858
Evaluating and Predicting Outcome of Acute
Cerebral Vascular Accident
HENRY G. DOVE, P H . D . , * KAREN C. SCHNEIDER, M.P.H.,* AND JAN D. WALLACE,
M.D.t
SUMMARY An innovative evaluation method is used to study the outcomes and clinical predictions for 97
patients with acute cerebral vascular accidents. The technique involved the participation of several professional disciplines in selecting baseline and treatment variables and making independent predictions about
the functional status of patients upon discharge from the stroke treatment center. The data suggest that (1)
baseline variables were more important than treatment variables in the participants' predictions about the
patients' short-term outcomes; (2) stroke unit staff members were generally successful in predicting
patients' functional status; and (3) stroke extensions and other complications are important factors which
affect stroke patients' short-term outcomes.
Stroke Vol 15, No 5, 1984
Downloaded from http://stroke.ahajournals.org/ by guest on July 28, 2017
THE EVALUATION of the quality of medical care for
stroke patients has involved a variety of approaches.
Some studies have compared stroke outcome in academic and community hospitals.1-2 Others have concentrated on examining the value of intensive care
units for acute stroke patients. M Cooper and associates
discuss the development of a combined heart-stroke
intensive care unit.9 Several investigators have evaluated inpatient rehabilitation units for patients with
completed stroke.]0~'2 A number of researchers have
been interested in predicting the outcome of acute
stroke. l3~16 Anderson and associates, using a modified
version of Williamson's outcomes-oriented assessment method,17 compared estimated outcomes with actual measured outcomes for stroke patients who participated in a rehabilitation program. 1819
This evaluation of the quality of medical care focuses on acute stroke patients cared for in a comprehensive stroke treatment center in a university-affiliated
Veterans Administration medical center.
The objectives of this study were 1) to analyze the
relationships among stroke patients' baseline characteristicst on admission, treatment variables and outcome; and 2) to compare the stroke unit staffs outcome predictions for patients with the patients' actual
outcomes.
The following research questions were addressed by
the study:
1. What are the achieved functional status outcomes for a sample of stroke patients?
2. Do any treatment variables relate to outcome?
3. Which kind of variables — baseline or treatment
— are most important in terms of the functional status
outcome?
4. Which staff member's outcome predictions most
closely match the stroke patients' actual outcomes?
{Baseline characteristics are patient attributes or clinical characteristics that describe the initial state of the patient on admission to the
hospital before any treatment is begun.
From Health Services Research Program,* and Stroke Treatment
Center, t VA Medical Center, West Haven, Connecticut. This project
was supported as part of VA HSR&D Project #519, "The Use of
Conditional Discriminators in Medical Audits."
Address correspondence to: Henry G. Dove, Ph.D., Health Services
Research Program, VA Medical Center, West Spring Street, West Haven, Connecticut 06516.
Received September 29, 1982; revision # 2 accepted February 29,
1984.
5. Do the outcome predictions differ among the
various medical disciplines represented by the stroke
center staff?
6. Do the treatment variables affect the staff's outcome predictions?
Methods
Patients were identified for possible inclusion in the
study from the daily report of admissions, discharges
and transfers at the West Haven Veterans Administration Medical Center (WHVA/MC) for the period September 6, 1977 through January 22, 1981. This report
enabled us to identify all one hundred seventy six (176)
patients who were admitted into the Stroke Treatment
Center during this period.8 Virtually all patients who
enter the WHVA/MC emergency room with suspected
stroke are admitted to the Stroke Treatment Center
shortly after their arrival at the hospital. Most patients
are admitted to the unit within 24 hours (mean: 6
hours) of the onset of symptoms. The average age of
the patients was 62.7 years, with a range of 36 to 92
years.
Seventy-nine (79) patients were excluded for the
following reasons: transient ischemic attack§ (38),
brain stem stroke (30), intracerebral hematoma with
coma (4), previous stroke with severe disability (4),
and subarachnoid hemorrhage (3). The study sample
consisted of 97 acute cerebral vascular accident patients, classified using a system similar to that adopted
by the ad hoc Committee on Cerebrovascular Disease
of the National Institute of Neurological and Communicative Disorders and Stroke.20
The Stroke Treatment Center at the WHVA/MC
includes a four-bed, acute-care area as well as a subacute area for intensive rehabilitation. Patients remain
in the acute care area until stabilized and are then
moved to the subacute area. The average total length of
stay in the Stroke Treatment Center was 29.5 days
(range: 1-120 days).
Throughout the study, a single outcome variable
was used: the functional status of the patient upon
discharge from the subacute area of the center. The
participants in the study believed that this measure,
§An additional 52 patients admitted to the Stroke Treatment Center
during the study period for diseases other than stroke. Transient ischemic attack (TIA) is defined in this study as an episode of a neurological
deficit which clears within 24 hours.21- n
EVALUATION AND PROGNOSIS OF STROKE/Dove et al
Downloaded from http://stroke.ahajournals.org/ by guest on July 28, 2017
although short-term in duration, was the single most
meaningful outcome variable for patients and staff
alike.
The study was conducted by the WHVA/MC Health
Services Research staff and included the participation
of two physicians, two nurses, a social worker and a
neuropsychologist from the Stroke Treatment Center.
After a review of the stroke literature and discussions
with the Stroke Treatment Center Director, a preliminary list of variables thought to affect recovery
from stroke were presented to the six stroke center
participants.
The Health Services Research staff abstracted these
variables from the medical records of five patients not
in the sample, and asked the six participants to make
predictions for each patient in this training sample in
terms of the outcome variable. Predictions were made
at two points: 1) upon admission into the acute area,
using only admission or baseline variables; and 2) after
a description of the patients' stay in the center, which
included treatment variables such as the administration
of aspirin on admission and complications. The participants were "blinded" as to other participants'
predictions.
The list of variables were revised, based on the
participants' reactions to the pilot study. The final variables adopted by the study participants are briefly
described in table 1.
Severity of stroke was a baseline variable which was
determined on admission to the Stroke Treatment Center and based on the first standardized neurological
assessment. The scoring system, which has been used
for more than ten years at the WHVA/MC, was determined heuristically to measure functional status, particularly relating to activities of daily living and ability
to function in the outside world. The scores are comprised of measures of Mental Status, Aphasic, Motor
Function, and Sensory Function. The scores of each
dimension are determined as follows:
1. Mental Status:
100 — fully awake
75 — awake, time disorientation or mild
confusion
50 — lethargic, or disorientation and
confusion
25 — responds to pain only
0 — coma
S — seizures
2. Aphasia:
a. Comprehension:
100 — understands language normally
75 — points to named or written objects; not
able to understand more complex
written language
50 — right-left orientation and three step
commands
25 — less than two step commands
0 — no comprehension, no imitation
b. Expression:
100 — normal speech
75 — speaks in phrases with hesitation
859
50 — names objects, no phrases
25 — says "yes/no" correctly and points to
objects
0 — no oral speech or ability to write
3. Motor Function Testing
100 — normal strength
75 — ability to hold limb elevated briefly
50 — ability to raise limb against gravity
25 — ability to move, but not raise limb
0 — paralysis
4. Sensory Function Testing:
a. Pin Stick
100 — normal
50 — altered
0 — absent
b. Position Sense:
100 — normal
75 — inaccurate appreciation of movement
at slow rate
50 — appreciation only with wide range or
rapid movement
25 — appreciation of movement at wrist but
not fingers
0 — absent
c. Double Simultaneous Stimulation:
(record as right = left, right over left, or
left over right)
Touch Localization:
100 — if right = left
75 — correct body part, but not as accurate
as contralateral "normal" side
50 — recognized as appropriate limb only
25 — recognize "correct" side only
0 — no touch recognition
5. Visual field examination, presence or absence
of:
Monocular deficit: Unable to count fingers in
any quadrant, one eye.
Homonymous deficit: Unable to count fingers
in one visual field or no response to visual
threat in one visual field.
This assessment was performed by the nursing staff
at admission and at regular intervals (every 2 hours
while patient awake) to document changes in neurological status, with subjective interpolation of subtle
change falling between the major grades.21 Important
changes were confirmed by the attending physician.
Appendix A provides additional information as to how
the Neurological Assessment Score was used to place
each patient into one of five severity categories:
Complication variables included stroke exacerbation/extension and "other complications." A stroke
exacerbation or extension was defined in this study as a
period of improvement or stabilization for at least 24
hours after the initial stroke, followed by a striking
deterioration in neurological status — a 30 percent or
greater drop in mental status, language or motor scores
on the standardized neurological test. Our definition of
860
STROKE
TABLE 1 Key Baseline, Treatment, Complication and Outcome
Variables Adopted by Study Participants
Downloaded from http://stroke.ahajournals.org/ by guest on July 28, 2017
Baseline, Treatment, Complication and Outcome Variables
Baseline
1. Severity of stroke at first neurological assessment (Classes IIV) (see Appendix A)
2. Age
3. Previous stroke with mild deficit
4. Presence of heart disease
5. Presence of hypertension
6. Presence of diabetes mellitus
7. Living arrangement at admission (home/other)
8. Onset of stroke ( a 24hours, < 24 hours, several hours before
admission)
Treatment
1. Placed on aspirin at admission (650 mgm BID)
2. Placed on low dose heparin at admission (2500-5000 u BID
SQ)
3. Hypertensive medications given when DBP > 110 mm Hg
4. Restriction of fluids when evidence of hyponatremia
5. Fluids administered when evidence of hypernatremia
6. Transfusion given when hematocrit s 30
7. Phlebotomy done when hematocrit a 50
Complication
1. Stroke exacerbation or extension (stabilized, then > 30%
drop in mental status, language or motor scores on the neurological test over 24 hours after the first neurological assessment)
2. Other complications (pneumonia, pulmonary edema, thrombophlebitis, urinary tract infection, gastrointestinal bleeding,
decubiti or contractures)
Outcome
Functional status at discharge from the Stroke Treatment Center:
1. ambulatory and independent in Activities of Daily Living
(ADL)
2 ambulatory and needs some help with ADL
3. ambulatory with personal assistance and ADL with assistance
4 not walking at all (wheelchair bound) and ADL with assistance
5. total care required
6. deceased
a stroke extension is less restrictive than a "late exacerbation," a term defined by Jones and Millikan22
which requires 48 hours of stability or improvement
before consideration of a new event. We required some
period of stability or a measure of improvement to
separate the stroke extension from stroke in evolution.
"Other complications" included pneumonia, pulmonary edema, thrombophlebitis, urinary tract infection, gastrointestinal bleeding, decubiti or
contractures.
The variables listed in table 1 were abstracted from
the medical records of the 97 patients in the sample.
The six participants predicted functional status outcomes for these patients as they did in the pilot study.
VOL 15, No 5, SEPTEMBER-OCTOBER
1984
Each of the participants was given the appropriate data
for each patient in the study in the same order. They
recorded their functional status predictions for the patients using the six-point scale shown in table 1. Their
predictions were made blindly, i.e., the participants
were given no information about the others' predictions or the patients' identities. A reliability check on
the participants was included as a part of the predicting
activity.
Results
Baseline, Treatment and Outcome Relationships
The achieved functional status outcomes for the
sample of stroke patients at discharge from the Stroke
Treatment Center are shown in table 2. Approximately
half of the patients had a highly successful outcome
after stroke — ambulatory and independent in activities of daily living.
When the baseline variable, severity of stroke, was
crosstabulated with functional status outcome, the chisquared statistic of 25.2 was highly significant (p <
.0001). Table 3 shows crosstabulations using the Mantel-Haenszel method23 of the baseline, treatment and
complication variables with functional status outcome
controlling for the major baseline variable, severity of
stroke after admission to the Stroke Treatment Center.
It is important to emphasize that the outcome is determined at discharge from the Stroke Treatment Center
and is thus a measure of recovery from the acute event.
However, the long term prognosis is altered by most if
not all of the complication variables.24 None of the
baseline variables was significantly associated with
outcome after adjusting for severity of stroke. The
treatment variable — place on aspirin at admission to
the center — was significantly associated with outcome. The data suggest that patients started on aspirin
at admission were more likely to have a successful
outcome (ambulatory and independent) than those who
were not, after adjusting for severity of stroke. Unfortunately, there were not enough data for the other treatment variables listed in table 1 to include them in the
analysis.
Stroke extensions and "other complications"t were
both significantly related to outcome. Table 3 shows
that even after the severity of the stroke is taken into
account, patients with stroke extensions or other complications were more likely to have a poor outcome.
The prevention of a stroke extension or other complication is important in improving the short-term outcome of stroke patients.
Table 4 gives the results of a stepwise logistic
regression procedure using baseline, treatment and
complication variables as independent variables and
outcome (successful/unsuccessful) as the dependent
variable. The most important variables were initial
•Thirty-three (33) of the patients in the sample had a stroke extension.
tTwenty-five (25) of the patients had at least one complication:
urinary tract infection (10), pneumonia (8), gastrointestinal bleeding
(6), decubiti (5), pulmonary edema (2), thrombophlebitis (1) and
contracrures (I).
861
EVALUATION AND PROGNOSIS OF STROKE/Dove et al
TABLE 2 Functional Status Outcomes for the Sample of 97 Stroke
Unit Patients
Outcome
N
1. Ambulatory and independent in activities of daily
living (ADL)
2. Ambulatory and needs some help with ADL
49
50.5
14
14.4
3. Ambulatory with personal assistance and ADL
with assistance
4. Not walking at all and ADL with assistance
5. Total care required
7
7.2
12
12.4
7
7.2
6. Deceased
8
8.2
Total
97
100.0
severity of stroke and stroke extension. The adjusted
odds ratios shown in this table are of particular interest. For example, a patient with a class V stroke has a
five-fold risk of having an unsuccessful outcome comDownloaded from http://stroke.ahajournals.org/ by guest on July 28, 2017
TABLE 3 Summary of Cross-tabulations of Selected Baseline,
Treatment and Complication Variables with Functional Status Outcome (SuccessfuUUnsuccessful)* Controlling for Severity of Stroke
at First Neurological Assessment after Admission to Stroke Treatment Center
Variable
Overall
MantelHaenszel
X2
Baseline
Age
0.4
<60 (n = 43)
>60 (n = 54)
Previous stroke (no/yes)
2.1
no (n = 74)
yes (n = 23)
Heart disease
3.0
absent (n = 58)
present (n = 45)
Hypertension
1.0
absent (n = 52)
present (n = 45)
Process
Placed on aspirin
8.5t
yes (n = 62)
no (n = 35)
13.2$
no (n = 64)
yes (n = 33)
Other complications
Adjusted
odds
ratio
Variables
Initial severity of stroke
l.Ot
class II
class III
1.7
class IV
3.0
class V
5.2
6.6
Stroke extension
•Patients with successful outcomes were ambulatory and independent in ADL at discharge from the Stroke Treatment Center.
tReference category.
pared to a patient with a class II stroke. A patient
having a stroke extension is 6.6 times more likely to
have an unsuccessful outcome than someone without
this complication.
Expected Versus Achieved Outcomes — Goal Setting
The correlation between participants' predictions^
and patients' actual outcomes is shown in table 5.
Participants' predicted outcomes were highly correlated with patients' achieved outcomes, particularly
when both baseline data and data describing the patients' stay in the Stroke Treatment Center were used.
The data indicate that no participant stood out as being
more accurate than others in predicting patients' actual
outcomes and that no particular medical discipline provided better results than other disciplines.
The correlation analysis revealed that the mistakes
in predictions by participants were not large and the
predictions improved when information on baseline
and process variables were made available to them.
The ridit analysis*25 in table 6 shows that when participants used only baseline variables in making their
judgments, they did not make systematic errors. However, when using all of the variables, the study participants tended to systematically downgrade their predictions, i.e., become consistently pessimistic about the
patients' outcome. For four of the six participants,
these shifts were statistically significant.
Discussion
Complication
Stroke extension
TABLE 4 Stepwise Logistic Regression Using Initial Severity,
Age, Previous Stroke, Hypertension, Heart Disease, Living Arrangement, Diabetes, Onset, Aspirin, Stroke Extension and Complications Against Functional Status at Discharge from Stroke
Treatment Center (Successful/Unsuccessful Outcome*)
4.4§
Baseline, Treatment and Outcome
This study showed that about half of the patients
admitted with strokes (excluding TIA's, previous
strokes with severe disabilities, brain stem strokes,
none (n = 72)
yes (n = 25)
•Patients discharged from the subacute area of the center who
were ambulatory and independent in activities of daily living were
classified as having a successful outcome.
tp < 0.01.
Xp < 0.001.
§p < 0.05.
tA reliability check on individual study participants using 10 of the
patients in the sample revealed a high degree of reliability in the predictions. There was no indication of significant upgrading or downgrading
for any of the participants.
•A ridit analysis is a statistical technique which is useful when one
has data from several samples where the subjects from each sample are
distributed over a number of ordered categories, i.e., functional status
outcomes scored on a scale from 1-6.
862
STROKE
TABLE 5 Correlation Between Participants' Predictions and Patients' Actual Functional Status Outcomes (n = 97)
Correlation
coefficients
(prediction, actual
outcome)
Participants
Downloaded from http://stroke.ahajournals.org/ by guest on July 28, 2017
Goals using baseline variables only
physician 1
physician 2
nurse 1
nurse 2
neuropsychologist
social worker
Goals using all variables
physician 1
physician 2
nurse 1
nurse 2
neuropsychologist
social worker
+ 0.6056
+ 0.6147
+ 0.6441
+ 0.6865
+ 0.6472
+ 0.6621
+ 0.7574
+ 0.7183
+ 0.7707
+ 0.7743
+ 0.7606
+ 0.7739
intracerebral hemorrhages with coma and subarachnoid hemorrhages), had a successful outcome — ambulatory and independent at discharge. However, it is
difficult to compare these findings with those of other
studies for at least two reasons. First, criteria for accepting patients into a study sample vary considerably
from study to study. For example, some researchers
include patients with TIA's in their study samples
while others exclude this category (as did the present
investigation). Second, stroke units vary from special
intensive care units for a small number of critically ill
patients admitted soon after the development of symptoms to units designed specifically for rehabilitation of
patients with completed stroke.6
The baseline variables were found to be more imporTABLE 6 Comparing Participants' Goal Setting Using Ridit
Analysis for Ordered Data with the Distribution of Patients' Actual
Outcomes as a Standard
Mean
Goal
Goals using baseline only
physician 1
physician 2
nurse 1
nurse 2
neuropsychologist
social worker
Goals using all variables
physician 1
physician 2
nurse 1
nurse 2
neuropsychologist
social worker
ridit
Z
statistic Significance
0.531
0.492
0.525
0.501
0.488
0.513
1.07
-0.28
0.86
0.03
-0.41
0.45
0.565
0.551
0.589
0.582
0.542
0.579
2.24
1.76
3.07
2.83
1.45
2.72
NS
NS
NS
NS
NS
NS
p < 0.025
NS
p < 0.003
p < 0.005
NS
p < 0.007
VOL 15, No 5, SEPTEMBER-OCTOBER
1984
tant than the treatment variables in predicting stroke
outcomes as defined in this study. Stroke extensions
which occurred and were theoretically preventable,
however, were even more important factors relating to
outcome. As expected, the most powerful, predictive
baseline variable was the initial severity of the stroke.
Although a comparison with other studies cannot be
accurately made, other researchers have found severity
of stroke to be a key factor in stroke outcome.13"16
The treatment variable showing the strongest relationship with reduced stroke disability was treatment
with aspirin on admission to the center. The reduced
stroke disability was likely secondary to a decrease in
incidences of stroke extensions; however, the small
sample size precluded further statistical analyses for
this relationship. Much work has been done in attempting to determine the effectiveness of aspirin in reducing the risk of completed strokes among patients with
transient ischemic attacks.26"30 However, little has been
reported about the use of aspirin in acute stroke. The
results of this study should not be mistaken for those of
a controlled experiment involving aspirin; however,
our results, in our opinion, suggest that this avenue
should be explored in future research.
The complication having the greatest effect on outcome was stroke extension. The data indicate that patients having exacerbations of the initial stroke are over
six times more likely not to be ambulatory and independent on discharge from the center, regardless of the
initial severity. Thirty-four percent of the patients (33)
had stroke extensions. Preventing this complication of
stroke would substantially improve the short-term
functional status of the patient after a stroke. It is worth
repeating that the definition of stroke extension used in
this study was broader than that of Jones and Millikan
who defined a "late exacerbation" as a significant
worsening in neurological status after 48 hours of stable course.22 In this investigation, a stroke exacerbation was a major worsening following a relatively stable period of at least 24 hours.
Predicted Versus Achieved Outcomes
The six participants were generally very successful
in predicting patient's actual outcomes. No participant
produced significantly better results than the others; no
medical profession (nurse, physician, social worker
and neuropsychologist) provided more accurate predictions than another. Participants improved their predictions when they had access to both admission data
and data describing the patients' stay in the center,
particularly knowledge about stroke extensions and
other complications.
The similarity of predictions made by the different
participants and disciplines may reflect the fact that all
members of the team meet daily to discuss each patient's progress and, therefore, develop similar expectations for patients. Even though each participant made
his/her predictions blindly, independently, individually and without any knowledge of the others' predictions and indeed years after discharge of patients, the
"team atmosphere" in the Stroke Treatment Center
EVALUATION AND PROGNOSIS OF STROKE/Dove et al
probably helped the participants to make accurate —
and similar — predictions.
Generalizeability to Other Institutions
The method of evaluating patient care attracted the
interest of the participants and offered insights into the
care provided at the West Haven VA Medical Center.
The method may be useful on other clinical problems.
At other institutions, participants may choose different
outcomes and/or process variables, depending on their
perceived clinical relevance. It is likely — even desirable — that other outcome variables, for example,
would be chosen.
Downloaded from http://stroke.ahajournals.org/ by guest on July 28, 2017
Conclusion
The focus of this patient care evaluation was on
acute stroke patients admitted to a stroke treatment
center in a university-affiliated VA medical center.
The study technique involved the participation of six
stroke center staff members representing the professional disciplines of medicine, nursing, social work
and neuropsychology. The participants helped to
choose baseline, treatment and outcome variables
thought to be important in recovery from stroke. They
also predicted outcome for a sample of stroke patients.
The statistical analysis of the sample provided useful results:
1. Baseline variables — particularly initial severity
of stroke — were more important than treatment variables in predicting short-term outcome from stroke.
2. The treatment variable that seems to warrant further experimentation is prescribing aspirin on admission to the center.
3. Preventing late exacerbations would substantially improve stroke outcome.
4. Stroke Treatment Center staff members were
generally successful in predicting patients' functional
status outcomes.
5. The method used to assess the care provided was
both practical and flexible because it enabled the participants to suggest process and outcome variables that
were clinically meaningful to them.
Appendix A
Severity of Stroke Based on Neurological
Assessment Scoring*
Used at the West Haven Medical Center
Class I
TIA
Class II
Mental Status 75-100. Other defects mild to moderate.
Class III
Mental Status 50-75 OR minimal expressive or receptive
aphasia (50-75) OR motor defects 25 or less.
Class IV
Mental Status 25-50 OR aphasia 25-50. Other defects
need not be evident.
Class V
Mental Status 0-25 OR aphasia 0-25. Other defects need
not be evident.
•Scoring is done every 2 hours on patients (while awake) in the acute
area of the Stroke Treatment Center. Scores range from 0-100 in areas
including mental status, receptive aphasia, expressive aphasia, and
motor-arm, motor-leg.
863
Guidelines for Deriving Severity of Stroke Classes
for Stroke Audit Patients
1. Determine the five possible classes a patient would fall into based on
individual scores in the following five areas: mental status, receptive
aphasia, expressive aphasia, motor-arm, and motor-leg.
2. A patient's class (I, II, III, IV, or V) was the lowest class that he
clearly fell into.
3. If a patient's lowest class fell between two classes in mental status or
receptive or expressive aphasia, the other possible classes in the other
two areas were used. If these were higher than the lowest of the ' 'tied''
classes, the patient was placed in the higher class. If all the mental status
and aphasia classes were between the same two classes, then the motor
scores were considered. If these were high, the higher of the two classes
was used to classify the seventy of the stroke.
References
1. Carpenter RR, Reed DE: The Outcome for Patients with Cerebrovascular Disease in University and Community Hospitals. Stroke
3: 747-758, 1972
2. Feigenson JS, Feigenson MD, Gitlow HS, McCarthy ML, Greenberg SD: Outcome and Cost for Stroke Patients in Academic and
Community Hospitals. JAMA 1878-1880, 1978
3. Drake WE, Hamilton MJ, Carlsson M, Blumenkrantz J: Acute
Stroke Management and Patient Outcome. The Value of Neurovascular Care Units (NCU). Stroke 4: 933-945, 1973
4. Kennedy FB, Pozen TJ, Gabelman EH, Tuthill JE, Zaentz SD:
Stroke Intensive Care — An Appraisal. Amer Heart Jnl 80: 188—
196, 1970
5. Millikan CH: Stroke Intensive Care Units: Objectives and Results.
Adv Neurol 25: 361-365, 1979
6. Norris JW, Hachinski VA: Intensive Care Management of Stroke
Patients Stroke 7: 573-577, 1976
7. Pitner SE and Mance CJ: An Evaluation of Stroke Intensive Care:
Results in a Municipal Hopsital. Stroke 4: 737-741, 1973
8. Taylor RR: Acute Stroke Demonstration Project in a Community
Hospital. J SC Med Assoc 66: 225-227, 1970
9. Cooper SW, Olivet JA, Woolsey FM Jr Establishment and Operation of Combined Intensive Care Unit NY J Med 72: 2215-2220,
1972
10. Adler M, Hamaty D, Brown CC, Potts H: Medical Audit of Stroke
Rehabilitation: A Critique of Medical Care Review J Chron Dis
30: 4 6 1 ^ 7 1 , 1977
11. McCann BC, Culbertson RA: Comparison of Two Systems for
Stroke Rehabilitation in a General Hospital. J Am Geriatr Soc 24:
211-216, 1976
12. Feigenson JS, Gitlow HS, Greenberg SD: The Disability Oriented
Rehabilitation Unit — A Major Factor Influencing Stroke Outcome. Stroke 10: 5-8, 1979
13. Britton M, deFaire V, Hclmers C, Mian K: Prognostication in
Acute Cerebrovascular Disease. Acta Med Scand 207: 37—42,
1980
14. Oxbury JM, Greenhall RCD, Grainger KMR: Predicting the Outcome of Stroke: Acute Stage After Cerebral Infarction. Brit Med
Jnl 3: 125-127. 1975
15 Rankin J. Cerebral Vascular Accidents in Patients Over the Age of
60: II. Prognosis. Scot Med Jnl 2: 200-215, 1957
16. Truscott BL, Kretschmann CM, Toole JF, Pajak TF: Early Rehabilitative Care in Community Hospitals Effect on Quality of Survivorship Following a Stroke. Stroke 5: 623-629, 1974
17. Williamson JW. Evaluating Quality of Patient Care: Strategy Relating Outcome and Process Assessment. JAMA 218: 564-569,
1971
18. Anderson TP, Baldridge M, Ettinger MG: Quality of Care for
Completed Stroke Without Rehabilitation: Evaluation of Assessing
Patient Outcomes. Arch Phys Med Rehabil 60: 103-107, 1979
19. Anderson TP, McCIure WJ, Athelstan G, Anderson E, Crewe N,
Amdts L, Ferguson MB, Baldridge M, Gullickson G Jr, Kottke FJ:
Stroke Rehabilitation: Evaluation of Its Quality by Assessing Patient Outcomes. Arch Phys Med Rehabil 59: 170-175, 1978
20. Classification and Outline of Cerebrovascular Diseases 2, Report
864
21.
22.
23.
24.
25.
STROKE
of the Ad Hoc Committee on Cerebrovascular Disease of the National Institute of Neurological and Communicative Disorders and
Stroke. Stroke 6: 564-616, 1975
Wallace JD, Levy LL: Blood Pressure After Stroke. JAMA 246:
2177-2180, 1981
Jones HR, Millikan CH: Temporal Profile (Clinical Course) of
Acute Carotid System Cerebral Infarction. Stroke 7: 64-71, 1976
Mantel N, Haenszel W. Statistical Aspect of the Analysis of Data
from Retrospective Studies of Disease. JNCI 22: 719-748, 1959
Sacco RL, Wolfe PA, Kannel WB, McNamara PM: Survival and
Recurrence Following Stroke: The Framingham Study. Stroke 13:
290-295, 1982
Fleiss J: Statistical Methods for Rates and Proportions New York:
VOL
15,
No
5,
SEPTEMBER-OCTOBER
1984
John Wiley and Sons, 1981
26. Bamett HJM, McDonald JWD, Sackett DL: Aspirin — Effective in
Males Threatened with Stroke. Stroke 9: 295-298, 1978
27. Canadian Cooperative Study Group: A Randomized Trial of Aspirin and Sulfinpyrazone in Threatened Stroke. NEJM 299: 53-59,
1978
28. Fields WS, Lemak NA, Frankowski RF, Hardy RJ: Controlled
Trial of Aspinn in Cerebral Ischemia. Stroke 8: 301-315, 1977
29. Fields WS, Lemak NA, Frankowski RF, Hardy RJ: Controlled
Trial of Aspinn in Cerebral Ischemia. Part II. Surgical Group.
Stroke 9: 309-318, 1978
30. Millikan CH, McDowell FH: Treatment of Transient Ischemic
Attacks. Stroke 9: 299-308, 1978
A Randomized Trial of Team Care Following Stroke
SHARON WOOD-DAUPHINEE, P H . D . , * STANLEY SHAPIRO, P H . D . , !
Downloaded from http://stroke.ahajournals.org/ by guest on July 28, 2017
EDWARD BASS,
M.D., F.R.C.P.,$
PAMELA GEORGES,
R.N.,§
CLAUDETTE FLETCHER, B . S C , N . E D . , §
VALERIE HENSBY, B . S C .
P.T., |
AND BERNICE MENDELSOHN, M.S.P.A.U
SUMMARY A randomized controlled trial was conducted to examine the effects of interdisciplinary team
care on acute hospitalized stroke patients. After obtaining baseline information on 42 stroke victims
receiving conventional care in a general hospital, 130 stroke patients were stratified and randomly assigned
either to Traditional or Team care. Assessments by independent evaluators permitted comparisons between
Team and Traditional groups with reference to patient survival, motor performance and functional abilities. Data obtained prospectively from charts and treatment logs allowed the care process across groups to
be compared. Results demonstrated that Team and Traditional patients fared similarly in survival. However there was an unexpected difference in survival depending upon sex. For motor performance, male
survivors performed better with Team care and female survivors with the Traditional method. In terms of
functional abilities, male patients receiving Team care again performed better than their Traditional
counterparts, whereas in women there was no difference between the treatment groups.
Stroke Vol 15, No 5, 1984
DURING THE 1970s substantial attention was given
to the value of stroke units in enhancing patient outcomes. As proposed by Bonner,1 a stroke unit may be
either a multi-disciplinary team of stroke specialists
who consult throughout a hospital and provide services
wherever the patient is situated, or a defined geographic area of a hospital where stroke patients are admitted
and care is provided by a stroke team.
Some attempts to evaluate the effects of stroke units
on patients with acute lesions, have involved the comparison of care provided in specialized neurovascular
intensive care units with regular ward care. Although
three studies2"4 failed to show a significant decrease in
From the School of Physical & Occupational Therapy,* Epidemiology and Health,! McGill University and from the Department of Neurology, t the Department of Nursing,§ Physical Therapy Department, |
and Department of Speech Pathology and Audiology,'' Sir Mortimer B.
Davis Jewish General Hospital.
Supported by Grant #HS04072, National Center for Health Services
Research of the U.S. Public Health Service, t Dr. Bass is currently at the
College of Medicine, University of South Florida, Department of Neurology, Box 55, 12901 North 30th Street, Tampa, Florida 33612.
Address correspondence to Dr. S. Wood-Dauphinee, School of Physical and Occupational Therapy, McGill University, 3654 Drummond
Street, Montreal, Quebec H3G 1Y5.
Received October 3, 1983: revision #1 accepted February 28, 1984.
case fatality when patients received care in these units,
a reduction in post-stroke complications was noted.5-6
In contrast, two studies" l6 reported a significant decline in mortality after the establishment of a stroke
intensive care unit in their institutions.
Other programs have departed from the model of the
intensive care unit and called themselves "stroke
care" or "stroke disability" units. These units provide
early and comprehensive evaluation, care and rehabilitation for stroke patients. Comparative studies evaluating the effects of these units have, with one exception,7
determined that patients achieved better functional
outcomes.8"10 Patients cared for in these units were
sometimes, 8 but not always, 9 " more frequently discharged to home. In-hospital mortality,7"" and length
of stay" were not positively influenced, but the units
appeared to be effective in promoting family participation in patient care and the rehabilitation process.8"10
A careful examination of this literature provides no
consensus about the value of stroke units. Several studies suggest that the major effort exerted on behalf of the
patient is helpful, but since only one of these studies
was strictly randomized numerous questions remain.
The establishment of a stroke unit at the Sir Mortimer B. Davis Jewish General Hospital in Montreal,
Evaluating and predicting outcome of acute cerebral vascular accident.
H G Dove, K C Schneider and J D Wallace
Stroke. 1984;15:858-864
doi: 10.1161/01.STR.15.5.858
Downloaded from http://stroke.ahajournals.org/ by guest on July 28, 2017
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1984 American Heart Association, Inc. All rights reserved.
Print ISSN: 0039-2499. Online ISSN: 1524-4628
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://stroke.ahajournals.org/content/15/5/858
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Once the online version of the published article for which permission is being requested is located, click Request
Permissions in the middle column of the Web page under Services. Further information about this process is
available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Stroke is online at:
http://stroke.ahajournals.org//subscriptions/