(Stroke. 1997;28:543-549.)
© 1997 American Heart Association, Inc.
Articles |
Nonmedical Bed-Days for Stroke Patients Admitted to Acute-Care Hospitals in Montreal, Canada
From the Division of Geriatrics (N.E.M., D.G.) and School of Physical and Occupational Therapy, Department of Medicine (N.E.M., S.W.-D.), McGill University, and Division of Clinical Epidemiology, Royal Victoria Hospital (N.E.M, S.W-D., D.G., S.C.S.), Montreal, Canada.
Correspondence to Dr Nancy E. Mayo, Division of Clinical Epidemiology, Ross Pavilion, 4th Floor, Royal Victoria Hospital, 687 Pine Ave W, Montreal, QC, H3A 1A1. E-mail mdnm{at}musica.mcgill.ca.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionAppendix 1References |
|---|
Background and Purpose Reducing the amount of nonoptimal time stroke patients spend in the hospital should be a priority because prolonged hospitalization is not only costly but may be detrimental for persons with stroke through deconditioning, social isolation, and the fostering of dependent relationships. The purpose of this study was to determine the amount of time spent by stroke patients in acute-care hospitals that was not justified for medical reasons and to identify mechanisms contributing to nonmedical bed-days.
Methods A retrospective cohort study was performed with 2232 persons admitted for acute stroke to one of 13 hospitals in Montreal, Canada, during 1991. Information was collected on the patient, the stroke, functional status, course in hospital, services, and discharge. Nonmedical bed-days were calculated as the difference between the time to meet specified criteria and time of discharge. Associations with nonmedical bed-days were estimated with adjustment for patient mix.
Results Acute-care stay averaged 27 days, yielding 60 279 bed-days. Almost 50% of the cohort remained in the hospital after meeting criteria for medical discharge, resulting in 43% of total bed-days not accounted for medically. Fifty percent of persons with delayed discharge did not go home but were discharged to another acute-care hospital or to rehabilitation or long-term care, accounting for 66% of the nonmedical bed-days. Hospital and discharge destination remained strongly associated with nonmedical days, even after adjustment for patient mix.
Conclusions The single greatest contributor to excessive nonmedical stay appeared to be the need in Quebec for increased access to alternate levels of care, including skilled nursing facilities and rehabilitation centers.
Key Words: Canada • hospitalization • outcome • stroke management
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1References |
|---|
Stroke patients in Canada spend, on average, twice as many days in acute-care hospitals as do stroke patients in many other parts of the world.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 An extended period of acute-care hospitalization may genuinely be detrimental for persons with stroke, by increasing social isolation,19 by fostering dependent relationships and, through prolonged inactivity, by contributing to muscle weakness. Thus, persons with stroke would benefit from ways of reducing the length of time spent in the hospital to only that amount needed for acute medical care, and the system as a whole would benefit through increasing efficiency of care. Strategies for reducing bed-stay can only be devised through discriminating between the amount of time that is medically necessary and the amount of time not necessary on medical grounds and by identifying factors contributing to the latter.
The concept of nonmedical bed-days has been extensively studied in the United States20 21 and has been termed inappropriate bed-stay. These studies have, for the most part, examined inappropriate bed-stay for particular services but not for specific conditions. The proportion of total bed-stay deemed inappropriate is reported to range from 9% to 35%, with methodological differences such as review methods, instruments, reviewers, and sampling schemes being identified as the major contributor to variability across studies.20 Within studies, physician and hospital factors have been identified as the major determinants of inappropriate bed-stay, more important than patients' or family members' desires.20 The reporting of inappropriate bed-stay in Canada has been limited to data from single hospitals.22 23 24 Because of differences in how health care is provided in Canada compared with the United States, we cannot generalize results from American studies to Canadian institutions; however, the provision of high-quality care in the most efficient manner is a goal shared by healthcare providers worldwide.
Nonmedical bed-stay for stroke has received little examination. In a recent article by Goldman et al,25 a computerized algorithm based on a combination of patient characteristics and physician-identified needs for specific services was used to identify "unjustified hospital stay." The 177 patients admitted to one hospital in Wisconsin contributed 1841 bed-days, 41% of which were unjustified. The average length of stay was 5.6 days among the 57 patients with no unjustified stay but 12.6 days among the other 120 patients, who had an average of 4.3 unjustified days. This study illustrates that it is possible to examine unjustified stay rigorously. In the present report we describe a method for determining unjustified stay based on clinically oriented data.
The objective of this study was therefore to determine the proportion of time spent in the acute-care hospital after stroke that was not justified for medical reasons and to identify mechanisms contributing to nonmedical bed-days.
We have chosen to describe bed-days as "nonmedical" because, even under ideal circumstances, some acute-care stay is devoted to waiting for placement in another facility or for services in the community to be put in place. These days may in fact be appropriate even if they are not medically justified.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1References |
|---|
A retrospective chart review was performed on 2232 persons admitted for acute stroke to one of 13 Montreal area acute-care hospitals between January 1 and December 31, 1991. Charts were selected with the use of records of hospital discharges obtained from Quebec's hospital database, MEDECHO. All records with primary discharge diagnosis coded 430 to 437 (excluding 435: transient ischemic attack) according to the International Classification of Diseases, 9th Revision (ICD-9)26 were screened for evidence of acute stroke. Acute stroke was defined as an episode characterized by rapidly developing clinical signs of focal (or global in the case of subarachnoid hemorrhage without focal neurological deficits) disturbance of cerebral functions, lasting more than 24 hours or leading to death, with no apparent cause other than of vascular origin.27 All such events that had occurred not more than 30 days before admission or while already in hospital were included. Summaries of charts not fulfilling the conditions for acute stroke were reviewed by a study neurologist, who decided whether to exclude the individual. The accuracy of the procedure used for defining acute stroke based on MEDECHO codes and a retrospective review of medical charts has previously been validated.28 29 The chart abstraction form had been developed and pretested. Permission to review the medical charts was obtained from each hospital after ethical review.
Of the 3529 medical charts screened, 32.9% were excluded because the
diagnosis was something other than acute stroke, and an additional
3.8% could not be recovered. Thus, 2232 charts were reviewed.
Information on sociodemographic descriptors; coexisting disease; type
of stroke; presenting signs and symptoms; newly arising conditions and
progression of stroke; functional status; tests and procedures
performed and services and consultations received; medication use
before admission, during hospitalization, and at discharge; procedures
for discharge planning and destination at discharge; subsequent
emergency visits; and previous and subsequent admissions was collected.
The nine major Montreal area teaching hospitals were targeted for
review first, but owing to depleted resources, only four nonteaching
hospitals were included. The inclusion of community hospitals was based
on the timing of agreements, solicited unsystematically, with the
archival departments. Table 1
compares participating and
nonparticipating hospitals on information available on admissions for
cerebrovascular disease before screening for acute stroke.
|
Criteria for Medical Discharge
The criteria for medical discharge were selected through a
three-stage process. First, a consensus group was held following a
nominal group process as described by Delbeq et al.30 Ten
healthcare professionals representing neurology, cardiology,
geriatrics, internal medicine, nursing, physical therapy, occupational
therapy, speech therapy, and social work arrived at a consensus
regarding the important factors contributing to length of stay for
stroke in Montreal. Ten other persons, named by the first 10 as having
roles in caring for stroke patients similar to their own, were asked to
rate the identified variables in terms of their importance for length
of stay for stroke and to add any other relevant variables. Step 2 was
performed during the data analysis phase. Any factors found to be
significantly associated with length of stay or with discharge
destination but not identified in step 1 were reviewed by the group
and, if judged to constitute medically relevant factors, were added to
the criteria. Step 3 involved the finalization of the criteria as a
working algorithm based on coherence with clinical practice, summarized
by the Stroke Ready for Medical Discharge Checklist (see
"Appendix"). An individual's need for the specialized services
available in acute care was deemed met when (1) diagnosis was known;
(2) newly arising conditions were stable or improving, and tests had
been completed or could be completed as an outpatient (criteria 3
through 7 had been met and at least 3 days had passed since any newly
arising episode of items 8 to 15); and (3) 3 days had passed since
criteria 1 and 2 were met.
Statistical Analyses
Data were analyzed (1) to characterize the cohort, (2) to
determine length of stay, (3) to identify when individuals had met the
criteria for medical discharge, and (4) to identify factors explaining
the time between meeting these criteria and discharge.
Length of stay was calculated as time from admission (or from onset of stroke for 53 persons whose strokes occurred while hospitalized for other reasons) to discharge. Nonmedical bed-days were calculated as the difference between the time to meet criteria and the time of actual discharge. Each day of an individual's stay was evaluated as being justified or not, according to the set criteria. Once the person had met medical criteria, all subsequent time in hospital was classified as nonmedical, regardless of whether new episodes developed that would have justified care in the acute setting.
To examine variability in nonmedical bed-days, multiple linear regression would have been the preferred method of analysis; however, the highly skewed distribution of nonmedical bed-days prohibited this approach. Instead we used logistic regression, defining an event as more than 7 nonmedical bed-days. Once the best fitting logistic model was identified, a proportional odds ordinal regression model31 32 with five levels of outcome (0, 1 to 7, 8 to 21, 22 to 35, and >35 nonmedical bed-days) was used to determine whether the estimated associations depended on the dichotomization selected for defining an event. Persons transferred to another acute-care hospital before meeting criteria for medical discharge were not included in this analysis. Neurological deficits and conditions arising in the hospital were modeled as present or absent within a time window set as the first 7 days after admission (or onset of stroke for persons whose stroke occurred in the hospital). Comorbidity was defined according to the procedure described by Romano et al.33
A multivariate model was created consisting of (1) patient-mix variables (age, sex, type of stroke, comorbidity, and living arrangement); (2) neurological deficits and conditions arising during hospitalization; and (3) system-related variables (hospital and discharge destination). We believed it appropriate to adjust for the effect of hospital and discharge destination when considering the impact of patient-related factors on nonmedical bed-days because, in the absence of adjustment for system-related factors, the patient-related effects would be in part explained by these variables. Because of the large number of hospitals (n=13) and because there was no interest in any particular hospital, those with similar proportions of patients having more than 7 nonmedical bed-days after adjustment for case mix were grouped together. Only those neurological deficits and arising conditions that remained significant at the .05 level were retained in the final model. Two-way interactions between age and sex and each of the other variables were also examined.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1References |
|---|
Table 2
presents the characteristics of the cohort,
as well as differences in the distribution of these characteristics
across hospitals. There was an uneven distribution of patient mix. For
example, while 10% of persons were older than 85 years at the time of
their stroke, across hospitals this proportion ranged from 2% to 26%.
Mean length of acute-care stay was 27.0 days (SD, 30.8) and varied by
more than 13 days across hospitals. Mean length of stay also varied by
discharge disposition: persons discharged home were hospitalized on
average for 18.1 days, compared with 46.4 days for persons discharged
to long-term care or rehabilitation facilities.
|
On average, the time to meet criteria for medical discharge (or to discharge if discharge occurred before our criteria were met) was 16.2 days (SD, 20.0), with a range over hospitals of 10.6 to 28.7 days and a range over subjects of 0 to 339 days. Mean time to meet criteria also varied by discharge disposition: 11.9 days among persons discharged home, 18.0 days for those discharged to an outside facility for long-term care, 22.0 days for persons transferred to another acute-care facility, 23.2 days for persons discharged to a rehabilitation center, and 29.5 days for persons transferred to a long-term care ward within the acute-care hospital.
The Figure displays cumulative proportions of persons
(1) discharged and (2) who met criteria for medical discharge over
time. Fifty percent of individuals were hospitalized for 17 days
(median) or more and 25% for more than 36 days, whereas 25% of
persons had met medical criteria for medical discharge within 6 days,
50% had met criteria within 10 days (median), and 75% had met
criteria by day 19.
|
The total number of bed-days for this cohort was 60 279, representing
approximately three quarters of all bed-days attributed to stroke for
Montreal in that year. Table 3 presents the distribution
of individuals and bed-days in relation to meeting criteria for medical
discharge. Twenty percent of subjects (n=450) died before they met
criteria; 4% (n=87) were transferred to another acute-care hospital
before they met criteria; and 28% (n=466+37+125) were otherwise
discharged before all criteria on our checklist were met. The remaining
48% of persons (range across hospitals, 27.5% to 66.1%) were
discharged (including 34 deaths) after meeting the criteria, and their
stay accounted for 69% [(16 206+25 668)/60 279] of all bed-days,
although only 39% [16 206/(16 206+25 668)] of their bed-days were
medically justified. Of these 1067 persons with any nonmedical
bed-stay, 72% had more than 7 days of stay not medically justified,
and 39% had more than 3 weeks. Overall, 43% (25 668/60 279) of all
bed-days were not medically justified.
|
Associations of patient-related and system-related factors with
nonmedical bed-stay are presented in Table 4. Age, sex,
and comorbidity were not independent predictors of nonmedical bed-days;
they served only as adjustment factors for the other variables. Type of
stroke was not significantly associated with nonmedical bed-days once
other adjustment factors were considered. Persons admitted from other
institutions had a significantly lower probability of nonmedical
bed-days, likely because their discharge destination was predetermined.
Persons with any of confusion/disorientation, sensory disturbance,
aphasia, or lower-extremity paresis were more likely to have more than
7 nonmedical bed-days (odds ratio [OR], 1.3 to 2.1), as were persons
having difficulty with bowel movements and persons exhibiting signs of
depression during the course of hospitalization (OR, 1.4 and 1.7,
respectively).
|
Even after adjustment for case mix, neurological deficits, and additional events arising during the course of hospitalization, hospital factors remained strongly associated with nonmedical bed-stay. Compared with the three hospitals with the lowest proportions of nonmedical bed-days, ORs for hospital groups ranged from 1.4 to more than 8 for one outlier hospital.
Discharge destination was also strongly associated with nonmedical bed-days. In comparison to persons discharged home, the odds of having more than 7 nonmedical bed-days were 5.7 times higher for persons discharged to a rehabilitation facility. Persons discharged to long-term care also had a higher probability of nonmedical bed-days than persons discharged home, and this was higher for persons being discharged to an outside long-term care institution (OR, 3.4) than for persons being transferred to a long-term care ward within the institution (OR, 1.7). Persons who died in the hospital after their stroke were very unlikely to incur nonmedical bed-days (OR, 0.14). Proportional odds regression analysis confirmed the predictability of the variables identified by logistic regression with other dichotomizations of nonmedical bed-days. The magnitude of the effect of discharge destination on nonmedical bed-days was found to vary depending on the dichotomization of nonmedical bed-days selected.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1References |
|---|
A considerable proportion of the hospitalization for stroke was not medically justified according to the criteria for this project. The mean time to meet criteria or to be discharged was 16.2 days compared with the actual mean length of stay of 27.0 days. Fifty percent of patients were discharged or ready for medical discharge within 10 days, which is close to the 8 days covered in the United States for the diagnosis-related group (DRG) for stroke,6 34 which could be considered a proxy for need for acute care among typical cases of stroke.
The proportion of bed-stay deemed nonmedical was 43%, larger than what has been reported in the literature.20 21 35 36 However, it is in agreement with what was reported recently by Goldman et al,25 although the length of stay for stroke in the Goldman study, performed in the United States, was much shorter than in Quebec (11 days versus 27 days).
One factor that could explain the discrepancy between results from our
study and others in the literature is the criteria used to determine
readiness for discharge. Previous studies mainly used the
Appropriateness Evaluation Protocol, a tool designed for use with many
different types of patients.35 However, there was a high
degree of similarity of our criteria with these more general criteria.
We do not think our criteria were too stringent because, as indicated
in Table 3
, more than 28% of persons were discharged alive from acute
care on or before meeting the criteria. The major reason for leaving
acute care before the criteria were met was that discharge was on the
same day that the intravenous equipment was removed, whereas we allowed
a 3-day grace period. The choice of a 3-day grace period was arbitrary
and was an attempt to prevent the criteria from being unreasonably
rigid. We were comfortable with this choice after verifying that
reducing the grace period did not result in substantial differences in
the proportion of persons discharged or deceased before criteria were
met (49.4% versus 46.5% and 41.9%, for 2 days and 1 day,
respectively), nor did changing the grace period have a dramatic impact
on the proportion of bed-stay considered nonmedical (changing from
42.6% to 45.5% for a 2-day grace period and to 48.7% for a 1-day
grace period). The proportion of subjects with nonmedical bed-days who
died in the hospital would have increased from 3.2% to 3.5% as the
days of grace were decreased.
The calculation of nonmedical bed-stay was based on considering as nonmedical all days of stay beyond the point at which criteria for medical discharge were met. Indeed, when a 3-day grace period was used, 47% of persons hospitalized past the point of meeting criteria subsequently experienced events that would have been considered to require medical care; however, almost one third of these events did not occur until more than 30 days after criteria were met. If persons had been discharged at the time criteria were met, then potentially 45% of persons would have required supplementary medical intervention; however, the vast majority of events could readily be managed outside of the acute-care setting. There were very few serious events, and those that did occur would not have been predictable (eg, two persons with myocardial infarction and two persons with a second stroke). Only two events occurred with a frequency of 10% or more: new episodes of bowel incontinence (28%) and vomiting (11%). Dysrhythmias were not uncommon, occurring at a frequency of 4% for arrhythmias, 8% for bradycardia, and 9% for tachycardia. Also, while 34 of the 1067 persons with nonmedical bed-stay subsequently died in the hospital, 75% died 14 or more days, 50% died 25 or more days, and 25% died 48 or more days after criteria were met. On the basis of the type of events documented after criteria were met, it would not seem imprudent to discharge persons when criteria are met and deal with later acute events through existing services. It is an expensive undertaking to keep persons hospitalized in case complications develop.
There were a number of limitations to this study; however, they do not
negate the internal validity of the study or the importance of the
finding that 43% of bed-days for stroke were not medically justified.
Although the majority of hospitals included were teaching hospitals,
judging from the differences between participating and nonparticipating
hospitals, the inclusion of more community hospitals would not have
reduced the proportion of nonmedical bed-days. There are very real
differences between community and teaching hospitals, and these
differences are reflected in the type of stroke patient admitted and in
their management (see Table 1). However, even though this study was not
able to include persons with stroke admitted to all hospitals, the
variability among the more than 2000 patients in terms of age,
functional status, and type of stroke suggests that the results are
generalizable to the range of presentation of stroke in the population.
The range of duration of stay reflects the duration of stay throughout
Canada,37 and thus the results are likely to hold
elsewhere in this country.
We are confident that the medical charts reviewed pertained to persons with acute strokes; however, there may have been some underascertainment of events. Using administrative data, we selected for review only charts with ICD-9 codes for cerebrovascular disease. We did not review charts that listed transient ischemic attack as the only stroke-related diagnosis because this ICD-9 rubric has been shown not to have a high yield in terms of acute stroke.29 A small number of strokes may also "hide" in codes such as vertigo or coma.29
Data for this study were collected retrospectively and thus represent a true portrait of typical care and not the type of care received under an experimental situation. The disadvantages of working retrospectively are that some data are not available and other data may not be complete. We did not have information relating to physician, the type of service to which the patient was admitted, or severity of stroke. For severity of stroke, the data available pertained to the presence or absence of certain deficits (eg, leg or arm weakness, dysphagia, aphasia). Severity of these deficits was not well documented and remains a limitation of the retrospective approach. However, severity of stroke is, to a certain extent, subsumed in the time to meet medical criteria, which ranged from 0 to 339 days across patients.
Data on patients' economic situation were also unavailable; however, this was unlikely to introduce a confounding bias because Canada's healthcare system guarantees universal access, and there are only limited private resources for stroke care available for purchase by more economically advantaged patients. Data on whether community-dwelling patients lived alone or with a companion were often missing among persons at the extremes of the range of severity; the association of nonmedical bed-days with living alone could therefore not be estimated. Missing information was not a problem for the outcome variable or the main variables of interest. There was evidence on every chart that a standard neurological examination had been performed, and thus the documentation of the common neurological deficits was likely to be complete. While there could have been error in documenting events arising during the course of hospitalization, events important enough to require medical or nursing care or to affect level of care were unlikely to be missed.
Although the data collection for this project was retrospective, the Stroke Ready for Medical Discharge Checklist is meant to be used in a prospective fashion. Its aim is to provide a mechanism to monitor outcome in terms of medical stability and to ensure that discharge occurs expediently. The prospective evaluation of this checklist in reducing length of stay would be an essential next step.
Why, then, are persons who medically appear to be ready for discharge not discharged? The strongest reasons appear to be system related as opposed to patient related. After we controlled for sociodemographic and stroke-related variables, hospital and discharge destination were important contributors to stay that was not medically justified. The hospital factors remain unmeasured at present but are potentially amenable to modification.
The single greatest contributor to excessive nonmedical stay appeared to be the need for alternate levels of care. Quebec does not have skilled nursing facilities that would likely meet the needs even of persons who are eventually discharged home. There are few places available in rehabilitation facilities, and waiting lists can be long. Discharge to long-term care is not undertaken lightly. Persons not ready or unable to return to the community, either because of severe disability or because there is no one at home to provide care, must wait for a bed, and they wait in the acute-care setting. For persons who could return home, the treatment team is reluctant to recommend discharge when there is limited access to rehabilitation services other than as an inpatient at an acute-care hospital.
The development of skilled nursing facilities would be a solution for some types of patients. For others, it would be more expedient to have rehabilitation services more readily available outside of the acute-care setting. Given that the acute-care hospitals already offer physical and occupational therapy on an inpatient basis, this could be achieved with a change in emphasis from inpatient to outpatient. Since the aim of rehabilitation is to facilitate community reintegration, it would appear logical that the sooner the patient leaves the protected environment of the hospital, the sooner the reintegration process can commence.
Although these data reflect the situation during 1991, little has changed in Quebec's healthcare system that would improve the situation for stroke care. Hospitals are being closed, and resources are being shifted toward the provision of community-based services; however, the type of services that stroke patients need are not yet in place. While inpatient rehabilitation beds are being reduced, however, these facilities are being encouraged to expand outpatient services. Thus, some seemingly negative events are being coupled with potentially positive changes. It remains to be seen whether there will be a net gain in the actual allocation of healthcare resources and in the outcome of stroke.
|
Acknowledgments |
|---|
This study was supported by Health Canada, National Health, Research and Development Program 6605-392-57. The authors would like to thank Drs Irena Danys and Joseph Carlton for their contribution to the development of the research question and in screening stroke charts for inclusion. We would also like to thank Claudette Corrigan, who coordinated the project, including the data abstraction and the computerization of the data. In addition, the authors would like to thank the following data abstractors: Luc Tellier, Alex Hazzi, Monique Pépin, Barbara Zenobie, Nancy Perrin, Linda Perrin, Sara McEwen, Nancy Béland, Kathy Daley, Cathy Berg, Chantal Pelletier, Diane Lalancette, Sylvie Courtemanch, and Laura Barré, as well as Sara Mayo and Stephanie Goldbaum for computerizing the data.
|
Appendix 1 |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1References |
|---|
|
Received June 19, 1996; revision received November 19, 1996; accepted November 19, 1996.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1References |
|---|
1. Anderson C, Stewart-Wynne E, Jamrozik K, Burvill P, Chakera T. Perth Community Stroke Study: design and preliminary results. Clin Exp Neurol. 1990;27:125-129. [Medline] [Order article via Infotrieve]
2.
Wolfe CD, Taub NA, Woodrow J, Richardson E, Warburton
FG, Burney PG. Patterns of acute stroke care in three districts
of southern England. J Epidemiol Community Health. 1993;47:144-148.
3. Wade DT, Langton-Hewer R, Skilbeck CE, Bainton D, Burns-Cox C. Controlled trial of a home-care service for acute stroke patients. Lancet. 1985;1:323-326. [Medline] [Order article via Infotrieve]
4. Wade DT, Wood VA, Langton-Hewer R. Use of hospital resources by acute stroke patients. J R Coll Physicians Lond.. 1985;19:48-52. [Medline] [Order article via Infotrieve]
5. Gillum RF, Gomez-Marin O, Kottke TE, Jacobs DR Jr, Prineas RJ, Folsom AR, Luepker RV, Blackburn H. Acute stroke in a metropolitan area, 1970 and 1980: the Minnesota Heart Survey. J Chron Dis.. 1985;38:891-898. [Medline] [Order article via Infotrieve]
6. Lagoe RJ, Lauko SJ. Stroke hospitalization under prospective payments: analysis of Diagnosis Related Group 14. Arch Phys Med Rehabil. 1985;66:773-776. [Medline] [Order article via Infotrieve]
7. Shah SK, Bain C. Admissions, patterns of utilization and disposition of cases of acute stroke in Brisbane hospitals. Med J Aust.. 1989;150:256-260. [Medline] [Order article via Infotrieve]
8.
Ward G, Jamrozik K, Stewart-Wynne E. Incidence
and outcome of cerebrovascular disease in Perth, Western
Australia. Stroke. 1988;19:1501-1506.
9. Cable EP, Mayers SP Jr. Discharge planning effect on length of hospital stay. Arch Phys Med Rehabil. 1983;64:57-60. [Medline] [Order article via Infotrieve]
10.
Sacco RL, Foulkes MA, Mohr JP, Wolf PA, Hier DB, Price
TR. Determinants of early recurrence of cerebral infarction: the
Stroke Data Bank. Stroke. 1989;20:983-989.
11.
Kahn KL, Keeler EB, Sherwood MJ, Rogers WH,
Draper D, Bentow SS, Reinisch EJ, Rubenstein LV, Kosecoff J, Brook
RH. Comparing outcomes of care before and after implementation
of the DRG-based prospective payment system. JAMA. 1990;264:1984-1988.
12. Adelman SM. The national survey of stroke: economic impact. Stroke. 1981;12(suppl):69-78.
13.
Dombovy ML, Basford JR, Whisnant JP, Bergstralh
EJ. Disability and use of rehabilitation services following
stroke in Rochester, Minnesota, 1975-1979. Stroke. 1987;18:830-836.
14.
Britton M, Roden A. Progression of stroke after
arrival at hospital. Stroke. 1985;16:629-632.
15.
Strand T, Asplund K, Eriksson S, Hagg E, Lithner F,
Wester PO. A non-intensive stroke unit reduces functional
disability and the need for long-term hospitalization.
Stroke. 1985;16:29-34.
16. Friedman PJ. Stroke rehabilitation in the elderly: a new patient management system. N Z Med J. 1990;103:234-236. [Medline] [Order article via Infotrieve]
17. Osberg JS, Haley SM, McGinnis GE, DeJong G. Characteristics of cost outliers who did not benefit from stroke rehabilitation. Am J Phys Med Rehabil.. 1990;69:117-125. [Medline] [Order article via Infotrieve]
18. Schluep M, Bogousslavsky J, Regli F, Tendon M, Prud'hom LS, Kleiber C. Justification of hospital days and epidemiology of discharge delays in a department of neurology. Neuroepidemiology.. 1994;13:40-49. [Medline] [Order article via Infotrieve]
19. Fisher RH, Zorzitto ML. Placement problem: diagnosis, disease or terms of denigration? Can Med Assoc J.. 1983;129:331-333. [Medline] [Order article via Infotrieve]
20. Payne SMC. Identifying and managing inappropriate hospital utilization: a policy synthesis. Pew Memorial Trust Policy Synthesis: 4. Health Serv Res. 1987;22:709-769. [Medline] [Order article via Infotrieve]
21. Payne SMC, Restuccia JD, Ash A, Shwartz M, Tarr L, Williams B. Using utilization review information to improve hospital efficiency. Hosp Health Serv Admin. 1991;36:473-490. [Medline] [Order article via Infotrieve]
22. Sheps SB, Anderson G, Cardiff K. Utilization management: a literature review for Canadian health care administrators. Healthcare Manage Forum. 1991;4:34-39. [Medline] [Order article via Infotrieve]
23. O'Neill ML. Bed utilization: a regional trial. Leadership. 1994;2:35-39.
24. Collins PE, Skafel JR, Atkinson DI, Jenkins S. Continuum: measuring and managing the patient care process. Healthcare Manage Forum. 1994;7:27-40. [Medline] [Order article via Infotrieve]
25.
Goldman RS, Hartz AJ, Lanska DJ, Guse CE.
Results of a computerized screening of stroke patients for unjustified
hospital stay. Stroke. 1996;27:639-644.
26. World Health Organization. International Classification of Diseases, 9th Revision: Manual of the International Statistical Classification of Diseases, Injuries, and Causes of Death. Geneva, Switzerland: World Health Organization; 1978.
27. Aho K, Harmsen P, Hatano S, Marquardsen J, Smirnov VE, Strasser T. Epidemiology of cerebrovascular diseases in the community: results of a WHO collaborative study. Bull World Health Organ.. 1980;58:113-130. [Medline] [Order article via Infotrieve]
28. Mayo NE, Danys I, Carlton J, Scott SC. Validity of hospital discharge coding for stroke. Can J Cardiol. 1993;9(suppl D):121-124.
29. Mayo NE, Chockalingam A, Reeder BA, Phillips S. Surveillance for Stroke in Canada. Health Reports. 1994;6:62-72. Statistics Canada 82-003. [Medline] [Order article via Infotrieve]
30. Delbeq A, Van de Ven A, Gustafson D. Group Techniques for Program Planning: A Guide to Nominal Group and Delphi Processes. Dallas, Tex: Scott, Foresman & Co; 1975.
31.
Armstrong BG, Sloan M. Ordinal regression models
for epidemiologic data. Am J Epidemiol. 1989;129:191-204.
32. Scott SC, Goldberg MS, Mayo NE. Statistical assessment of ordinal outcomes in comparative studies. J Clin Epidemiol. In press.
33. Romano PS, Roos LL, Jollis JG. Adapting a clinical comorbidity index for use with ICD-9-CM administrative data: differing perspectives. J Clin Epidemiol. 1993;46:1075-1079. [Medline] [Order article via Infotrieve]
34. Department of Health and Human Services. Medicare program: changes to the hospital inpatient prospective payment systems and fiscal year 1994 rates. Federal Register. September 1, 1993;58:46270-46580.
35. Gertman PM, Restuccia JD. The appropriateness evaluation protocol: a technique for assessing unnecessary days of hospital care. Med Care. 1981;19:855-871. [Medline] [Order article via Infotrieve]
36. Santos-Eggimann B. Hospital utilization reviews under field conditions: potential and improvements. Int J Technol Assess Health Care. 1993;9:514-521. [Medline] [Order article via Infotrieve]
37.
Mayo NE, Neville D, Kirkland S, Ostbye T, Mustard CA,
Reeder B, Joffres M, Brauer G, Levy AR. Hospitalization and
case-fatality rates for stroke in Canada from 1982 through
1991. Stroke. 1996;27:1215-1220.
This article has been cited by other articles:
 |
|
 |
|
  L. De Wit, K. Putman, E. Dejaeger, I. Baert, P. Berman, K. Bogaerts, N. Brinkmann, L. Connell, H. Feys, W. Jenni, et al. Use of Time by Stroke Patients: A Comparison of Four European Rehabilitation Centers Stroke, September 1, 2005; 36(9): 1977 - 1983. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. M. Wilson and C. D. Truman Evaluating Institutionalization by Comparing the Use of Health Services before and after Admission to a Long-Term-Care Facility Eval Health Prof, September 1, 2004; 27(3): 219 - 236. [Abstract] [PDF]
|
|
|
|
|
|
N. E. Mayo, S. Wood-Dauphinee, R. Cote, D. Gayton, J. Carlton, J. Buttery, and R. Tamblyn There's No Place Like Home : An Evaluation of Early Supported Discharge for Stroke Stroke, May 1, 2000; 31(5): 1016 - 1023. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||