Vol. XXIV, No. 5 Printed in U S A .
JOURNAL OF THE AMERICAN GERIATRICS SOCIETY Copyright 0 1976 by the American Geriatrics Society
Comparison of Two Systems for Strokc Rehabilitation in a General Hospital B. CAIRBRE M c C A ” , MD* and RICHARD A. CULBERTSON, MHA**
Rhode Island Hospital, Providence, Rhode Island ABSTRACT: This study was designed to compare the effectiveness of stroke rehabilitation therapy in a specialized Stroke Unit with that provided on the medical service of a general hospital (Rhode Island Hospital). The &bed Stroke Unit is staffed by a multidisciplinary team, and a weekly conference is held for evaluation and planning. On the basis of data obtained from the hospital records, two groups of patients were studied: 224 who were treated in the Stroke Unit, and 110 who were evaluated and approved for admission to the Unit but were not accommodated. A rigid “first come, fvst served” policy for admission to the Unit was o b served. Hypothesis testing was performed with reference to the patient’s medical condition, socioeconomic status, demographic characteristics, and difficulties during hospital stay to determine whether the groups were comparable. A patient was considered to have improved if his condition decreased in severity between the time of admission to therapy and the time of discharge. Severity was rated as: mild (level one), moderate (level two), severe (level three), and profound (level four). No significant difference in rehabilitation results was found between the two treatment systems at severity levels two (moderate) and four (profound). However, the Stroke Unit attained significantly better results with level-three patients (severe stroke). This group received more sessions of physical therapy and remained in the hospital longer than did the level-three patients treated on the general medical service. Physicians referred patients selectively to the Stroke Unit, although the Unit had no policy of screening patients for admission, and this may have had some influence on the achievement of better results with level-three patients. Levelfour patients did not do well in either setting. Stroke is the third leading cause of death in the United States and the greatest long-term crippler. It is estimated that over 200,000 persons die of stroke each year, accounting for 11 percent of all deaths in this country (1). Although the acute phase of stroke has captured attention as a public health problem, the chronic phase and its consequences are just as significant. More than 2 million people now alive have survived this illness, with or without residual impairment of function. Through programs of rehabilitation,
many of these patients may be returned to satisfactory lives in their communities. This in turn may avoid admission to an institution for longterm care, with the attendant loss of human resources. The study presented here sought to evaluate comparatively the effects upon patients of two systems for delivery of stroke rehabilitation care. In our general hospital, one group of stroke patients received treatment in the general medical units, whereas the others received treatment in a specialized Stroke Unit. Through research design, comparable samples of patients were obtained from the two distinct settings. The two key questions in this study are: What effects do stroke rehabilitation programs have
* Director, Department of Rehabilitation Medicine.
** Correspondence to be addressed to: Richard A. Culbertson MHA, Assistant Vice President, m o d e Island Hospital, 593 Eddy Street, Providence, RI 02902.
211
McCANN AND CULBERTSON
upon outcome? What differences in the treatment programs lead to these results? BACKGROUND STATISTICS The study was conducted a t Rhode Island Hospital, a 730-bed voluntary general hospital affiliated with Brown University. Large numbers of patients are admitted to the hospital with the diagnosis of stroke. For purposes of the study, the term stroke is defined as “any disease in which one or more of the blood vessels of the brain is primarily implicated in a pathological process” (2). Hospital medical records indicated that during 1971, 435 patients were admitted to the institution with a primary diagnosis of stroke, and 439 with stroke as a secondary diagnosis. Many cases fall into the category of “transient ischemic attack” or “small stroke,” and the patients require no more therapy than admission for observation and follow-up by a physician after discharge. Others fail to survive the acute phase of the disease and die shortly after admission. During 1971, 59 of the patients admitted to the hospital with stroke as a primary diagnosis died within the first 10 days, and 32 with stroke as a secondary diagnosis died during this period. In the same year, 90 patients of the primary group were discharged as improved within 10 days, as were 103 of the secondary group. These patients account for less than half of the total group of admissions. The others may remain in the hospital because of continuing problems in medical management or for rehabilitation activities. There are 2 full-time specialists in the area of rehabilitation medicine and 10 physical therapists for the treatment of adults. This staffing pattern has remained consistent for the last three years, a period encompassing the time of the study. This staff treats patients a t the request of their physicians, either in the department of rehabilitation medicine or at the bedside. THE STROKE UNIT In an attempt to improve the pattern of delivery of rehabilitation services to stroke patients, a separate Stroke Unit was formed in February 1970. It was defined a t that time as a “geographic location within the hospital designated for stroke and stroke-like patients who are in need of rehabilitation services and skilled professional care that a regular unit cannot provide.” It was proposed that these services could be provided by
212
VOl.XXIV
a multidisciplinary team consisting of a physiatrist, neurologist, nurses, physical therapists, social workers and a speech pathologist in cooperation with the primary care physician. The Unit was regarded as an alternative means of providing services to patients through a different organizational form. No services not already available to stroke patients in other areas of the hospital were proposed. Similarly, no additional personnel were added to the staff already available. Patients continued to be seen only upon request of a private physician, and transfers to the Unit could be made only with his approval. During the patient’s stay on the Unit the private physician continued to be responsible for the medical management of the patient and for the final decision concerning discharge. This is not to imply, however, that such an organizational change is not without consequences. Upon admission to the Unit, consultations with rehabilitation medicine, social service and speech pathology are obtained routinely rather than on request. In addition, a program of physical therapy is set up for each patient, as determined by the therapist and physiatrist. A major difference is also apparent in the provision of nursing services. The program for the Stroke Unit gives the assigned nursing staff an opportunity to concentrate upon stroke patients and their specific nursing problems. An interdisciplinary weekly conference is convened to review the management and evaluate the progress of each patient during his stay, and to make recommendations to the private physician. This group also performs discharge planning for the patient in a methodical way. Severity ratings (Table 1) are assigned to the patient upon entering the unit and updated weekly as a measurement of progress. The ratings are based upon the “PULSES” profile, a graphic measure of disabilities which displays the patient’s overall clinical state in easily understood form. The profile evaluates the patient in terms of Physical condition, Upper and Lower limb function, Sensory function for vision and communication, Excretory functions, and Psychologic status (2). The Unit consists of 8 beds, in two 4-bed rooms on the ninth floor in the main building at Rhode Island Hospital. One room is set aside for male patients, and the room directly across the hall is for the female patients. Requirements for admission are that the medical condition of the patient be stable and that he not be comatose. He or she
May 1976
COMPARISON OF STROKE REHABILITATION SYSTEMS TABLE 1 Stroke Seuerity Ratings
1. Mild 2 . Moderate
3 . Severe .
4. Profound
Patient has minor disabilities that place slight restrictions on his ability to function at home and in the community, but he can live independently. Patient has moderate disabilities that place restrictions on his ability to function at home and in the community, but he can function with the assistance of another person, although this assistance is required only on occasion. Patient has severe disabilities that limit his ability to function at home and in the community, except with supervision. Patient is completely dependent upon the assistance of at least one person to carry out daily needs, and is not functionally competent in the home or in the community. -
must also be seen by a neurologist and physiatrist before admission. Admission to the Unit takes place under a first-come, first-served policy. Thus no overt attempt is made to exclude more difficult cases in favor of cases with a more favorable prognosis. Differences in the treatment of patients in the Stroke Unit and those in general medical areas exist principally in the system by which rehabilitative care is delivered. Both groups continue under the primary management of their private physicians during rehabilitation; as a consequence, general medical management may be considered equivalent. All patients in both groups receive physical therapy treatment during hospitalization. Against these similarities are differences in treatment and environment. The primary difference in treatment is the adoption by the Stroke Unit staff of an aggressive philosophy of rehabilitation involving the use of physical therapy on a twice-daily basis if the patient’s condition permits. This approach is not necessarily duplicated for the medical-ward stroke patients whose physical therapy continues to be ordered by the private physician and may well be on a less intensive basis. A second major difference is partially therapeutic, partially environmental. This is the specialization of personnel in dealing with stroke in the Stroke Unit. Here specialized nursing care is available, with the nursing staff assigned solely to this Unit. In contrast, patients treated in other areas of the hospital are cared for by personnel who are engaged in the care of a variety of disorders. As a consequence, their familiarity with stroke may not be great. Acutely ill patients will receive more of the nurses’ attention than stroke patients in the rehabilitation phase of their illness. Functional assessment and active stimulation in the activities of daily living are rarely evident in the day-to-day concerns of general medical nursing. In the Stroke Unit, these are of daily concern. The patient’s residual abilities in
the elements of daily living (e.g., basic mobility in and around the bed, feeding, dressing, grooming and toileting) are measured, and realistic independence goals are set. Within the framework of these goals, a realistic physical restorative program can be established immediately. The nurse/patient relationship involves much more nurse/patient contact than might otherwise be the case. The outcome in a case of stroke hinges critically on family involvement. This may range from an expression of interest or concern for the patient, to willingness or ability to receive the hemiplegic patient in a home family setting. The general tempo of the hospital offers little opportunity for family education in the many ramifications of such a disability as stroke, or for any involvement in care. In the case of the Stroke Unit patient, the relationship with family is a special one. Responsible family members observe the various elements of care, ranging from feeding to physical therapy and methods of transferring and walking. When the patient has progressed sufficiently, the family members are taught techniques of care so that they can manage the patient. Family involvement of this kind is as much the staffs concern as any other part of the care plan. Such an experience results in some discharges home which otherwise would have been unlikely. The social worker’s task of interpretation and support for the family becomes easier. Families and patients also understand that the key resources of the speech therapist, physical therapist, physiatrist and social worker will be available after discharge if necessary. This process of family involvement is not available in the case of the stroke patient treated in the general medical wards where an organized system is not in,use. A final difference is the environmental one of a “therapeutic community” made up of patients with similar disorders and undergoing a similar c o m e of therapy. This setting may enhance
213
McCANN AND CULBERTSON
patient morale and motivation, as has been shown by Abramson et a1 (3). FINDINGS In undertaking this study, we wished to compare the results achieved in rehabilitation of stroke patients by two different treatment systems. It is hoped that the findings of this study may be of benefit in adding to our knowledge about stroke rehabilitation and also be useful as a point of reference for future comparative studies in this complex area. Major findings of the study are presented here, with statistical documentation provided in Table 2.
Comparability of the two groups of patients Chi-square and Student t-tests were performed on 16 variables suggested by the rehabilitation personnel and reports in the literature, to see if the Stroke Unit group could be considered equivalent to the group approved for admission to the Stroke Unit but treated elsewhere. It was found that the samples could be judged equivalent for all but one variable, viz, the severity rating at the time of evaluation for Stroke Unit admission. A difficult problem was created by the finding that the two groups showed significantly different distributions of patients as measured by the severity level. This was particularly troublesome because the severity rating (Table 1) is presumed to provide a comprehensive picture of the patient’s condition, taking into account medical status, function in movement and psychologic status (2). The group in the Stroke Unit had a TABLE 2 Summarv of Chi-Sauare Results in Selected Variables to Determine Adsociation of Outcome and Site of Rehabilitation Treatment, i.e., Stroke Unit or General Medical Wards Variable
Test Statistics
DF
N
P
Site of treatment Outcome-severity level two
X* =
3.04
1
90
P > .05
Site of treatment Outcome-severity level three
X* =
4.46
1
175
P < .05t
Site of treatment Outcome-severity level four
X* =
2.08
1
42
P > .10
?Site of treatment and outcome, severity level three, is significant at an alpha level of .05.
214
V O l . xxzv
disproportionate number of patients at severity level three (severely impaired). The general medical group had a greater-than-expected representation at level two (moderate) and level four (profound). Several possible explanations for this finding are proposed. These variations might be due in part to the natural course of stroke. Moderately impaired patients might recover quickly before entry to the Stroke Unit was possible, whereas profoundly impaired patients would fall into the general medical group and be unable to accept transfer to the unit for rehabilitation. Thus, severity level three would still include the patients who would require more extensive hospitalization while being able to gain from rehabilitation treatment. Another possible explanation is that physicians refer patients selectively to the Stroke Unit and that these physicians screen out level-two and level-four cases. The successful outcomes achieved by the Unit with the levelthree patients reinforce this idea. The medical staff may well have become aware of the good results, on an informal basis.
Comparison of outcomes
A significant difference was observed in the results attained with level-three patients. The Stroke Unit had better results with this group, which constituted a majority of patients treated on the Unit. No significant difference was found between the outcomes achieved by the Stroke unit and the general medical service when patients of all severity levels were combined. This also was true when level-two and level-four patients were considered separately from the others. The success with the level-two cases in either treatment setting strengthens McCann’s assertion that some patients require little assistance in order to recover (4). In such cases, improvement to level one or to complete independence in daily living is defined as a successful outcome. Moskowitz et a1 (5) also have stated that a considerable number of stroke patients improve spontaneously with little effort on the part of rehabilitation personnel. Thus, although similar good results are achieved with the Stroke Unit patients, the expenditure of resources to produce this effect is greater and perhaps unnecessary. The Unit’s efforts may be geared more toward the severely impaired patient who appears to benefit most from such treatment rather than toward the less impaired patient. The level-two patient may be
May 1976
COMPARISON OF STROKE REHAl3ILITATION SYSTEMS
more analogous to the general medical patient with limited physical-therapy needs than to the candidate for intensive rehabilitation. A significantly better result was obtained with the level-three patients. The Stroke Unit group showed a more favorable outcome a t this level than did the group on the general medical wards. However, a number of factors entered into this result. The first was the significantly greater representation of level-three patients in the Stroke Unit group than would be expected by chance. It is possible that some “specialization” had occurred in the Unit because the number of these cases seen by the personnel had increased their ability to care for such patients. If so, it represents the achievement of one of the objectives of the Unit. It was also suspected that physicians were chiefly referring level-three patients to the Stroke Unit for therapy. This might be the result of either their own preconceptions concerning the patients for whom the Unit is of benefit, or their knowledge of successes achieved by the Unit with these patients. Level-three Stroke Unit patients remain in the hospital for a longer period than do level-three patients on the general medical service, and also undergo more sessions of physical therapy. Probably these factors also influence the outcome. For level-three patients, success entailed advancement from a phase in which they require substantial assistance in order to function independently, to a phase in which they require only minimal assistance (level two). An advance of this sort is important, since it enhances the possibility that the patient can continue life in the community rather than in an institution. Patients who advance to level two can live at home with less effort contributed by family and friends, whereas those who remain at level three require major assistance from these persons if they return to the community. The outcome for level-four patients (profoundly impaired and totally unable to function without assistance at all times) showed little variation between the two systems of rehabilitation. Interestingly, it is at this level that there appear to be significant differences in the commitment of treatment resources between the systems. The importance of this finding is somewhat obscured by the small size of the samples in the study. Nevertheless, the indications are that favorable outcomes with level-four patients are not achieved in either setting. This may be due to the initial severity of the illness and a deterioration of
function that makes rehabilitation exceedingly difficult. Moreover, these patients may be regarded as so severely impaired that the staff members “give up’’ in the matter of treatment. The former explanation appears more tenable, for although the number of physical therapy sessions received by this group is not as great as the number received by level-three patients, the mean number of hospital days for level-four patients exceeds those for the other two levels. This may indicate that the patient’s medical condition is not sufficiently stable to permit intensive rehabilitation procedures. If so, it would tend to support Wylie’s assertion (6) that patients with poorer scores on admission to a rehabilitation program make less progress than those with better admission scores. IMPLICATIONS OF FINDINGS Centralization of services for stroke rehabilitation in a Stroke Unit appears to be a better organization for delivery of these services to a certain stratum of patients than does the system of rehabilitation provided in the hospital’s regular medical wards. The patients who benefit from such services are those at level three (severely impaired) of the severity rating at the time of evaluation for admission to the Stroke Unit. This evaluation occurs after the acute phase of the stroke has passed, usually about the sixth hospital day. The greater number of sessions of physical therapy and the increased length of hospital stay offered to level-three patients by the Stroke Unit may be influential in achieving a better outcome with these patients. For stroke patients at levels two and four, there is no difference in rehabilitation outcome between those treated in the centralized organization of the Stroke Unit and those treated in the decentralized medical service. Level-two patients appear to do well in either rehabilitation setting, and show no better rate of improvement in the Stroke Unit despite the more intensive commitment of resources to their treatment. Level-four patients do poorly in either rehabilitation setting. An informal screening system seems to exist among physicians, to provide appropriate candidates for treatment in the Stroke Unit. The formal requirement for admission to such a Unit is that the patient’s condition be stable and that he be evaluated by a neurologist and a physiatrist. Yet referrals of level-three patients outnum-
215
V0l.XXIV
McCANN AND CULBERTSON
ber all other referrals combined, indicating that physicians in the hospital acknowledge a particular role for the Stroke Unit. The course of the disease process further removes level-two patients from the Stroke Unit candidates because of early discharge. It also removes level-four stroke patients through deterioration of their medical condition to the point where they are no longer eligible for transfer. It is significant that this referral system is not a result of decisions and policies made by the committee which oversees the Stroke Unit. The Unit in this case is dependent upon private physicians and the house staff for referrals, and the initial decision to refer or not to refer to the Unit always lies with these physicians. In the literature are many reports showing that specialized facilities make policies for acceptance of patients which will enhance the success of their program and provide for the most efficient use of their services. This is not the case for the Stroke Unit, for collaborative decisions are being made continually in the referral of individual patients. The important finding is that such decisions are made by the persons who manage the comprehensive treatment of the patient together with the persons providing such specialized care as is found in the Stroke Unit. The patient’s primary care physician has the greatest influence in deciding whether this resource is to be used. These perceptions of appropriate cases for treatment in the Stroke Unit may later be formalized by a specific policy on admission or rejection for specialized services, although this has not yet occurred in the case of the Rhode Island Hospital Stroke Unit. This development probably follows a general model for the development of special care units, the roles of which are informally defined on the basis of their performance. An example is the evolution of the role of the coronary care unit from one originally designated to deal with the most severe cases of heart disease. As these units developed, it was found that their services were most appropriate for cases of mild cardiac infarction. These patients were able to benefit from the emergency services in these units, whereas the
216
patients for whom the units were originally intended, were not. COMMENT Medical care resources in American society are limited, and consequently should be allocated with maximal benefit for the greatest number of persons requiring such services. The study presented here indicates that a Stroke Unit for the intensive rehabilitation of stroke patients is of benefit in delivering these services to a definite group of persons afflicted with this disease. The services it provides exceed those provided by the general medical care of stroke patients at Rhode Island Hospital, regardless of the level of impairment. It may be that such intensive rehabilitation units should serve primarily as referral centers for the severely impaired, drawn from a regional area to form a pool of candidates who would benefit from these units. In this manner, these services would become available to persons in smaller hospitals, particularly in rural areas where hospitals cannot justify the operation of a unit of this type. A teaching hospital seems a logical place for the location of such a unit. As Wylie (7) has said: “Since intensive rehabilitation efforts are costly and personnel are scarce, a critical first step to improve existing programs may be to ensure that the care reaches those who will benefit most.” REFERENCES 1. U. S. Senate Appropriations Committee: Title 11, Depart-
2. 3. 4.
5.
ment of Health, Education, and Welfare. Washington, DC, U. S. Government Printing Office, 1974, p 5-26. Moskowitz E and McCann BC: Classification of disability in the chronically ill and aging, J Chron Dis 5: 342, 1957. Abramson A, Kutner R, Rosenberg P et al: A therapeutic community in a general hospital: adaptation to a rehabilitation service, J Chron Dis 16: 179, 1963. McCann BC: Post stroke rehabilitation, Rhode Island Med J 52: 23, 1969. Moskowitz E, Lightbody FE and Freitag NS: Long-term follow-up of the poststroke patient, Arch Phys Med Rehab
53: 167, 1972. 6. Wylie CM: Gauging the response of stroke patients to rehabilitation, J Am Geriatrics SOC15: 797, 1967. 7. Wylie CM: Rehabilitative care of stroke patients, JAMA 196: 1117, 1966.
JOURNAL OF THE AMERICAN GERIATRICS SOCIETY Copyright 0 1976 by the American Geriatrics Society
Comparison of Two Systems for Strokc Rehabilitation in a General Hospital B. CAIRBRE M c C A ” , MD* and RICHARD A. CULBERTSON, MHA**
Rhode Island Hospital, Providence, Rhode Island ABSTRACT: This study was designed to compare the effectiveness of stroke rehabilitation therapy in a specialized Stroke Unit with that provided on the medical service of a general hospital (Rhode Island Hospital). The &bed Stroke Unit is staffed by a multidisciplinary team, and a weekly conference is held for evaluation and planning. On the basis of data obtained from the hospital records, two groups of patients were studied: 224 who were treated in the Stroke Unit, and 110 who were evaluated and approved for admission to the Unit but were not accommodated. A rigid “first come, fvst served” policy for admission to the Unit was o b served. Hypothesis testing was performed with reference to the patient’s medical condition, socioeconomic status, demographic characteristics, and difficulties during hospital stay to determine whether the groups were comparable. A patient was considered to have improved if his condition decreased in severity between the time of admission to therapy and the time of discharge. Severity was rated as: mild (level one), moderate (level two), severe (level three), and profound (level four). No significant difference in rehabilitation results was found between the two treatment systems at severity levels two (moderate) and four (profound). However, the Stroke Unit attained significantly better results with level-three patients (severe stroke). This group received more sessions of physical therapy and remained in the hospital longer than did the level-three patients treated on the general medical service. Physicians referred patients selectively to the Stroke Unit, although the Unit had no policy of screening patients for admission, and this may have had some influence on the achievement of better results with level-three patients. Levelfour patients did not do well in either setting. Stroke is the third leading cause of death in the United States and the greatest long-term crippler. It is estimated that over 200,000 persons die of stroke each year, accounting for 11 percent of all deaths in this country (1). Although the acute phase of stroke has captured attention as a public health problem, the chronic phase and its consequences are just as significant. More than 2 million people now alive have survived this illness, with or without residual impairment of function. Through programs of rehabilitation,
many of these patients may be returned to satisfactory lives in their communities. This in turn may avoid admission to an institution for longterm care, with the attendant loss of human resources. The study presented here sought to evaluate comparatively the effects upon patients of two systems for delivery of stroke rehabilitation care. In our general hospital, one group of stroke patients received treatment in the general medical units, whereas the others received treatment in a specialized Stroke Unit. Through research design, comparable samples of patients were obtained from the two distinct settings. The two key questions in this study are: What effects do stroke rehabilitation programs have
* Director, Department of Rehabilitation Medicine.
** Correspondence to be addressed to: Richard A. Culbertson MHA, Assistant Vice President, m o d e Island Hospital, 593 Eddy Street, Providence, RI 02902.
211
McCANN AND CULBERTSON
upon outcome? What differences in the treatment programs lead to these results? BACKGROUND STATISTICS The study was conducted a t Rhode Island Hospital, a 730-bed voluntary general hospital affiliated with Brown University. Large numbers of patients are admitted to the hospital with the diagnosis of stroke. For purposes of the study, the term stroke is defined as “any disease in which one or more of the blood vessels of the brain is primarily implicated in a pathological process” (2). Hospital medical records indicated that during 1971, 435 patients were admitted to the institution with a primary diagnosis of stroke, and 439 with stroke as a secondary diagnosis. Many cases fall into the category of “transient ischemic attack” or “small stroke,” and the patients require no more therapy than admission for observation and follow-up by a physician after discharge. Others fail to survive the acute phase of the disease and die shortly after admission. During 1971, 59 of the patients admitted to the hospital with stroke as a primary diagnosis died within the first 10 days, and 32 with stroke as a secondary diagnosis died during this period. In the same year, 90 patients of the primary group were discharged as improved within 10 days, as were 103 of the secondary group. These patients account for less than half of the total group of admissions. The others may remain in the hospital because of continuing problems in medical management or for rehabilitation activities. There are 2 full-time specialists in the area of rehabilitation medicine and 10 physical therapists for the treatment of adults. This staffing pattern has remained consistent for the last three years, a period encompassing the time of the study. This staff treats patients a t the request of their physicians, either in the department of rehabilitation medicine or at the bedside. THE STROKE UNIT In an attempt to improve the pattern of delivery of rehabilitation services to stroke patients, a separate Stroke Unit was formed in February 1970. It was defined a t that time as a “geographic location within the hospital designated for stroke and stroke-like patients who are in need of rehabilitation services and skilled professional care that a regular unit cannot provide.” It was proposed that these services could be provided by
212
VOl.XXIV
a multidisciplinary team consisting of a physiatrist, neurologist, nurses, physical therapists, social workers and a speech pathologist in cooperation with the primary care physician. The Unit was regarded as an alternative means of providing services to patients through a different organizational form. No services not already available to stroke patients in other areas of the hospital were proposed. Similarly, no additional personnel were added to the staff already available. Patients continued to be seen only upon request of a private physician, and transfers to the Unit could be made only with his approval. During the patient’s stay on the Unit the private physician continued to be responsible for the medical management of the patient and for the final decision concerning discharge. This is not to imply, however, that such an organizational change is not without consequences. Upon admission to the Unit, consultations with rehabilitation medicine, social service and speech pathology are obtained routinely rather than on request. In addition, a program of physical therapy is set up for each patient, as determined by the therapist and physiatrist. A major difference is also apparent in the provision of nursing services. The program for the Stroke Unit gives the assigned nursing staff an opportunity to concentrate upon stroke patients and their specific nursing problems. An interdisciplinary weekly conference is convened to review the management and evaluate the progress of each patient during his stay, and to make recommendations to the private physician. This group also performs discharge planning for the patient in a methodical way. Severity ratings (Table 1) are assigned to the patient upon entering the unit and updated weekly as a measurement of progress. The ratings are based upon the “PULSES” profile, a graphic measure of disabilities which displays the patient’s overall clinical state in easily understood form. The profile evaluates the patient in terms of Physical condition, Upper and Lower limb function, Sensory function for vision and communication, Excretory functions, and Psychologic status (2). The Unit consists of 8 beds, in two 4-bed rooms on the ninth floor in the main building at Rhode Island Hospital. One room is set aside for male patients, and the room directly across the hall is for the female patients. Requirements for admission are that the medical condition of the patient be stable and that he not be comatose. He or she
May 1976
COMPARISON OF STROKE REHABILITATION SYSTEMS TABLE 1 Stroke Seuerity Ratings
1. Mild 2 . Moderate
3 . Severe .
4. Profound
Patient has minor disabilities that place slight restrictions on his ability to function at home and in the community, but he can live independently. Patient has moderate disabilities that place restrictions on his ability to function at home and in the community, but he can function with the assistance of another person, although this assistance is required only on occasion. Patient has severe disabilities that limit his ability to function at home and in the community, except with supervision. Patient is completely dependent upon the assistance of at least one person to carry out daily needs, and is not functionally competent in the home or in the community. -
must also be seen by a neurologist and physiatrist before admission. Admission to the Unit takes place under a first-come, first-served policy. Thus no overt attempt is made to exclude more difficult cases in favor of cases with a more favorable prognosis. Differences in the treatment of patients in the Stroke Unit and those in general medical areas exist principally in the system by which rehabilitative care is delivered. Both groups continue under the primary management of their private physicians during rehabilitation; as a consequence, general medical management may be considered equivalent. All patients in both groups receive physical therapy treatment during hospitalization. Against these similarities are differences in treatment and environment. The primary difference in treatment is the adoption by the Stroke Unit staff of an aggressive philosophy of rehabilitation involving the use of physical therapy on a twice-daily basis if the patient’s condition permits. This approach is not necessarily duplicated for the medical-ward stroke patients whose physical therapy continues to be ordered by the private physician and may well be on a less intensive basis. A second major difference is partially therapeutic, partially environmental. This is the specialization of personnel in dealing with stroke in the Stroke Unit. Here specialized nursing care is available, with the nursing staff assigned solely to this Unit. In contrast, patients treated in other areas of the hospital are cared for by personnel who are engaged in the care of a variety of disorders. As a consequence, their familiarity with stroke may not be great. Acutely ill patients will receive more of the nurses’ attention than stroke patients in the rehabilitation phase of their illness. Functional assessment and active stimulation in the activities of daily living are rarely evident in the day-to-day concerns of general medical nursing. In the Stroke Unit, these are of daily concern. The patient’s residual abilities in
the elements of daily living (e.g., basic mobility in and around the bed, feeding, dressing, grooming and toileting) are measured, and realistic independence goals are set. Within the framework of these goals, a realistic physical restorative program can be established immediately. The nurse/patient relationship involves much more nurse/patient contact than might otherwise be the case. The outcome in a case of stroke hinges critically on family involvement. This may range from an expression of interest or concern for the patient, to willingness or ability to receive the hemiplegic patient in a home family setting. The general tempo of the hospital offers little opportunity for family education in the many ramifications of such a disability as stroke, or for any involvement in care. In the case of the Stroke Unit patient, the relationship with family is a special one. Responsible family members observe the various elements of care, ranging from feeding to physical therapy and methods of transferring and walking. When the patient has progressed sufficiently, the family members are taught techniques of care so that they can manage the patient. Family involvement of this kind is as much the staffs concern as any other part of the care plan. Such an experience results in some discharges home which otherwise would have been unlikely. The social worker’s task of interpretation and support for the family becomes easier. Families and patients also understand that the key resources of the speech therapist, physical therapist, physiatrist and social worker will be available after discharge if necessary. This process of family involvement is not available in the case of the stroke patient treated in the general medical wards where an organized system is not in,use. A final difference is the environmental one of a “therapeutic community” made up of patients with similar disorders and undergoing a similar c o m e of therapy. This setting may enhance
213
McCANN AND CULBERTSON
patient morale and motivation, as has been shown by Abramson et a1 (3). FINDINGS In undertaking this study, we wished to compare the results achieved in rehabilitation of stroke patients by two different treatment systems. It is hoped that the findings of this study may be of benefit in adding to our knowledge about stroke rehabilitation and also be useful as a point of reference for future comparative studies in this complex area. Major findings of the study are presented here, with statistical documentation provided in Table 2.
Comparability of the two groups of patients Chi-square and Student t-tests were performed on 16 variables suggested by the rehabilitation personnel and reports in the literature, to see if the Stroke Unit group could be considered equivalent to the group approved for admission to the Stroke Unit but treated elsewhere. It was found that the samples could be judged equivalent for all but one variable, viz, the severity rating at the time of evaluation for Stroke Unit admission. A difficult problem was created by the finding that the two groups showed significantly different distributions of patients as measured by the severity level. This was particularly troublesome because the severity rating (Table 1) is presumed to provide a comprehensive picture of the patient’s condition, taking into account medical status, function in movement and psychologic status (2). The group in the Stroke Unit had a TABLE 2 Summarv of Chi-Sauare Results in Selected Variables to Determine Adsociation of Outcome and Site of Rehabilitation Treatment, i.e., Stroke Unit or General Medical Wards Variable
Test Statistics
DF
N
P
Site of treatment Outcome-severity level two
X* =
3.04
1
90
P > .05
Site of treatment Outcome-severity level three
X* =
4.46
1
175
P < .05t
Site of treatment Outcome-severity level four
X* =
2.08
1
42
P > .10
?Site of treatment and outcome, severity level three, is significant at an alpha level of .05.
214
V O l . xxzv
disproportionate number of patients at severity level three (severely impaired). The general medical group had a greater-than-expected representation at level two (moderate) and level four (profound). Several possible explanations for this finding are proposed. These variations might be due in part to the natural course of stroke. Moderately impaired patients might recover quickly before entry to the Stroke Unit was possible, whereas profoundly impaired patients would fall into the general medical group and be unable to accept transfer to the unit for rehabilitation. Thus, severity level three would still include the patients who would require more extensive hospitalization while being able to gain from rehabilitation treatment. Another possible explanation is that physicians refer patients selectively to the Stroke Unit and that these physicians screen out level-two and level-four cases. The successful outcomes achieved by the Unit with the levelthree patients reinforce this idea. The medical staff may well have become aware of the good results, on an informal basis.
Comparison of outcomes
A significant difference was observed in the results attained with level-three patients. The Stroke Unit had better results with this group, which constituted a majority of patients treated on the Unit. No significant difference was found between the outcomes achieved by the Stroke unit and the general medical service when patients of all severity levels were combined. This also was true when level-two and level-four patients were considered separately from the others. The success with the level-two cases in either treatment setting strengthens McCann’s assertion that some patients require little assistance in order to recover (4). In such cases, improvement to level one or to complete independence in daily living is defined as a successful outcome. Moskowitz et a1 (5) also have stated that a considerable number of stroke patients improve spontaneously with little effort on the part of rehabilitation personnel. Thus, although similar good results are achieved with the Stroke Unit patients, the expenditure of resources to produce this effect is greater and perhaps unnecessary. The Unit’s efforts may be geared more toward the severely impaired patient who appears to benefit most from such treatment rather than toward the less impaired patient. The level-two patient may be
May 1976
COMPARISON OF STROKE REHAl3ILITATION SYSTEMS
more analogous to the general medical patient with limited physical-therapy needs than to the candidate for intensive rehabilitation. A significantly better result was obtained with the level-three patients. The Stroke Unit group showed a more favorable outcome a t this level than did the group on the general medical wards. However, a number of factors entered into this result. The first was the significantly greater representation of level-three patients in the Stroke Unit group than would be expected by chance. It is possible that some “specialization” had occurred in the Unit because the number of these cases seen by the personnel had increased their ability to care for such patients. If so, it represents the achievement of one of the objectives of the Unit. It was also suspected that physicians were chiefly referring level-three patients to the Stroke Unit for therapy. This might be the result of either their own preconceptions concerning the patients for whom the Unit is of benefit, or their knowledge of successes achieved by the Unit with these patients. Level-three Stroke Unit patients remain in the hospital for a longer period than do level-three patients on the general medical service, and also undergo more sessions of physical therapy. Probably these factors also influence the outcome. For level-three patients, success entailed advancement from a phase in which they require substantial assistance in order to function independently, to a phase in which they require only minimal assistance (level two). An advance of this sort is important, since it enhances the possibility that the patient can continue life in the community rather than in an institution. Patients who advance to level two can live at home with less effort contributed by family and friends, whereas those who remain at level three require major assistance from these persons if they return to the community. The outcome for level-four patients (profoundly impaired and totally unable to function without assistance at all times) showed little variation between the two systems of rehabilitation. Interestingly, it is at this level that there appear to be significant differences in the commitment of treatment resources between the systems. The importance of this finding is somewhat obscured by the small size of the samples in the study. Nevertheless, the indications are that favorable outcomes with level-four patients are not achieved in either setting. This may be due to the initial severity of the illness and a deterioration of
function that makes rehabilitation exceedingly difficult. Moreover, these patients may be regarded as so severely impaired that the staff members “give up’’ in the matter of treatment. The former explanation appears more tenable, for although the number of physical therapy sessions received by this group is not as great as the number received by level-three patients, the mean number of hospital days for level-four patients exceeds those for the other two levels. This may indicate that the patient’s medical condition is not sufficiently stable to permit intensive rehabilitation procedures. If so, it would tend to support Wylie’s assertion (6) that patients with poorer scores on admission to a rehabilitation program make less progress than those with better admission scores. IMPLICATIONS OF FINDINGS Centralization of services for stroke rehabilitation in a Stroke Unit appears to be a better organization for delivery of these services to a certain stratum of patients than does the system of rehabilitation provided in the hospital’s regular medical wards. The patients who benefit from such services are those at level three (severely impaired) of the severity rating at the time of evaluation for admission to the Stroke Unit. This evaluation occurs after the acute phase of the stroke has passed, usually about the sixth hospital day. The greater number of sessions of physical therapy and the increased length of hospital stay offered to level-three patients by the Stroke Unit may be influential in achieving a better outcome with these patients. For stroke patients at levels two and four, there is no difference in rehabilitation outcome between those treated in the centralized organization of the Stroke Unit and those treated in the decentralized medical service. Level-two patients appear to do well in either rehabilitation setting, and show no better rate of improvement in the Stroke Unit despite the more intensive commitment of resources to their treatment. Level-four patients do poorly in either rehabilitation setting. An informal screening system seems to exist among physicians, to provide appropriate candidates for treatment in the Stroke Unit. The formal requirement for admission to such a Unit is that the patient’s condition be stable and that he be evaluated by a neurologist and a physiatrist. Yet referrals of level-three patients outnum-
215
V0l.XXIV
McCANN AND CULBERTSON
ber all other referrals combined, indicating that physicians in the hospital acknowledge a particular role for the Stroke Unit. The course of the disease process further removes level-two patients from the Stroke Unit candidates because of early discharge. It also removes level-four stroke patients through deterioration of their medical condition to the point where they are no longer eligible for transfer. It is significant that this referral system is not a result of decisions and policies made by the committee which oversees the Stroke Unit. The Unit in this case is dependent upon private physicians and the house staff for referrals, and the initial decision to refer or not to refer to the Unit always lies with these physicians. In the literature are many reports showing that specialized facilities make policies for acceptance of patients which will enhance the success of their program and provide for the most efficient use of their services. This is not the case for the Stroke Unit, for collaborative decisions are being made continually in the referral of individual patients. The important finding is that such decisions are made by the persons who manage the comprehensive treatment of the patient together with the persons providing such specialized care as is found in the Stroke Unit. The patient’s primary care physician has the greatest influence in deciding whether this resource is to be used. These perceptions of appropriate cases for treatment in the Stroke Unit may later be formalized by a specific policy on admission or rejection for specialized services, although this has not yet occurred in the case of the Rhode Island Hospital Stroke Unit. This development probably follows a general model for the development of special care units, the roles of which are informally defined on the basis of their performance. An example is the evolution of the role of the coronary care unit from one originally designated to deal with the most severe cases of heart disease. As these units developed, it was found that their services were most appropriate for cases of mild cardiac infarction. These patients were able to benefit from the emergency services in these units, whereas the
216
patients for whom the units were originally intended, were not. COMMENT Medical care resources in American society are limited, and consequently should be allocated with maximal benefit for the greatest number of persons requiring such services. The study presented here indicates that a Stroke Unit for the intensive rehabilitation of stroke patients is of benefit in delivering these services to a definite group of persons afflicted with this disease. The services it provides exceed those provided by the general medical care of stroke patients at Rhode Island Hospital, regardless of the level of impairment. It may be that such intensive rehabilitation units should serve primarily as referral centers for the severely impaired, drawn from a regional area to form a pool of candidates who would benefit from these units. In this manner, these services would become available to persons in smaller hospitals, particularly in rural areas where hospitals cannot justify the operation of a unit of this type. A teaching hospital seems a logical place for the location of such a unit. As Wylie (7) has said: “Since intensive rehabilitation efforts are costly and personnel are scarce, a critical first step to improve existing programs may be to ensure that the care reaches those who will benefit most.” REFERENCES 1. U. S. Senate Appropriations Committee: Title 11, Depart-
2. 3. 4.
5.
ment of Health, Education, and Welfare. Washington, DC, U. S. Government Printing Office, 1974, p 5-26. Moskowitz E and McCann BC: Classification of disability in the chronically ill and aging, J Chron Dis 5: 342, 1957. Abramson A, Kutner R, Rosenberg P et al: A therapeutic community in a general hospital: adaptation to a rehabilitation service, J Chron Dis 16: 179, 1963. McCann BC: Post stroke rehabilitation, Rhode Island Med J 52: 23, 1969. Moskowitz E, Lightbody FE and Freitag NS: Long-term follow-up of the poststroke patient, Arch Phys Med Rehab
53: 167, 1972. 6. Wylie CM: Gauging the response of stroke patients to rehabilitation, J Am Geriatrics SOC15: 797, 1967. 7. Wylie CM: Rehabilitative care of stroke patients, JAMA 196: 1117, 1966.
