Psychosomatics 40:345-355, August 1999
© 1999 The Academy of Psychosomatic Medine
Integrating Medical and Psychiatric Treatment in an Inpatient Medical Setting
The Type IV Program
Yasuhiro Kishi, M.D., and
Roger G. Kathol, M.D.
Received June 16, 1998; revised November 4, 1998; accepted December 4, 1998. From the Department of Psychiatry, Nippon Medical School Chiba Hokusoh Hospital, Chiba, Japan; and the Departments of Psychiatry and Internal Medicine, University of Iowa Hospitals, Iowa City, Iowa. Address correspondence and reprint requests to Dr. Kathol, Departments of Psychiatry and Internal Medicine, University of Iowa Hospitals, C429–GH1, 200 Hawkins Drive, Iowa City, IA 52242; e-mail: cartsolu{at}tutti.inav.net
|
ABSTRACT
|
This study compares the treatment of patients with comorbid medical and psychiatric illness admitted to a high-acuity (Type IV) integrated medicine and psychiatry inpatient program with patients having psychiatric symptoms on general internal medicine wards (IMWs). More patients in the Type IV program had agitation, suicidal ideation, or psychosis as psychiatric admissionbehaviors when compared to IMW patients. Medical symptom improvement was comparable in the two settings, whereas, psychiatric symptoms improved more in the Type IV Program than on the IMWs despite more significant illness and comparable lengths of stay. Integrated care on the Type IV unit allowed shorter total lengths of stay for medical patients with serious psychiatric illness than would have occurred had the traditional sequential approach to care been used. The integrated Type IV medicine and psychiatry treatment program represents an efficient and effective process improvement in the way that medical patients with comorbid medical and psychiatric illness can be treated.
Key Words: Inpatients • Medical Psychiatry • Consultation Psychiatry
|
INTRODUCTION
|
Integrated inpatient medicine and psychiatry treatment programs are specifically designed to address the needs of patients with medical and psychiatric comorbidity. Kathol et al.1 categorized these programs into four types (Table 1). Type I programs treat patients with acute psychiatric disorders who also have stable medical problems for which nonacute attention is required. Most general psychiatric wards fall into this category, since about one-third accept patients totally dependent in daily activities.2 Type II programs, conversely, treat patients with acute medical disorders in whom nonacute psychiatric problems co-occur. While many patients on medical wards fall into this category, only those in which on-site psychiatric liaison is available are the psychiatric problems routinely addressed. These two types of units constitute the traditional approach to patient care for those in whom nonacute medical and psychiatric difficulties in one discipline is coupled with acute difficulties in the other.
A population of inpatients, however, exists in whom the acuity of medical and psychiatric illness is not easily addressed in traditional medical or psychiatric settings, even when there is a desire to work with patients with comorbid illness, as in Type I and Type II programs. Examples of such patients are prevalent—a hyperactive manic patient with cellulitis requiring intravenous antibiotics, a suicidal patient on peritoneal dialysis, a psychotic patient after a burst of glucocorticoids for arthritis, and a hypoxemic delirious patient after a pulmonary embolus, just to mention a few. Type III and IV programs introduce real change in clinical capabilities by providing increased levels of both medical and psychiatric services in the same setting. These programs require the active and sustained involvement of primary care physicians and psychiatrists. Both medical and psychiatric safety features form a prerequisite for the physical settings in which these units are housed, and nursing personnel must receive extra training in psychiatric and medical nursing techniques.
Most Type III programs are administered through psychiatry, since the clinical needs of inpatients with combined medical and psychiatric illness are commonly encountered by consultation-liaison psychiatrists. While Type III programs strive to assume responsibility for medical and psychiatric care, they are hampered by their inability to handle high-acuity medical illness. This problem results because these programs often lack an adequate level of medical nursing expertise, access to timely emergency medical procedures (including advanced cardiac life support), and the availability of medical physician coverage.
Reimbursement disincentives, introduced by the segregation of medical treatment, also play a role in limiting medical and psychiatric services in Type III programs. Most behavioral health care companies do not compensate for medical services. When patients have both illnesses at the same time, behavioral health care organizations refuse responsibility for medical care, pointing out that the medical difficulties should be covered under the patient's medical benefits package.
To maximize income, Type III programs usually choose a Medicare diagnostic-related group (DRG) exempt status. Under this and other "carved out" reimbursement schedules, medical services are not paid for even when appropriately used. Reimbursement for medical procedures is disallowed when provided in a psychiatric setting or in response to orders by psychiatrists, the required admitting physicians in DRG-exempt programs.
Finally, Type III programs frequently face challenges in providing treatment to acutely ill medical patients because some of these programs are located in stand-alone psychiatric facilities. As a result, nonpsychiatric supervision of patient care can be remote, laboratory and X-ray services are difficult to access, and adverse medical events often necessitate transfer. These and other aforementioned problems invariably lead to a psychiatric, rather than medical, referral base. While Type III programs add value to patient care, the fact that barriers to primary care physician admissions are created by who, where, and how these programs are set up, makes them less accessible to many medical patients who may benefit by the care that they provide.
Type IV programs are capable of treating patients with any level of medical acuity that can be handled on general medicine wards and any level of psychiatric acuity addressed on inpatient psychiatric intensive care services. Though there are currently few Type IV programs in general hospitals, interest in their development is growing in health care systems because of the program's ability to provide cost-effective—yet quality—medical services to complicated and expensive patients.
This study directly compares characteristics and clinical outcomes of medical patients with psychiatric comorbidity admitted to internal medicine wards (IMWs) with patients admitted to a Type IV integrated medicine and psychiatry treatment program (Type IV program) administered by the internal medicine department.
|
METHODS
|
Setting
The University of Iowa Hospitals and Clinics is a 900-bed, tertiary-care, academic medical center. It had 206 internal medicine beds at the time of the study. Of these, 42 beds from the general internal medicine wards, 20 beds from the gastroenterology ward, 16 beds from the pulmonary ward, 11 beds from the endocrinology ward, 16 beds from the cardiology ward, and all 12 beds from the Type IV integrated medicine and psychiatry treatment program were used to recruit patients. The average lengths of stay for patients admitted to these units at the time this study was performed were general medicine: 5.8 days, gastroenterology: 6.5 days, pulmonary: 8.6 days, endocrinology: 5.5 days, cardiology: 3.8 days, and the integrated medicine and psychiatry program: 10.3 days.
In 1986 the University of Iowa Hospitals and Clinics and the University of Iowa College of Medicine collaborated in the creation of a 12-bed Type IV integrated medicine and psychiatry treatment program.1 Central characteristics of this clinical inpatient service included physician coverage by members of the Department of Internal Medicine and the Department of Psychiatry by using a co-attending model, the recruitment and training of nursing personnel who have skills in both medical and psychiatric nursing techniques, and the renovation of a clinical location within the general hospital to accommodate patients with comorbid medical and psychiatric illness. The Type IV program was designated DRG nonexempt to allow admission of patients by both primary care physicians and psychiatrists. With these components, it has become recognized by providers and insurers as both a medicine or a psychiatry service provider.
This Type IV integrated treatment program differs from most existing programs1,3 that combine medical and psychiatry services in several critical ways. First, the Type IV program is administered through the Department of Internal Medicine, which ensures that primary care physician expertise and coverage and the physical organization of the Type IV program is on par with other medical services in the hospital. The Type IV program was designed to address all medical needs that could be taken care of on other general IMWs throughout the hospital. Second, the Type IV program was supported by the Department of Psychiatry, which ensured that psychiatric physician coverage was available with timely response to patient needs 24 hours a day.
Third, patients could only be admitted to the Type IV program if they had active medical and psychiatric illness. The program did not take overflow patients from either medicine or psychiatry. Furthermore, patients in whom it was anticipated that either the medical or psychiatric illness would require prolonged care (arbitrarily 5 to 7 days) after one of the illnesses had improved were transferred to a medicine-only or psychiatry-only treatment location.
Fourth, before the Type IV program opened all nurses were hired to work in this service area because they wished to combine medical and psychiatric nursing skills. Each received training in medical and psychiatric nursing techniques, with more time devoted to the clinical discipline from which they had not come. Training and coverage were administered through the internal medicine nursing division in collaboration with and the support of psychiatry nursing.
Fifth, the Type IV program was located in the midst of other primary care inpatient units in the general hospital. The program unit was renovated to provide clinical and safety components that would allow the same type of care provided on other general IMWs and on the acute general psychiatry wards. Core medical features included medical gases in all rooms, wide doorways, an infectious isolation room, a physical examination room, clean and dirty utility rooms, and space for medical equipment. Core psychiatric features included an activities room; a seclusion room; a nourishment room; a group/family therapy room; patient observation and staff communication capabilities throughout; sturdy furniture; shatterproof windows; barricade-proof doors; shortable electric sockets; tamperproof ceilings; breakaway curtain rods; the absence of potentially dangerous objects, such as cords, sharp objects, plastic bags, etc.; and laundry facilities.
Consistent with the aim of treating high-acuity patients, the Type IV program had a locked entrance and involuntary admission capabilities. The program did not provide cardiac monitoring, treat patients on ventilators, or take patients requiring isolation with positive pressure laminar flow. But the program did provide most other medical and surgical services. All psychiatric assessment techniques and treatments were available, including the development and implementation of sophisticated behavior-modification programs, amytal/benzodiazepine interviews, and electroconvulsive therapy.
Finally, all patients admitted to the Type IV program required active medical and psychiatric physician involvement using a co-attending model. These individuals ("co-attenders") reviewed patient progress and saw patients together daily. Since the Type IV program was organized in a teaching hospital, residents from internal medicine and from psychiatry worked together to provide patient care.4
By the time of the study, referral patterns to the Type IV program were well established. About 70% were admissions either directly from physicians in Iowa (40%) or from the emergency room or clinics (30%), and 30% were in-house transfers, primarily from IMWs. There was a bias for patients with a more severe or prolonged interaction of psychiatric and medical illness to be directed to the Type IV program.
Protocol
This protocol is a prospective, structured chart assessment comparing clinical characteristics and outcomes of patients admitted to the Type IV integrated medicine and psychiatry treatment program with patients admitted to internal medicine units who were identified by the nursing personnel caring for the patients as having some form of psychiatric problem. Charts of patients admitted to the Type IV program were reviewed within 24 hours of admission and then every 3 days until discharge, unless the patient was in the hospital for more than 2 weeks, at which time follow-up assessments were performed weekly. When the clinical notes were unclear, nurses providing care to the patients were personally interviewed to complete the data set. Charts of patients admitted to the other IMWs were reviewed within 24 hours of when the patient was identified as having a psychiatric contribution to his/her medical presentation. Follow-up reviews were performed on inpatients admitted to the Type IV program.
Subject Recruitment
After obtaining human subject review approval, 105 consecutive medical inpatients admitted to the 12-bed Type IV integrated medicine and psychiatry treatment program between April 1994 and September 1994 were entered into the study. All patients had simultaneous and active medical and psychiatric illnesses or an interaction of medical illness and psychiatric behaviors that made treatment in the general medical setting difficult.
Also, 105 patients who were identified between April 1994 to December 1994 by nursing supervisors taking daily census from charge nurses on the general medicine, pulmonary, gastrointestinal, endocrine, and cardiac services as having psychiatric or behavioral problems associated with the medical disorder requiring admission served as a comparison population. Recruitment of these patients depended on the cooperation and time of the nursing supervisors and the availability of the investigators to review charts. Because of these constraints, priority was given to the entry of the 634 patients admitted to the general medicine wards. Most of these patients were screened for the study. When time allowed, patients from specialty wards were also entered into the study. Sixty-three patients from general medicine, 16 patients from endocrinology, 16 patients from gastroenterology, 8 patients from pulmonary, and 2 patients from cardiology comprised the comparison population. All of these patients were confirmed to have active medical and psychiatric difficulties by one of two study raters (YK and JC) before entry.
Patients with a history of psychiatric illness but no active psychiatric symptoms were excluded from the study; however, patients who developed emotional or behavioral symptoms during the course of hospitalization were included. Patients on telemetry or ventilators were also excluded.
Measures
Two rater/investigators (YK and JC), trained in university psychiatric residency programs, documented the information in the data set. At the start of the study, the charts of 10 patients were reviewed by both raters independently. The results of these assessments were compared. Where discrepancies arose, rules were developed to consistently score future patients. Also, when questions about how to record information on a complicated patient occurred during the course of the study, this was discussed by the study investigators and resolved by consensus.
Age, gender, marital status, employment status, tobacco and alcohol use, medical and psychiatric diagnoses, the patient's behavioral or emotional disturbance, history of personal and/or family psychiatric illness, and severity of medical and psychiatric illness were recorded at study entry. If the medical or psychiatric diagnoses changed during the course of hospitalization, study data were altered to most accurately reflect the patient's illnesses. An abbreviated laboratory version of the Acute Physiology and Chronic Health Evaluation, or APACHE-III, designed to assess medical severity in general medicine, rather than intensive care unit, patients was also administered upon admission.5 Higher scores indicate more serious illness.
Since this study was approved by the human subjects committee as a chart review without direct patient contact, it was not possible to confirm diagnoses through patient interviews with structured instruments. The two study investigators (YK and JC) gathering data, however, were medical doctors and had both been trained in the use of DSM-IV6 as a psychiatric diagnostic classification system. They also had direct and concurrent access to the medical and psychiatric health professionals familiar with the patient's complaints and symptoms. As a result, the psychiatric diagnoses recorded were considered accurate based on timely and confirmable information.
During the course of hospital stay, the severity of the patient's medical and psychiatric illness was documented by using a modified Karnofsky Performance Scale7 and the Global Assessment of Functioning (GAF) Scale,6 respectively. The Karnofsky Performance Scale was altered to exclude hospitalization as an indicator of medical severity, since all were inpatients. Both scales record severity on a scale of 0 to 100, with higher scores indicating less severe impairment.
Disposition location or death was recorded as patients exited the study. Disposition categories include personal residence; nursing or county home; long-term, specialized treatment facilities; or jail.
Statistical Analyses
Continuous data were analyzed by using the Student's t-test unless nonnormally distributed, in which case the Welch analysis of variance (ANOVA) was substituted. The chi-square test was used to compare categorical data unless small sample size necessitated the use of Fisher's exact test.
|
RESULTS
|
Demographics
Characteristics of the IMW and Type IV program patients are summarized in Table 2. Thirty-one patients were transferred to the Type IV program from other locations within the University of Iowa Hospitals, compared with 14 of the IMW patients ( 2=8.17, df=1,208, P=0.004). Within the University Hospitals, the Type IV program provided service to other medicine wards (n=17: 5 intensive care unit, 12 other), surgery wards (n=6), neurology wards (n=4), psychiatry wards (n=3), and ophthalmology wards (n=1). Ten of the 14 transfers to the IMWs were from the medical intensive care unit. Twenty-two of the IMW patients had psychiatric consultations.
Seventy-four Type IV program patients were admitted directly. When these patients were compared with the 91 directly admitted IMW patients, there were no significant differences in patient characteristics between each group or in the entire study sample.
Primary Medical and Psychiatric Illness
Table 3 summarizes the types of primary and secondary medical conditions the patients experienced. These findings indicate that neurological presentations, medication adjustment and adverse drug effects (pharmacological), intoxications, or overdoses were more frequently the admitting medical problems for patients entering the Type IV program, whereas pulmonary, gastrointestinal, and endocrine problems were more often the admission reason on the IMWs. Since 40 IMW patients were recruited from the endocrine, gastrointestinal, and pulmonary wards, the numbers in Table 3 do not reflect the frequency of illness encountered on the general medicine wards. When patients on the general medicine wards were analyzed independently, pulmonary, gastrointestinal, and endocrine disease accounts for 12%, 15%, and 3% of the primary medical disorders, respectively. The need for acute medication adjustment or care of adverse drug reactions was the admitting medical reason in 26% of general medicine ward patients, whereas neurological conditions and intoxications/overdoses remained low at 2% and 10%, respectively.
View this table:
[in this window]
[in a new window]
|
TABLE 3. Primary medical illness necessitating admission and the presence of specific medical conditions in patients treated on the internal medicine wards (IMWs) and in the Type IV integrated medicine and psychiatry treatment program
|
Table 4 summarizes the psychiatric conditions for which patients required attention. Mood disorders and psychosis were more frequently the psychiatric reason for admission in patients admitted to the Type IV program, whereas substance abuse was the most frequent psychiatric condition requiring attention in patients on the IMWs. Delirium was equally distributed between IMW and the Type IV program patients; however, delirium constituted a much smaller percentage of total admissions to the IMWs.
View this table:
[in this window]
[in a new window]
|
TABLE 4. Primary psychiatric illness necessitating attention and the presence of specific psychiatric conditions in patients treated on the internal medicine wards (IMWs) and in the Type IV integrated medicine and psychiatry treatment program
|
Nineteen patients admitted to the Type IV program and 10 admitted to the IMW did not fit into a convenient psychiatric category. The Type IV program patients who were difficult to classify often had less common but debilitating or costly combinations of illnesses. For example, four patients had catatonia of uncertain etiology, five patients had nonsuicidal self-induced medical conditions (factitious disorder), two had agitated organic mental disorders, two had unmanageable mixed personality disorders, and one had hypersexuality. Most IMW patients falling into this group were either anxious or had a combination of mental retardation with other behavioral difficulties.
If psychiatric problems are listed by their behavioral or emotional presentation rather than diagnostic subtype, it becomes more apparent why IMW nursing personnel were concerned about the patients or why they were admitted or transferred to the Type IV program (Table 5). Agitated, uncooperative, suicidal, or psychotic patients were more likely to be treated in the Type IV program than on the IMW. Conversely, patients with anxiety tended to remain on IMWs.
View this table:
[in this window]
[in a new window]
|
TABLE 5. Emotional or behavioral problems in patients on the internal medicine wards (IMWs) and in the Type IV integrated medicine and psychiatry treatment programa
|
Delirium was frequently the study entry reason on both the IMW and in the Type IV program; however, on a per admission basis, delirium was more commonly treated in the Type IV program. Unexplained somatic complaints were uncommonly considered psychiatric difficulties that would qualify the patient for study entry.
Medical and Psychiatric Acuity
Table 6 shows medical and psychiatric acuity at entry and at discharge from the units. IMW patients were more likely to be considered medically "disabled, requiring special care and assistance," whereas patients admitted to the Type IV program just "required considerable assistance and frequent medical care.". Although both showed improvement during hospitalization, patients on the IMW were significantly more likely to be unable to carry on normal activity or to do active work, even though they were able to care for themselves at discharge, compared with patients in the Type IV program.
View this table:
[in this window]
[in a new window]
|
TABLE 6. Severity of medical and psychiatric symptoms on the internal medicine ward (IMW) and in the Type IV integrated medicine and psychiatry treatment program patients at study entry and exit
|
Patients admitted to the Type IV program demonstrated impaired reality testing or communication that prevented effective work, school, social, and personal performance, compared with IMW patients who had serious symptoms that only limited social, occupational, and school performance. The patients in the Type IV program showed significantly more improvement during hospitalization than the IMW patients.
Length of Stay
IMW study patients and Type IV program patients had lengths of stay 7 and 12 days longer than patients without psychiatric comorbidity admitted to the IMWs, respectively (Table 7). The total length of stay in the hospital is nearly 5 days longer for patients in the Type IV program (17.5 days) when compared with patients on the IMWs (12.5 days). The longer length of stay in the Type IV program is nearly entirely due to the time spent on other units before transfer to the Type IV program. When patients who had been internally transferred to the ward in which they were studied were excluded, the average length of stay was 10.5±9.7 days for the 91 IMW patients and 13.5±12.9 days for the 74 Type IV program patients, an insignificant difference. The total length of stay for the 31 Type IV program patients who were transferred was 18.8±14.3 days, whereas their length of stay in the Type IV program was 12.6±11.3 days.
View this table:
[in this window]
[in a new window]
|
TABLE 7. Length of stay for patients admitted to the internal medicine wards (IMWs) and the Type IV integrated medicine and psychiatry treatment program
|
|
DISCUSSION
|
This is the first study that describes the contribution that an acute-care, inpatient integrated medicine and psychiatry treatment program makes when commitment and participation by primary care is as great as that by psychiatry. Ten of 11 inpatient integrated treatment programs reviewed by Harsch et al.3 were administered through psychiatry, with varying levels of commitment by primary care specialties to cover medical needs. Only one program was administered through the Department of Internal Medicine (the program described in this report). Certainly, other integrated programs exist that cater to the psychiatric needs of medical patients, for example, one described by Stoudemire et al.;8 however, most are Type II units1 with limited psychiatric capabilities. Most integrated treatment programs are designed by psychiatry to serve the needs of psychiatrists faced with the challenges these patients present.2,9–18
The University of Iowa admits about 35,000 patients per year. The 12-bed Type IV integrated treatment program accounts for about 1% of these admissions. With a 97% bed-occupancy rate between 1986 and 1993, and a routine waiting list, the number of patients who could be served in a Type IV integrated program is likely much larger than the admissions reflect. Despite reduction in the hospital's bed complement during the past 4 years by nearly 15%, the integrated treatment program is scheduled to increase from 12 to 18 beds.
The Type IV program serves as a sieve for complicated and expensive patients in the University of Iowa system. Attempts to benchmark its activity have met with limited success, because few similar programs exist in the United States, or the world. In a sense, the Type IV program serves as the repository for patient outliers, which is reflected by longer lengths of stay when compared with patients without psychiatric comorbidity admitted to the general medicine wards. Referring physicians recognize the complicated nature of these patients and refer them for the intensive capabilities found in the Type IV program. In fact, it is the difficult nature of these patients that led to the development of the integrated program at Iowa. These patients were ineffectively and inefficiently treated in other locations in the hospital.
This report finds that patients admitted to the Type IV program are qualitatively and quantitatively different from patients with psychiatric comorbidity on the general IMWs. For instance, they are younger. This finding may be somewhat surprising to some, since patients over 65 often exhibit psychiatric symptoms during the course of a medical illness. Several factors probably influence this finding. First, dementia, per se, is not a reason for admission to the Type IV program at Iowa. Demented patients must also have delirium, depression, or some other psychiatric problem that warrants admission, in addition to an active medical illness. Second, many psychiatric illnesses have their onset between the ages of 20 and 40. Often, these disorders are complicated by medical conditions that require acute intervention, such as suicide attempts in depressed patients, hyponatremia in polydipsic schizophrenia patients, and severe malnutrition in anorexic patients.
Our findings also suggest that patients admitted to the Type IV program are significantly more likely to have a personal history of psychiatric treatment and other family members with psychiatric illness. These findings are understandable given the higher severity of psychiatric symptoms in patients in the Type IV program. Interestingly, patients on the IMWs are as likely to have a personal history of psychiatric illness.
All patients admitted to the Type IV program had medical conditions that would have required treatment in the general medical setting. However, they would have been difficult or impossible to treat in a general psychiatry setting. The type of medical illness that precipitated the Type IV program admission was most often neurologic, often delirium, or related to medication/substance use or abuse.
While the most common reason for admission may have been neurologic or pharmacologic, a combination of medical factors frequently complicated the picture. For instance, a patient with Cushing's syndrome may have overdosed on a medication due to the common occurrence of depression in patients with hypercortisolemia. The admission reason would be "overdose," but the Cushing's syndrome would have required treatment at the same time. Similar circumstances arise in patients on renal dialysis, with congestive heart failure, etc.
The head-to-head comparison of severity of medical illness demonstrated that patients in the Type IV program were less medically ill than patients admitted to the IMWs. This finding often reflects the fact that some patients are referred to the unit by clinicians who think there might be an organic etiology for the patient's behavior but are incorrect in this impression. This mistake is also compounded by the fact that significant psychiatric comorbidity, which makes it difficult to treat less severe medical conditions without a supervised setting, leads to referral. Physicians admitting patients recognize that the Type IV program provides an environment that encourages patient cooperation and adherence. The environment plays a role in the effective treatment of existing illness and prevention of complications or medical worsening, which could lead to prolonged functional impairment and more frequent hospitalizations. For instance, treating depression in a patient with insulin-dependent diabetes who repeatedly enters the hospital with sugars greater than 400 may lessen the likelihood of future admissions. In the Type IV program, insulin management can be maximized while depression is treated. The effectiveness with which this is accomplished in the Type IV program at Iowa is reflected by equivalent Karnofsky improvement but lower Karnofsky severity at discharge.
While Table 6 demonstrates that medical severity is less in the Type IV program overall, it does not reflect that very medically ill patients are also treated in the Type IV setting. For the integrated program to maximize value to the University of Iowa system, it must be capable of addressing the needs of patients who would be unmanageable without high medical-acuity capabilities. For this reason, the Type IV program also gets its share of patients with pulmonary emboli, with renal failure on peritoneal or hemodialysis, with central lines, and the like. This capability can only be provided because of the involvement of internal medicine in the clinical and administrative structure.
While the level of severity of medical illness would make it impossible to treat patients in the psychiatric setting, the level of psychiatric illness also makes it difficult to manage these patients in the medical setting. General medical wards commonly see patients with substance abuse (mainly alcoholism), delirium, mood disorders, and anxiety. These illnesses are debilitating, as is reflected by the GAF scores of patients entering the IMWs. The ability to effectively improve patient's psychiatric comorbidity, however, is less than in the Type IV program. Interestingly, only 22 patients had psychiatric consultations, despite recognition by the medical staff that behavioral or emotional problems were present.19,20
There are certain psychiatric conditions with which medical personnel felt particularly uncomfortable. Nearly all patients with severe mood disorder, especially when suicidal ideation was present, psychosis, or severe agitation with violence potential were treated in the Type IV program.
Whether patients with both medical and psychiatric illness were treated on the IMWs or in the Type IV program, they required longer hospital stays than patients without psychiatric comorbidity. When the days spent on other hospital units before transfer are excluded from the total length of stay, patients in the Type IV program have an equivalent number of days of hospitalization to IMW patients.
The health care system is moving from a fee-for-service to a managed care-and-capitated model. By using this new model, emphasis is placed on decreasing hospitalizations, reducing the length of hospital stay, and maximizing outpatient intervention. On the face of it, Type IV programs appear expensive, because they have longer lengths of stay than other units in the hospital. Our data, however, do not bear this out.
The Type IV program at the University of Iowa admits patients who contribute substantially to medical care expenditures, the outliers, which would make one wonder about the adequacy of treatment if hospital stays were shorter, given the population served. Excluding the time in hospital before admission to the Type IV program, patients could be treated in 13 days in the integrated setting, not significantly different than patients with less severe psychiatric comorbidity on the IMWs. With a traditional approach to medical care, the same patients would require up to 21 days of hospitalization with sequential admissions (6 days on medicine and 15 days on psychiatry), an average savings of 8 days of hospitalization. If 1 day of hospitalization costs the average patient $500 (a conservative figure), then a 12-bed Type IV program with 350 admissions/year could save the health care system $1.4 million per year.
While this is the first study, to our knowledge, that illustrates the contribution that Type IV programs can make to inpatient medical services, several limitations should be recognized. First, the study is a prospective chart review in which direct information from patients was not obtained. Patients were enrolled as they were identified to have psychiatric comorbidity with their medical illness; however, incomplete information missing from the chart could be obtained from health care personnel working with the patients to clarify symptom presentation.
Second, patients on the IMW were identified by nursing supervisors after consultation with the head nurse on each unit but not by a physician. Since we did not evaluate patients directly, some patients on the IMW may have been missed. Third, patients without active psychiatric problems were excluded, even though they had a past psychiatric history, for example, psychiatrically and symptomatologically stable patients on serotonin reuptake inhibitors for depression would not have been studied. Fourth, interviewers were not blind to the location from which patients were recruited. This factor could have led to recording bias as diagnosis and outcome was assessed.
Fifth, patients on IMWs were not consecutively inducted into the study, and they were recruited from several general and specialty medicine units. In an attempt to decrease population bias, preference was given to the recruitment of patients from the general medical wards. Despite this procedure, a higher number of patients with pulmonary, gastrointestinal, and endocrine problems made up the IMW population. When appropriate selective analysis of the general medical patients alone was performed, admissions to the IMWs for pulmonary conditions remained higher than in the Type IV program, whereas gastrointestinal and endocrine conditions became comparable.
Finally, data collection occurred in 1994, when lengths of hospitalization were longer than today. While this limitation may decrease the generalizability to hospital stays in 1999, this factor does not change the patient characteristics that influenced outcome. Efficiencies in medical care have decreased the total lengths of stay in most hospitals. Despite shorter stays, patients with comorbid medical and psychiatric illness remain a subset of patients who use resources at a higher level than those without.
|
Summary
|
This study suggests that more than 1% of patients admitted to a general hospital may benefit from integrated medical and psychiatric services in the medical setting. Involvement of both internal medicine and psychiatry facilitates a high-acuity environment that increases the value of patient care. The Type IV integration model provides care to a group of patients who are difficult to treat in the general medicine setting and the general psychiatry setting. By adding the capabilities inherent in the Type IV program, patients can be expected to have comparable improvement in medical illness and greater improvement of psychiatric illness than that provided on IMWs. Yet patients have similar lengths of stay even when a more severe psychiatric illness is present. Such treatment reduces the cost of care for these complicated patients, since concurrent rather than consecutive care can be provided.
Type IV programs introduce a new way of addressing the needs of patients with comorbid illness in the primary care inpatient setting. These programs are unlike the psychosomatic units of the past or even some current "psychiatric medicine units," many of which have closed or are closing, because they focus on the treatment of illnesses of insufficient severity to warrant inpatient or specialized care.
Despite inherent advantages, a limited number of health care organizations have adopted the integration model. As purchasers, insurers, and providers better recognize the impact that psychiatric comorbidity has on clinical, social, and economic outcomes, this model will take on greater importance in the changing health care environment.
|
ACKNOWLEDGMENTS
|
The authors thank Jon Cooney, M.D., for assistance with patient recruitment and Bonnie Eicher for help with manuscript preparation.
|
REFERENCES
|
-
Kathol RG, Harsch HH, Hall RCW, et al: Categorization of types of medical/psychiatry units based on level of acuity. Psychosomatics 1992; 33:376–386[Abstract/Free Full Text]
-
Fogel B: A psychiatric unit becomes a psychiatric/medical unit: administrative and clinical implications. Gen Hosp Psychiatry 1985; 7:26–35[Medline]
-
Harsch HH, Koran LM, Young LD: A profile of academic medical-psychiatric units. Gen Hosp Psychiatry 1991; 13:91–95
-
Kathol RG, Krummel S, Shakespeare A: Psychiatry and internal medicine resident education in an acute care medical-psychiatry unit. Gen Hosp Psychiatry 1989; 11:23–30[Medline]
-
Knaus WA, Wagner DP, Draper EA, et al: The APACHE-III prognostic system. Risk prediction of hospital mortality for critically ill hospitalized adults. Chest 1991; 100:1619–1636
-
American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th Edition. Washington, DC, American Psychiatric Association, 1994
-
Yates JW, Chalmer B, McKegney FP: Evaluation of patients with advanced cancer using the Karnofsky Performance Scale. Cancer 1980; 45:2220–2224
-
Stoudemire A, Kahn M, Brown JT, et al: Masked depression in a combined medical-psychiatric unit. Psychosomatics 1985; 26:221–228[Abstract/Free Full Text]
-
Hoffman RS: Diagnostic errors in the evaluation of behavioral disorders. JAMA 1982; 248:964–967[Abstract/Free Full Text]
-
Molnar G, Fava GA, Zielenzny MA: Medical-psychiatric unit patients compared with patients in two other services. Psychosomatics 1985; 26:193–209[Abstract/Free Full Text]
-
Young LD, Harsch HH: Length of stay on a psychiatric-medicine unit. Gen Hosp Psychiatry 1989; 11:31–35[Medline]
-
Swenson JR, Mai FM: A Canadian medical-psychiatric inpatient service. Can J Psychiatry 1992; 37:326–334[Medline]
-
Gertler R, Kopec-Schrader EM, Blackwell CJ: Evolution and evaluation of a medical psychiatric unit. Gen Hosp Psychiatry 1995; 17:26–31[Medline]
-
Fogel BS: Med-Psych units. Financial viability and quality assurance. Gen Hosp Psychiatry 1989; 11:17–22[Medline]
-
Fava GA, Wise TN, Molnar G, et al: The medical-psychiatric unit: a novel psychosomatic approach. Psychother Psychosom 1985; 43:194–201[Medline]
-
Summergrad P: Medical psychiatry units and the roles of the inpatient psychiatric service in the general hospital. Gen Hosp Psychiatry 1994; 16:20–31[Medline]
-
Fava GA: Medical Psychiatric service. Psychother Psychosom 1987; 48:96–100[Medline]
-
Harsch HH, LeCann AF, Ciaccio S: Treatment in combined medical psychiatry units: an integrative model. Psychosomatics 1989; 30:312–317[Abstract/Free Full Text]
-
Goldberg RJ, Daly J, Golinger RC: The impact of psychiatric comorbidity on Medicare reimbursement for inpatient medical care. Gen Hosp Psychiatry 1994; 16:16–19[Medline]
-
Mayou R, Hawton K, Feldman E: What happens to medical patients with psychiatric disorder? J Psychosom Res 1988; 32:541–549[Medline]
This article has been cited by other articles:
|
|
|
|
 
M Kannabiran, S Deshpande, A Walling, J Alagarsamy, D Protheroe, and P Trigwell
Cross-sectional survey of disturbed behaviour in patients in general hospitals in Leeds
Postgrad. Med. J.,
August 1, 2008;
84(994):
428 - 431.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
C. P. Carney and L. E. Jones
Medical comorbidity in women and men with bipolar disorders: a population-based controlled study.
Psychosom Med,
September 1, 2006;
68(5):
684 - 691.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. S. Rubin, B. Littenberg, R. Ross, S. Wehry, and M. Jones
Effects on Processes and Costs of Care Associated With the Addition of an Internist to an Inpatient Psychiatry Team
Psychiatr Serv,
April 1, 2005;
56(4):
463 - 467.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
G. Schneider, M. Wachter, G. Driesch, A. Kruse, H.-G. Nehen, and G. Heuft
Subjective Body Complaints as an Indicator of Somatization in Elderly Patients
Psychosomatics,
April 1, 2003;
44(2):
91 - 99.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
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]
|
|
|
|
|
|
|
Y. Kishi, R. G. Robinson, and J. T. Kosier
Suicidal Ideation Among Patients With Acute Life-Threatening Physical Illness: Patients With Stroke, Traumatic Brain Injury, Myocardial Infarction, and Spinal Cord Injury
Psychosomatics,
October 1, 2001;
42(5):
382 - 390.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. Stoudemire
Integrating Medical and Psychiatric Treatment in an Inpatient Medical Setting
Psychosomatics,
August 1, 2000;
41(4):
366 - 367.
[Full Text]
|
|
|
Get information about faster international access.
Privacy Policy
Copyright © 1999
Academy of Psychosomatic Medicine.
All rights reserved.
Home
| Search
| Current Issue
| Past Issues
| Subscribe
| All APPI Journals
| Help
| Contact Us
|
TOP
ABSTRACT
INTRODUCTION
