15.6; t(2321) = 16.79, p < .001). Separate {chi}2 analyses were conducted to examine the effect of gender and ethnicity on comorbid physical illness. There was a significant effect for gender {chi}2, (1, N) = 14.52, p > .001. Post hoc analyses indicated that men were more likely not to have comorbid physical illness compared with women. No significant {chi}2 effect was found for ethnicity. The mean LOS for all hospital admissions was 18.20 days.


View this table:
[in this window]
[in a new window]
  		Table 1. Demographic Characteristics for All Patient Admissions by Psychiatric Diagnostic Category
 
The MCMI-II was used to examine the reliability of psychiatric diagnoses made by ward psychiatrists. The MCMI-II is a self-report measure of psychiatric clinical syndromes. Research has found that the MCMI-II reliably discriminates between populations of psychiatric inpatients at broad levels of the DSM-III-R (13) and has equal or greater accuracy for identifying psychiatric disorders compared with the Minnesota Multiphasic Personality Inventory (14). A {kappa} analysis (15) was performed with 157 patients, using the seven diagnostic categories included in the study. {kappa} statistics provide information on the percent agreement of diagnoses between the MCMI-II and the unit psychiatrists, adjusted for chance agreement.

To ascertain that LOS was normally distributed, skewness and kurtosis were calculated and a scatterplot was prepared. All three procedures indicated normality, making analysis of variance an appropriate statistical procedure for the variable of hospital LOS. To examine the effect of comorbid physical illness on LOS a 2 (gender) x 2 (physical diagnosis) x 7 (psychiatric diagnosis) ANOVA was performed, using hospital LOS, in days, as the dependent variable. Gender was included in the ANOVA due to significant {chi}2 finding.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHOD
 RESULTS
DISCUSSION
 NOTES
 REFERENCES
 
Agreement rate between patient psychiatric diagnoses obtained from the MCMI-II and the diagnoses made by the unit psychiatrists revealed a {kappa} statistic value of .82. This finding suggests that the diagnoses made by ward psychiatrists were reliable, compared with an independent patient self-report measure.

The 2 x 2 x 7 ANOVA revealed a significant main effect for comorbid physical illness, F(1, 2296) = 22.52, p < .001, indicating that patients with comorbid physical diagnoses had longer average LOS (mean = 20.01 days) than patients without comorbid physical illness (mean = 16.63 days). There was also a significant main effect for psychiatric diagnostic categories, F(6, 2296) = 27.92, p < .001. However, there was no significant interaction effect for physical diagnosis by psychiatric category, nor was a significant main effect or interaction effect found for gender.1

t tests between patients with and without comorbid physical diagnoses were conducted on each psychiatric diagnostic category to examine the a priori hypotheses concerning an extended hospital LOS associated with comorbid physical illness and the psychiatric diagnoses of depression and anxiety disorders. Bonferroni correction was used to protect against Type I errors, therefore, an {alpha} level of .007 was used for all t tests.

Patient hospital LOS in psychiatric diagnostic categories, broken down by presence or absence of comorbid physical illness, is illustrated in Figure 1. As expected, follow-up t tests revealed that depressed patients with comorbid physical illness had significantly longer hospital LOS (mean = 19.72) than depressed patients without a comorbid physical diagnosis (mean = 13.96; t(890) = 4.78, p < .001). No other diagnostic category evidenced a significant LOS difference between groups with and without comorbid physical illness. Moreover, depression with comorbid physical illness increased LOS by approximately 6 days over depression without comorbid disease. When examining the remaining six psychiatric groups, the largest increase in LOS between comorbid and noncomorbid groups was 3 days, thus adding considerable medical costs for the comorbid depressed group compared with other psychiatric groups.



View larger version (27K):
[in this window]
[in a new window]
  		Fig. 1. Effect of comorbid physical illness on length of stay. 1 = psychosis; 2 = depression; 3 = personality disorder; 4 = anxiety; 5 = adjustment disorder, not depressed; 6 = bipolar, not depressed; 7 = other.

 
Table 2 presents physical illness categories by psychiatric diagnostic categories. No substantial differences are evidenced among these categories, suggesting that the significant LOS finding among patients with comorbidity of depression and physical illness is not likely a result of a particular physical illness category among depressed patients.


View this table:
[in this window]
[in a new window]
  		Table 2. Psychiatric Diagnostic Category by Physical Illness Category: Percentage of Patients with Physical Illness
 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHOD
 RESULTS
 DISCUSSION
NOTES
 REFERENCES
 
Previous studies examining the comorbid mental illness effects among medical patients have found that depression results in significantly longer LOS on medical units (1, 2, 5) and the present study, along with the study by Schubert et al. (12), indicates that the increase of LOS also exists among depressed psychiatric patients. Contrary to our prediction, there was no significant difference in LOS between comorbid and noncomorbid anxiety patients. However, it is possible that a significant difference was not revealed due to the limited number of patients with a primary anxiety disorder. It will be important for future studies to replicate this finding, examining a larger number of patients with a primary anxiety disorder.

There are several possible explanations for the finding of a significant LOS increase in the depression group, such as differential degrees of physical illness severity, differential patient report related to type of patient psychopathology, and increased drug side effects due to physical illness comorbidity. Future research needs to explore factors that impact psychiatric hospital LOS among depressed patients with comorbid physical illness.

It is important to note that the high percentage of comorbid physical illness among the overall group of psychiatric patients in the present study is consistent with findings from previous studies conducted with both inpatients and outpatients (16, 17). Therefore, it seems that physical illness is highly prevalent among psychiatric patients and may complicate their treatment if not adequately addressed by performing routine medical evaluations with psychiatric patients.

The present study had several shortcomings. First, psychiatric diagnoses were not made using structured interviews. Although the present study included a diagnostic reliability measure on a percentage of the patients, it is possible that some diagnoses were not accurate. Severity of physical illness was not examined in this study, and so future investigations of psychiatric groups should control for this factor. The present study is also retrospective; prospective studies would strengthen the conclusions presented here. Finally, it should also be noted that the results of this study are based on one setting. Despite the large number of subjects in the present study, length of hospitalization may vary according to other nonpatient factors. Therefore, reliability of the results obtained in the present study would be increased by these results being replicated in other medical settings.

The results of this study, as well as the results of Schubert et al. (12), suggest that comorbidity of depression and physical illness is associated with a longer psychiatric LOS and clinicians should expect and plan for this complex challenge. To increase hospitalization efficiency, medical consultation should be requested as soon as physical illnesses are suspected and their recommendations implemented as soon as medical conditions are suspected. Future research is needed to determine why physical illness may have particularly severe effects on depressed patients compared with other psychiatric groups.


   NOTES
TOP
 ABSTRACT
 INTRODUCTION
 METHOD
 RESULTS
 DISCUSSION
 NOTES
REFERENCES
 
Because age between comorbid and noncomorbid patients was significant, a 2 x 2 x 7 covariate analysis of variance was performed using age as a covariate. The results did not differ from the main analysis in the study and the covariate of age was not significant. Back

Received for publication November 12, 1997.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHOD
 RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 

  1. Saravay SM, Lavin M. Psychiatric comorbidity and length of stay in the general hospital: a critical review of outcome studies. Psychosomatics 1994; 35: 233–52.[Abstract/Free Full Text]
  2. Lyons JS, Larson DB, Burns BJ, Cope N, Wright S, Hammer JS. Psychiatric comorbidities and patients with head and spinal cord trauma. Gen Hosp Psychiatry 1988; 10: 292–7.[Medline]
  3. Brezel BS, Kassenbrock M, Stein JM. Burns in substance abusers and in neurologically and mentally impaired patients. J Burn Care Rehab 1988; 9: 169–71.[Medline]
  4. Schubert DSP, Burns R, Paras W, Sioson E. Increase of medical hospital length of stay by depression in stroke and amputation patients: a pilot study. Psychother Psychosom 1992; 57: 61–6.[Medline]
  5. Fulop G, Strain JJ, Vita J, Lyons JS, Hammer JS. Impact of psychiatric comorbidity on length of hospital stay for medical/surgical patients: a preliminary report. Am J Psychiatry 1987; 144: 878–82.[Abstract/Free Full Text]
  6. Koranyi EK. Morbidity and rate of undiagnosed physical illness in a psychiatric clinic population. Arch Gen Psychiatry 1979; 36: 414–9.[Abstract]
  7. Imbroden JB, Canter A, Cliff LE, Rever RW. Brucellosis III psychologic aspects of delayed convalescence. Arch Intern Med 1959; 103: 406–14.
  8. Koenig HG, Shelp F, Goli V, Cohen HJ, Blazer DG. Survival and healthcare utilization in elderly medical inpatients with major depression. J Am Geriatr Soc 1989; 37: 599–606.[Medline]
  9. Mayou R, Hawton K, Feldman E, Ardern M. Psychiatric problems among medical admissions. Int J Psychiatry Med 1991; 21: 71–84.[Medline]
  10. Fields SD, MacKenzie CR, Charlson ME, Sax FL. Cognitive impairment: can it predict the course of hospitalized patients? J Am Geriatr Soc 1986; 34: 579–85.[Medline]
  11. Hall RC, Popkin MK, Devaul RA, Faillace LA, Stickney SK. Physical illness presenting as a psychiatric disease. Arch Gen Psychiatry 1978; 35: 1315–20.[Abstract]
  12. Schubert DSP, Yokley J, Sloan D, Gottesman H. Impact of the interaction of depression and physical illness on psychiatric unit’s length of stay. Gen Hosp Psychiatry 1995; 17: 326–334.[Medline]
  13. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition-Revised. Washington, DC, American Psychiatric Association, 1987.
  14. Libb JW, Murray J, Thurstin H, Alarcon RD. Concordance of the MCMI-II, the MMPI, and Axis I discharge diagnosis in psychiatric inpatients. J Pers Assess 1992; 58: 580–90.[Medline]
  15. Cohen J. A coefficient of agreement for nominal scales. Educ Psychol Meas 1960; 20: 37–46.
  16. Cyr JJ, Haley GA. Use of demographic and clinical characteristics in predicting length of psychiatric stay: a final evaluation. J Consult Clin Psychol 1983; 51: 637–40.[Medline]
  17. Roca RP, Breakey WR, Fisher PJ. Medical care of chronic psychiatric outpatients. Hosp Comm Psychiatry 1987; 38: 741–5.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 PsychosomaticsHome page
C. G. Lyketsos, G. Dunn, M. J. Kaminsky, and W. R. Breakey
Medical Comorbidity in Psychiatric Inpatients: Relation to Clinical Outcomes and Hospital Length of Stay
Psychosomatics, February 1, 2002; 43(1): 24 - 30.
[Abstract] [Full Text]

Home page
 Psychiatr. Serv.Home page
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]

Home page
 JWatch PsychiatryHome page
Physical Illness and Depression
Journal Watch Psychiatry, April 1, 1999; 1999(401): 5 - 5.
[Full Text]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sloan, D. M.
Right arrow Articles by Schubert, D. S. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sloan, D. M.
Right arrow Articles by Schubert, D. S. P.
Related Collections
Right arrow Consultation Psychiatry
Right arrow Depression

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS